Speaker 1: Writing your first academic paper can sometimes feel a little bit overwhelming, and so in this video I want to break down the standard components of a paper and give a little bit of tips and tricks for how to get started writing your first paper. This video will hopefully also be useful for people who are reading papers, trying to understand why people are making the different choices in the different sections of a paper. Now in this video I'm not going to get into actually typing up your paper. I'm a big fan of using LaTeX to write scientific papers and I have an entire video on how to write a thesis using LaTeX, so I'd encourage you to check out that if you're interested more in the mechanics of writing out a paper. And then final thing before we get into the video, I want to briefly thank the sponsor for today's video which is Overleaf. Overleaf is a cloud-based LaTeX editor and just really wonderful for being able to write, collaborate, revise your papers, so I encourage you to check them out at the promo code in the description. Alright so we're writing a paper. The first thing you're going to see as a reader is the title. My main hint here is just to be descriptive about what you're actually doing. There's no need to be clickbaity. This isn't YouTube. I know I sometimes write some clickbait titles, but you want to sort of neither undersell or oversell what your paper is. You just want to tell people what it's about. Sometimes people can go on the side of adding way too much technical jargon right into your title that nobody understands unless they've actually read the paper. Try to minimize that and focus on like the keywords that people in this discipline are going to understand so that they can sort of get a quick sense of what this paper might be about. Next up is the abstract which is this sort of like short summary that happens right at the top of a paper. And an abstract is really important. An abstract is where people are really going to genuinely decide do I want to read this paper? Is this relevant for me or is it not? So you want to tell them that in the abstract. This is where you're putting your selling points. Why should someone invest the time to read your paper? Similar to what I said with the title, in an abstract you want to describe what you're doing accurately. Not overselling, not underselling, but you would ideally do it in a way that somebody who has not read your paper is able to follow along and understand what the idea is. For example using lots of notation and jargon that's sort of relatively specific to your paper and isn't widely understood by people who are in the field who might be trying to read this paper. People should just be able to get the sort of gist of what your paper is about quickly from reading the abstract. Personally when I'm trying to find papers to read in a subject I'm going to skim through a whole bunch of different abstracts until I find something that's actually relevant for what I'm interested in and then I'm going to go and read the rest of the paper. The abstract is the part that helps me decide. One possible structure for an abstract is sort of a what, why, how structure. What is it that you did in this paper? Why did you do this? Why was it interesting to you? Maybe it will thus be interesting to the reader as well. And how did you do it? What sort of tools were you using to accomplish this particular result? Next up is the introduction and I'm really a big fan of the introduction. Someone reading an introduction has now read your abstract, they're invested in this paper, they want to know what you're writing about and you need to start taking them on a journey so they can understand every step along the way. Introductions often have a summary again, it's just a longer and more developed summary than the brief abstract was. For example it's been common to add a lot of background information and context, not new stuff that you're doing in the paper, but the relevant setting by which the paper exists compared to sort of other established results in the literature. You want to sort of catch the reader up on the definitions and theorems and results that are standard in the literature to which your paper is building on. And introductions are really valuable to read as well because you get to learn not just the result of the paper but the entire sort of broader landscape. It's like a little summary of this corner of research and the author has done the legwork to try and explain that to you. The references specifically in the section can be extremely important about finding out what are the big and important major results in this field, so I really like reading my introductions carefully. Introductions also commonly expand on the motivation, like why were you doing this, why were you writing this paper, what interested you about it. It goes back to this what, why, how mechanism I mentioned from the abstract. Knowing why you were doing a result is extremely important. The introduction is also the spot where you're establishing notation that you're going to use throughout your particular paper, jargon that's either you're introducing or sort of very specific and technical jargon. This is sort of catching the reader up so they'll be able to follow your main arguments later on in the paper. You sort of set the stage for all of that here. And then the final thing that's common introduction is just an outline of the paper. This is just a description of sort of the different major sections of the paper so someone can understand how is this organized, where are the different components. An outline is sort of like a descriptive table of contents. All right, so now we go to the body. The body is the most important part of your paper. It's where the thing that you actually are writing about goes. So whatever you have, if you're doing mathematical proofs, you're doing some sort of analysis, you're doing an experiment, all of that is going in the body. The standards for a body are going to depend a lot on the discipline and the audience and the type of paper that you're writing. So I just want to give just a few sort of pretty general points. The first is I'm a big fan of subdividing my body into a bunch of different sections. It shouldn't just be page after page after page sort of sort of of exposition. You want to logically break it up and in fact you want to have a logical flow between the different sections so that someone can sort of understand what you're doing and how you're breaking up whatever it is that you're writing about. These sections you might have already described a little bit in the outline that was done in the introduction, but nevertheless the point is I really want you to think about how to make the body accessible to an audience. This should be readable. This should be understandable. Somebody who's gone through the introduction and got that sort of groundwork should now be able to for the most part figure out what's going on through their body. I definitely take the view that papers should be understandable, that part of our goal of writing a paper is not just to establish the truth of a result, but so that people are able to read our papers, understand it, and benefit from them. Of course you should be focused on being rigorous and meeting the standards of the discipline, but I think this readability piece is important nevertheless. One thing that's often in a body is a discussion. This is sort of after the major results have sort of been determined and now we're going to have a discussion about those results, contextualizing the results, analyzing those results, and so forth. Next up we have a conclusion, which isn't quite the end of the paper very often, and the point of a conclusion is to once again summarize, but what you shouldn't do is just copy and paste the summary that you already wrote in the abstract, because they have very different purposes. For the abstract you're assuming somebody hadn't read your paper, didn't know all the technical jargon, wasn't completely familiar with all of the nuances of the paper, but the conclusion is now summarizing for someone who knows way more. They've read your paper and now you're putting together the big ideas and summarizing it for that person. You could also put in your conclusion, and sometimes in the body as well in the discussion, future work. So you might want to say, well this is what I've done and this is what could be done in the future. This helps other researchers look at what's being done and get ideas for moving forward in the future, and you can take those ideas of course and work on those yourself as well. After the conclusion there's an optional section. This is going to be appendices, and there's all sorts of different things that you'll sometimes find in an appendix. For example, one thing you might find is like side details. Interesting things that weren't quite worth going into the body, maybe they're sort of too technical or not quite as important, so they can go into an appendix. You'll sometimes put in the appendix introductory material to content that's not really well known. So this isn't new, this isn't something that you've added, but nevertheless you want to have a spot so that a reader who doesn't know about this can get up to speed, and there maybe isn't a sort of a great standard other thing that they can go and read, so you might put that introduction here. You might put elements of your code here, or extra computations, extra graphics, it could be all sorts of stuff that go in an appendix. And basically the the lens that I use here is that they would be somehow sort of disruptive or tedious or just sort of too long to put in the body of it, but they should still be somewhere and so you put them in the appendix. I'm actually a really big fan of people who put code in their appendix. This allows someone to come in and sort of build on your work and sