post_id
stringlengths 5
7
| domain
stringclasses 18
values | upvote_ratio
float64 0.5
1
| history
stringlengths 22
39.2k
| c_root_id_A
stringlengths 7
7
| c_root_id_B
stringlengths 7
7
| created_at_utc_A
int64 1.28B
1.67B
| created_at_utc_B
int64 1.28B
1.67B
| score_A
int64 2
43.5k
| score_B
int64 2
43.2k
| human_ref_A
stringlengths 0
10.7k
| human_ref_B
stringlengths 0
10.8k
| labels
int64 0
1
| seconds_difference
float64 0
145M
| score_ratio
float64 1
3.72k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
g3k9gg | askacademia_train | 0.98 | What does it mean for someone to be "good at research?" Sorry if the title is a bit ambiguous and broad. I was chatting last night with a couple of people from my lab. One of them is a recent PhD who's a year older than me and is one of the lab's "assets" in the sense that he's done some good work in our field and many people look up to him. Let's call him Tom. Tom also happens to be one of those people who I would call "pretentious" in the sense that his usual attitude is "I don't need people who work hard I need people who are actually good." Just for some background information, I'm a second semester MSCS student. I plan to applying to PhD programs abroad this fall so that when I graduate next spring I can start right away. The context of how the title of this post came to be is that Tom usually tells me things like "among your line of batchmates (people who entered the lab the same year) A and B are going to be the ones who shine" (I may be taking this the wrong way but the way he says it seems to imply that I'm not good enough), "I get that you enjoy research, but you really need to ask yourself if you're actually good at it," etc. and last night that really got me thinking. What exactly does it mean to be "good at research?" I asked him the same and he didn't really give me a straightforward answer after I retorted that everybody has their own measures of success. I took a look at this Quora question that's answered by one of my favorite Quora people but I was curious what other people on this subreddit may think. My personal criteria for one to be good at research is to first and foremost actually enjoy it (e.g. enjoy the feeling of realizing how "dumb" one is and enjoy the process of filling in knowledge gaps). I personally don't hold "publish a lot of papers" as a criterion but I've noticed the majority of people do. | fnt7cs0 | fnsakc6 | 1,587,236,023 | 1,587,217,239 | 3 | 2 | A. There will be a lot of Tom's in academia (but they are not only male), who will try to convince you that you're not good enough, won't make it, don't fit in, won't get your papers published, etc. Tom doesn't want you to find your potential. Tom is threatened by everyone. B. Ignore all the Toms (and for that matter most PhD student opinions). C. To be a good researcher is to be inquisitive, to develop ideas (keep a list of possible papers), to be self-motivated and passionate about your topics, and to believe that you're doing something worthwhile. The practical skills necessary are of course careful reading, good writing (what determines this is field specific), understanding the structure and rules of writing up journal articles or books, and being able to spot areas for new development or critique. D. Many PhDs take jobs that are not research heavy, so it is not the only path if you decide you don't love the research grind. | First, don't worry about Tom, as others have said. What I think makes a good researcher is the following: 1. Motivated 2. Knows how to do a good literature review and find the gap in the literature. 3. Ethical At the end of the day, research is knowing what's been done and knowing what needs to done, then filling that gap or part of that gap or even just attempting to fill the gap. | 1 | 18,784 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxup7jt | gxuosg7 | 1,620,827,091 | 1,620,826,890 | 217 | 182 | Theres a reason lots of people use that advice: it works. No the figures wont be perfect, but you need an order of events that makes sense, and figures allow you to present that story. Same reason shows put together a storyboard. I dont believe your PI meant that the figures shouldnt change at all while writing; if he did thats taking this too far, you're right. | It's a give and take. You need to start somehwere. I like putting the story together with figures first (these are just mock figures). I then write a draft of the results and continuously change the figure panels to make it all mesh. I 100% agree that sticking to pre-made figures is not good and forcing the writing to fit your premade figures is silly. All of my advisors agree with that statement too. | 1 | 201 | 1.192308 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuuhxj | gxuwq52 | 1,620,829,508 | 1,620,830,486 | 7 | 21 | It's highly field dependent, as well as dependent on the sheer volume of data you need to chew on (the more numbers you deal with, the more a graph becomes necessary to remotely grasp what the data is saying), but in my experience I'd say it's so obvious as to be less advice, and more "if you're not doing it this way you're literally doing it wrong." That's not to say that *every* graph and figure has to be made in advance, as occasionally you do run into a situation where you need a "novel" graph in order to illustrate some point. But generally speaking all of your major and/or "core" graphs should be made in the analysis phase, to the degree that you probably want all of your graphs to be made by your post-processing and analysis software. Granted, I'm in a field where the graphs genuinely don't change in terms of overall formatting from paper to paper, so I'm a bit spoiled, but still. | My experience comes from biology where not only is this a good strategy for writing a paper it's good for starting a project. Make up figures on a whiteboard of how the project could play out. Can you make sense of all the possible outcomes? Are there some answers that are better than others? If all the possible answers are interesting and interpretable then you have a good project design. If only some answers give you intelligible answers then you have a poor(er) design. Once you have data, drafting the figures gives you an outline of the paper. This is how people read papers. Often they just look at the figures to scrape the information they are interested in. It's the storyboard and the hard data that determines what you ultimately can say. | 0 | 978 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuuaja | gxuwq52 | 1,620,829,416 | 1,620,830,486 | 5 | 21 | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | My experience comes from biology where not only is this a good strategy for writing a paper it's good for starting a project. Make up figures on a whiteboard of how the project could play out. Can you make sense of all the possible outcomes? Are there some answers that are better than others? If all the possible answers are interesting and interpretable then you have a good project design. If only some answers give you intelligible answers then you have a poor(er) design. Once you have data, drafting the figures gives you an outline of the paper. This is how people read papers. Often they just look at the figures to scrape the information they are interested in. It's the storyboard and the hard data that determines what you ultimately can say. | 0 | 1,070 | 4.2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuuhxj | gxuxfa1 | 1,620,829,508 | 1,620,830,789 | 7 | 13 | It's highly field dependent, as well as dependent on the sheer volume of data you need to chew on (the more numbers you deal with, the more a graph becomes necessary to remotely grasp what the data is saying), but in my experience I'd say it's so obvious as to be less advice, and more "if you're not doing it this way you're literally doing it wrong." That's not to say that *every* graph and figure has to be made in advance, as occasionally you do run into a situation where you need a "novel" graph in order to illustrate some point. But generally speaking all of your major and/or "core" graphs should be made in the analysis phase, to the degree that you probably want all of your graphs to be made by your post-processing and analysis software. Granted, I'm in a field where the graphs genuinely don't change in terms of overall formatting from paper to paper, so I'm a bit spoiled, but still. | It’s shorthand for “make an outline (well)”. I find writing a paper is a dual process: constructing a narrative AND evaluating what, precisely, your data support. If you just go for the narrative, it’s all but inevitable that you will include conclusions and inferences not completely supported by your data. Mocking up a few figures helps with the dual task. It also reacquaints you with data that you may not have looked at in depth for some time. As others have said, it’s a starting point. The complex and nuanced stories we usually have to tell have a habit of changing on you as you pull them together. This is *one* technique to help you get to a shitty first draft. And by shitty I don’t mean poorly thought out or scraped off the floor - I mean a first pass at pulling together the threads of your work. ETA: YMMV. As I tell my trainees: until you find your own method and voice, borrow mine. It’s had some of the kinks worked out. | 0 | 1,281 | 1.857143 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuuaja | gxuxfa1 | 1,620,829,416 | 1,620,830,789 | 5 | 13 | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | It’s shorthand for “make an outline (well)”. I find writing a paper is a dual process: constructing a narrative AND evaluating what, precisely, your data support. If you just go for the narrative, it’s all but inevitable that you will include conclusions and inferences not completely supported by your data. Mocking up a few figures helps with the dual task. It also reacquaints you with data that you may not have looked at in depth for some time. As others have said, it’s a starting point. The complex and nuanced stories we usually have to tell have a habit of changing on you as you pull them together. This is *one* technique to help you get to a shitty first draft. And by shitty I don’t mean poorly thought out or scraped off the floor - I mean a first pass at pulling together the threads of your work. ETA: YMMV. As I tell my trainees: until you find your own method and voice, borrow mine. It’s had some of the kinks worked out. | 0 | 1,373 | 2.6 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv32jh | gxuuaja | 1,620,833,156 | 1,620,829,416 | 7 | 5 | I make all the figures and do all the analyses before starting writing. You need a story first and that comes from the interpretation of your results! | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | 1 | 3,740 | 1.4 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxv32jh | 1,620,831,505 | 1,620,833,156 | 2 | 7 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | I make all the figures and do all the analyses before starting writing. You need a story first and that comes from the interpretation of your results! | 0 | 1,651 | 3.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxv32jh | 1,620,833,039 | 1,620,833,156 | 2 | 7 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | I make all the figures and do all the analyses before starting writing. You need a story first and that comes from the interpretation of your results! | 0 | 117 | 3.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuuhxj | gxuuaja | 1,620,829,508 | 1,620,829,416 | 7 | 5 | It's highly field dependent, as well as dependent on the sheer volume of data you need to chew on (the more numbers you deal with, the more a graph becomes necessary to remotely grasp what the data is saying), but in my experience I'd say it's so obvious as to be less advice, and more "if you're not doing it this way you're literally doing it wrong." That's not to say that *every* graph and figure has to be made in advance, as occasionally you do run into a situation where you need a "novel" graph in order to illustrate some point. But generally speaking all of your major and/or "core" graphs should be made in the analysis phase, to the degree that you probably want all of your graphs to be made by your post-processing and analysis software. Granted, I'm in a field where the graphs genuinely don't change in terms of overall formatting from paper to paper, so I'm a bit spoiled, but still. | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | 1 | 92 | 1.4 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxuuaja | 1,620,868,512 | 1,620,829,416 | 6 | 5 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | 1 | 39,096 | 1.2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxxd2i8 | 1,620,858,408 | 1,620,868,512 | 5 | 6 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 10,104 | 1.2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxv7weo | 1,620,868,512 | 1,620,835,125 | 6 | 4 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | 1 | 33,387 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxxd2i8 | 1,620,846,300 | 1,620,868,512 | 5 | 6 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 22,212 | 1.2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxv7q4k | 1,620,868,512 | 1,620,835,055 | 6 | 3 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 1 | 33,457 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7z2e | gxxd2i8 | 1,620,835,155 | 1,620,868,512 | 3 | 6 | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 33,357 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxxd2i8 | 1,620,840,861 | 1,620,868,512 | 3 | 6 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 27,651 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvs3gy | gxxd2i8 | 1,620,843,403 | 1,620,868,512 | 3 | 6 | Works for me, everytime! | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 25,109 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxxd2i8 | 1,620,850,317 | 1,620,868,512 | 3 | 6 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 18,195 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxxd2i8 | 1,620,831,505 | 1,620,868,512 | 2 | 6 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 37,007 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxv2s7j | 1,620,868,512 | 1,620,833,039 | 6 | 2 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | 1 | 35,473 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv53kr | gxxd2i8 | 1,620,833,989 | 1,620,868,512 | 2 | 6 | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 34,523 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv8nl8 | gxxd2i8 | 1,620,835,433 | 1,620,868,512 | 2 | 6 | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 33,079 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxxd2i8 | 1,620,838,352 | 1,620,868,512 | 2 | 6 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 30,160 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvh2vv | gxxd2i8 | 1,620,838,857 | 1,620,868,512 | 2 | 6 | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 29,655 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxvoexn | 1,620,868,512 | 1,620,841,880 | 6 | 2 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | Its like storyboarding - you need initial figures to get your story in place, but no narrative goes through multiple writers/editors without things changing. I learned that it's easier to rip out any emotional attachment to the initial figures and be prepared to throw hours in exchange for a smoother writing experience | 1 | 26,632 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvvmqn | gxxd2i8 | 1,620,844,859 | 1,620,868,512 | 2 | 6 | It sounds like you are the one saying the figures can never change, which is an overly rigid interpretation. I would say ESPECIALLY for students it is important to start with the figures, and to see what the story you can build out of your data is. You don’t want to write a paper and then cherry pick only the data that fits what you wrote, you want to take a critical scientific look at your results and see what is true about them, even if some of it isn’t what you thought would happen, and write about that. | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | 0 | 23,653 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxw15su | 1,620,868,512 | 1,620,847,041 | 6 | 2 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | Yes! Make the figures first. The figures with their legends should be able to stand independently — it is all many people will read. The rest of the paper is just background, context and details for the figures. (and tables) | 1 | 21,471 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxxd2i8 | gxw35bs | 1,620,868,512 | 1,620,847,794 | 6 | 2 | Personally, by the time I am ready to actually write a manuscript, the data has been presented at conferences as either talks or posters. Meaning the figures already exist. Most data needs to be plotted pretty soon after data collection, otherwise the next steps you take are uncertain-- at least in my field. So the figures may not be polished, and I may get ideas for additional figures or details to add to current figures as I write, but for the most part, my figures come into being before my text does. I also like to outline papers, and a great way to do this is to plan what figures are needed to tell that story. Also, I pretty generally disagree with the statement: >But I have found that the story only comes together when pen hits paper... This sounds like a terribly inefficient way to write. This is why you present to your research group, in your department seminars, at meetings, in graduate student groups. If you can talk through your story first, tell it verbally, you will have a much easier time writing. Yes, the arguments in your paper become more refined and well developed as you write, but I don't see how you can even begin to write a paper with out a roadmap (at least in your head) of where you're headed. | I tend to think of my figures as my manuscript outline. I like to have them first so I know how the general flow of the paper will be and what data I’m going to include. Of course it’s fluid and sometimes I realize I want to talk about the data differently or group pieces together differently. But I always start with draft figures of some sort to guide me as I’m writing. | 1 | 20,718 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvfb23 | gxuuaja | 1,620,838,142 | 1,620,829,416 | 6 | 5 | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | It is excellent advice ... that should be thrown away once you actually start writing the paper. There are some parts of a paper you can write in advance, but you don’t really know what story you are going to tell until you have analyzed your data, and ideally the figures should tell the story on their own. Getting the figures right is therefore some of the most important work of writing the paper. BUT in the process of writing, sometimes you have insights that change the way you understand the data or how you want to present it, leading you to change the figures. I once wrote a paper and while putting together the talk about it I had a flash of insight about a new, better way to present the ideas. I ended up with a much better presentation based on completely different figures. The paper was mostly mathematical proofs and the talk ended up being mostly conceptual figures explaining the ideas in the proofs. Had I stuck with the original figures, it would have been a dud (honestly, I was even bored creating the original presentation). Instead, several people told me afterward it was their favorite conference talk. Bottom line: it is helpful to think about the story you want to tell and how to tell it through the figures before you start writing in earnest, sort of like storyboarding your paper, but be open to changing them as you go along. | 1 | 8,726 | 1.2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7weo | gxvfb23 | 1,620,835,125 | 1,620,838,142 | 4 | 6 | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | 0 | 3,017 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvfb23 | gxv7q4k | 1,620,838,142 | 1,620,835,055 | 6 | 3 | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 1 | 3,087 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7z2e | gxvfb23 | 1,620,835,155 | 1,620,838,142 | 3 | 6 | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | 0 | 2,987 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvfb23 | gxuz47a | 1,620,838,142 | 1,620,831,505 | 6 | 2 | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | 1 | 6,637 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxvfb23 | 1,620,833,039 | 1,620,838,142 | 2 | 6 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | 0 | 5,103 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv53kr | gxvfb23 | 1,620,833,989 | 1,620,838,142 | 2 | 6 | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | 0 | 4,153 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvfb23 | gxv8nl8 | 1,620,838,142 | 1,620,835,433 | 6 | 2 | >I tried to stick to pre-made figures and maintain the expectation that they weren't going to change you were being overly literal in how you interpreted their advice. They weren't saying that you can't ever make edits. | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | 1 | 2,709 | 3 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7weo | gxwrw9g | 1,620,835,125 | 1,620,858,408 | 4 | 5 | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 23,283 | 1.25 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxv7q4k | 1,620,858,408 | 1,620,835,055 | 5 | 3 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 1 | 23,353 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxv7z2e | 1,620,858,408 | 1,620,835,155 | 5 | 3 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | 1 | 23,253 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxvlyc1 | 1,620,858,408 | 1,620,840,861 | 5 | 3 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | 1 | 17,547 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvs3gy | gxwrw9g | 1,620,843,403 | 1,620,858,408 | 3 | 5 | Works for me, everytime! | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 15,005 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxwrw9g | 1,620,850,317 | 1,620,858,408 | 3 | 5 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 8,091 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxuz47a | 1,620,858,408 | 1,620,831,505 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | 1 | 26,903 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxv2s7j | 1,620,858,408 | 1,620,833,039 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | 1 | 25,369 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv53kr | gxwrw9g | 1,620,833,989 | 1,620,858,408 | 2 | 5 | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 24,419 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxv8nl8 | 1,620,858,408 | 1,620,835,433 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | 1 | 22,975 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxwrw9g | 1,620,838,352 | 1,620,858,408 | 2 | 5 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 20,056 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxvh2vv | 1,620,858,408 | 1,620,838,857 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | 1 | 19,551 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvoexn | gxwrw9g | 1,620,841,880 | 1,620,858,408 | 2 | 5 | Its like storyboarding - you need initial figures to get your story in place, but no narrative goes through multiple writers/editors without things changing. I learned that it's easier to rip out any emotional attachment to the initial figures and be prepared to throw hours in exchange for a smoother writing experience | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 16,528 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxvvmqn | 1,620,858,408 | 1,620,844,859 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | It sounds like you are the one saying the figures can never change, which is an overly rigid interpretation. I would say ESPECIALLY for students it is important to start with the figures, and to see what the story you can build out of your data is. You don’t want to write a paper and then cherry pick only the data that fits what you wrote, you want to take a critical scientific look at your results and see what is true about them, even if some of it isn’t what you thought would happen, and write about that. | 1 | 13,549 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw15su | gxwrw9g | 1,620,847,041 | 1,620,858,408 | 2 | 5 | Yes! Make the figures first. The figures with their legends should be able to stand independently — it is all many people will read. The rest of the paper is just background, context and details for the figures. (and tables) | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | 0 | 11,367 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxwrw9g | gxw35bs | 1,620,858,408 | 1,620,847,794 | 5 | 2 | >Based on my experience, this has been garbage advice. What the... no? Absolutely not. I'm sorry, but you're wrong about this being "garbage advice". It certainly isn't the only way to do things, but it's a good way to do it for a simple reason. What do you think the readers of your paper/thesis will look at? The (probably quite dull) technical word soup, or the pretty pictures? People give you the advice to make all the figures first for a simple reason - you need to tell your story with the figures alone. Someone needs to be able to understand your paper from just the figures, because they will read the abstract, read the conclusion, view the figures, then scan the body for details to answer the remaining questions they have about your work. We aren't writing novels here. The body of the text won't be entertaining. It should contain the details needed to reproduce our experiments, and the argument we're trying to make, written using words. But if that argument isn't obvious from your figures, people won't be willing to agree with your conclusions! The argument you write must, therefore naturally flows from viewing the figures, so producing them first will make your job a whole lot easier. If you think you should focus on writing the body of text and enjoy writing prose, good for you. Come back to me when you're writing your thesis literature review and let me know how many papers you read linearly, in their entirety, from abstract to conclusion! (It won't be many.) | I tend to think of my figures as my manuscript outline. I like to have them first so I know how the general flow of the paper will be and what data I’m going to include. Of course it’s fluid and sometimes I realize I want to talk about the data differently or group pieces together differently. But I always start with draft figures of some sort to guide me as I’m writing. | 1 | 10,614 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7weo | gxvz6ku | 1,620,835,125 | 1,620,846,300 | 4 | 5 | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 11,175 | 1.25 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7weo | gxv7q4k | 1,620,835,125 | 1,620,835,055 | 4 | 3 | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 1 | 70 | 1.333333 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxv7weo | 1,620,831,505 | 1,620,835,125 | 2 | 4 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | 0 | 3,620 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxv7weo | 1,620,833,039 | 1,620,835,125 | 2 | 4 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | 0 | 2,086 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv53kr | gxv7weo | 1,620,833,989 | 1,620,835,125 | 2 | 4 | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | Is it possible you're putting too much effort into the figures? My view is that I have a bunch of figures, like dozens, and it tells me which parts of the story are compelling, and then I write, and as I write I change figures and the story. | 0 | 1,136 | 2 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxv7q4k | 1,620,846,300 | 1,620,835,055 | 5 | 3 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 1 | 11,245 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7z2e | gxvz6ku | 1,620,835,155 | 1,620,846,300 | 3 | 5 | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 11,145 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxvz6ku | 1,620,840,861 | 1,620,846,300 | 3 | 5 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 5,439 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxvs3gy | 1,620,846,300 | 1,620,843,403 | 5 | 3 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | Works for me, everytime! | 1 | 2,897 | 1.666667 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxvz6ku | 1,620,831,505 | 1,620,846,300 | 2 | 5 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 14,795 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxv2s7j | 1,620,846,300 | 1,620,833,039 | 5 | 2 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | 1 | 13,261 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxv53kr | 1,620,846,300 | 1,620,833,989 | 5 | 2 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 12,311 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv8nl8 | gxvz6ku | 1,620,835,433 | 1,620,846,300 | 2 | 5 | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 10,867 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxvz6ku | 1,620,838,352 | 1,620,846,300 | 2 | 5 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 7,948 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvh2vv | gxvz6ku | 1,620,838,857 | 1,620,846,300 | 2 | 5 | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | 0 | 7,443 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxvoexn | 1,620,846,300 | 1,620,841,880 | 5 | 2 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | Its like storyboarding - you need initial figures to get your story in place, but no narrative goes through multiple writers/editors without things changing. I learned that it's easier to rip out any emotional attachment to the initial figures and be prepared to throw hours in exchange for a smoother writing experience | 1 | 4,420 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvz6ku | gxvvmqn | 1,620,846,300 | 1,620,844,859 | 5 | 2 | Nope, it's good advice. I should be able to read your figures alone and know what your conclusions are. Everything else is built around that. The writing (in any part of the paper) should not change your data or its interpretation at all. | It sounds like you are the one saying the figures can never change, which is an overly rigid interpretation. I would say ESPECIALLY for students it is important to start with the figures, and to see what the story you can build out of your data is. You don’t want to write a paper and then cherry pick only the data that fits what you wrote, you want to take a critical scientific look at your results and see what is true about them, even if some of it isn’t what you thought would happen, and write about that. | 1 | 1,441 | 2.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7q4k | gxuz47a | 1,620,835,055 | 1,620,831,505 | 3 | 2 | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | 1 | 3,550 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxv7q4k | 1,620,833,039 | 1,620,835,055 | 2 | 3 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | 0 | 2,016 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7q4k | gxv53kr | 1,620,835,055 | 1,620,833,989 | 3 | 2 | Our group does something called a "pre-paper talk", which is basically a research pitch we give before writing anything. This helps organize your thoughts and overall narrative, but only in slides used to communicate the ideas (which can sometimes be re-purposed into paper figures). This helps save time and effort from getting peer feedback early in the process, rather than waiting until after you've written an actual draft. YMMV based on your field and research practices, though. | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 1,066 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxv7z2e | 1,620,831,505 | 1,620,835,155 | 2 | 3 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | 0 | 3,650 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxv7z2e | 1,620,833,039 | 1,620,835,155 | 2 | 3 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | 0 | 2,116 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv7z2e | gxv53kr | 1,620,835,155 | 1,620,833,989 | 3 | 2 | If you use a reproducible approach (in R for example), it doesn't even matter because you can make quick adjustments on the fly if necessary. | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 1,166 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxvlyc1 | 1,620,831,505 | 1,620,840,861 | 2 | 3 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | 0 | 9,356 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxv2s7j | 1,620,840,861 | 1,620,833,039 | 3 | 2 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | 1 | 7,822 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxv53kr | 1,620,840,861 | 1,620,833,989 | 3 | 2 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 6,872 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxv8nl8 | 1,620,840,861 | 1,620,835,433 | 3 | 2 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | 1 | 5,428 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxvlyc1 | 1,620,838,352 | 1,620,840,861 | 2 | 3 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | 0 | 2,509 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvlyc1 | gxvh2vv | 1,620,840,861 | 1,620,838,857 | 3 | 2 | Do start with the figures, but remember Bobby Shaftoe's motto - display adaptability. They're not set in stone, just meant to provide tentative structure for the story. | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | 1 | 2,004 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxuz47a | gxvs3gy | 1,620,831,505 | 1,620,843,403 | 2 | 3 | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | Works for me, everytime! | 0 | 11,898 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv2s7j | gxvs3gy | 1,620,833,039 | 1,620,843,403 | 2 | 3 | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | Works for me, everytime! | 0 | 10,364 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvs3gy | gxv53kr | 1,620,843,403 | 1,620,833,989 | 3 | 2 | Works for me, everytime! | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 9,414 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv8nl8 | gxvs3gy | 1,620,835,433 | 1,620,843,403 | 2 | 3 | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | Works for me, everytime! | 0 | 7,970 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxvs3gy | 1,620,838,352 | 1,620,843,403 | 2 | 3 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | Works for me, everytime! | 0 | 5,051 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvs3gy | gxvh2vv | 1,620,843,403 | 1,620,838,857 | 3 | 2 | Works for me, everytime! | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | 1 | 4,546 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvoexn | gxvs3gy | 1,620,841,880 | 1,620,843,403 | 2 | 3 | Its like storyboarding - you need initial figures to get your story in place, but no narrative goes through multiple writers/editors without things changing. I learned that it's easier to rip out any emotional attachment to the initial figures and be prepared to throw hours in exchange for a smoother writing experience | Works for me, everytime! | 0 | 1,523 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxuz47a | 1,620,850,317 | 1,620,831,505 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | I think writing is an extremely personal process and you have to try what works for different people before finding your method. I don't like making my figures first, but I like making my section/subsection/subsubsection. I like writing my abstract first, even though many people say not to. Overall you need to write a cohesive story for your paper to be on its way to be good, but there's not one way that works. I think the "make your figures before writing the paper" is just another advice like this. Making your figures gives you an idea of where and how you want to take this paper. It gives you the story you need | 1 | 18,812 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxv2s7j | 1,620,850,317 | 1,620,833,039 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | I agree with most of the comments here, but in my experience, Figure 1 or maybe 2 (depending on the length of the paper) might change dramatically depending on the type of narrative and how it evolves during the writing process. These are usually intro figures and depend heavily on the written intro. Chemistry PhD candidate here (chem bio) which I think publication-wise should be relatively similar. | 1 | 17,278 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxv53kr | 1,620,850,317 | 1,620,833,989 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | My advisor makes us put together a slideshow presentation before we write anything. That way we easily move ideas around and change them while coming up with the "story" before we actually write. It becomes a visual outline. I'm not a huge fan of this approach but apparently it works. | 1 | 16,328 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxv8nl8 | gxw9c5y | 1,620,835,433 | 1,620,850,317 | 2 | 3 | I think it depends on how you write papers. Personally I find it very useful because it suits my writing approach, and this is how I usually do it. 1. Assuming that I have done all the analysis and looked at all the figures / tables, then the first thing that I will do is to think about the story that I want to tell based on the analysis. 2. Once I have that I will move on to the structure of the paper and think about what I want to include in the paper (and generate more figures / tables if necessary). 3. After that I put all of the figures into the paper, then start outlining what I want to say about each figure. I will go in as much details as possible at this step because I find it easier to think about the important takeaways from the figures when my mind is still fresh. 4. Once that's done, I will go and work on the Intro / Methodology etc. Additionally, because I have already analyzed the figures, I can write my Intro to cater it to the story that I want to tell. 5. Finally, I come back to the outlines that I have earlier and put them together into proper sentences and paragraphs. | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | 0 | 14,884 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvftud | gxw9c5y | 1,620,838,352 | 1,620,850,317 | 2 | 3 | I think it’s good advice tbh. It gives you a framework for writing. Obviously it’s flexible, so if you have a better idea, you can still change. | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | 0 | 11,965 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxvh2vv | 1,620,850,317 | 1,620,838,857 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | Yea my advisors also advocate this strategy. I'm in physics for context. The idea is that your figures should essentially tell your story, so once you have the idea of which figures your using and in what order, then you kind of have a skeleton built for the paper and you can fill in the details that make up the actual story. Of course, along the way you might find you need additional figures. It's just a method for getting moving in the right direction. | 1 | 11,460 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxvoexn | 1,620,850,317 | 1,620,841,880 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | Its like storyboarding - you need initial figures to get your story in place, but no narrative goes through multiple writers/editors without things changing. I learned that it's easier to rip out any emotional attachment to the initial figures and be prepared to throw hours in exchange for a smoother writing experience | 1 | 8,437 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxvvmqn | gxw9c5y | 1,620,844,859 | 1,620,850,317 | 2 | 3 | It sounds like you are the one saying the figures can never change, which is an overly rigid interpretation. I would say ESPECIALLY for students it is important to start with the figures, and to see what the story you can build out of your data is. You don’t want to write a paper and then cherry pick only the data that fits what you wrote, you want to take a critical scientific look at your results and see what is true about them, even if some of it isn’t what you thought would happen, and write about that. | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | 0 | 5,458 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw15su | gxw9c5y | 1,620,847,041 | 1,620,850,317 | 2 | 3 | Yes! Make the figures first. The figures with their legends should be able to stand independently — it is all many people will read. The rest of the paper is just background, context and details for the figures. (and tables) | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | 0 | 3,276 | 1.5 |
naoi57 | askacademia_train | 0.89 | "Make all the figures before you start writing the paper" is terrible advice Biomedical sciences student here. I am curious what people think of this statement. Personally, I have heard it from many PIs, including my own, leading me to believe it is a common bit of philosophy throughout academia. Based on my experience, this has been garbage advice. Sure, you need to know your data, and you need to have it analyzed before you start writing. But I have found that the story only comes together when pen hits paper, and some parts of the story end up needing to be emphasized while others can be ignored. In the past, I tried to stick to pre-made figures and maintain the expectation that they weren't going to change. Ultimately I found that this unnecessarily constrained the story I ended up being able to tell and became a waste of time since I just remake the figures anyway. | gxw9c5y | gxw35bs | 1,620,850,317 | 1,620,847,794 | 3 | 2 | It’s actually pretty decent advice. The PI isn’t telling you to make beautiful final figures that are submission quality. They’re advising you to lay out a plan for your manuscript. A rough ideas of what you want to present and how you want to present it. An outline of what figures you’d need is a great guidepost. You write around what you think you would present, and then adjust as you go. It’s a fluid and evolving process. | I tend to think of my figures as my manuscript outline. I like to have them first so I know how the general flow of the paper will be and what data I’m going to include. Of course it’s fluid and sometimes I realize I want to talk about the data differently or group pieces together differently. But I always start with draft figures of some sort to guide me as I’m writing. | 1 | 2,523 | 1.5 |
nc9x9b | askacademia_train | 0.96 | Not a professor but a graduate student. I would like to write a recommendation letter for my thesis advisor as she is an awesome mentor who really cares about students and is an inspiring scientist. I’m a PhD candidate in biology and in a research- focused university. I hope to seek some advice regarding writing recommendation letter for my thesis advisor to the Vice Dean in charge of faculty. My advisor ( relatively young PI) is an exemplary mentor who has inspired me in many ways in terms of scientific thinking and she truly cares a lot about students progress/and gives us both the space and support to develop our scientific minds. Therefore, I would like to write a recommendation letter for her in hope that it will be beneficial for her tenure consideration. My hesitation is that: how effective will this letter be in aiding her to get tenure? Or will it on the contrary, be something against her(?)- I am not sure how common is this practice of students writing recommendation letters for their advisors are in academia. | gy51dj3 | gy5db2j | 1,621,022,837 | 1,621,028,344 | 3 | 4 | In my experience, letters for exceptional mentors are best directed towards awards or funding for excellence in teaching at the University level. I wrote several for colleagues going up for a prestigious teaching awards, which are more helpful feathers in their cap (and occasionally with boosted funding) for T&P committees. Rather than writing the Vice Dean, I would approach your 1.) mentor about what will help their packet, and 2.) your dept. chair to say you want to nominate them for some award and contribute supporting material. As an aside, my experience on 3 different T&P committees at an R1 (college + university level) is that teaching was *very* rarely mentioned in review of application materials, unless the decision was close. Most people sail through based on the strength of their external letters, their publications, and their funding. Service and teaching is really ancillary and used for deciding weaker overall packets. | Everyone has touched on tenure, so I will also mention that universities and professional organizations often have teacher of the year and/or mentor of the year awards. Such recognition would also contribute positively toward tenure. | 0 | 5,507 | 1.333333 |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.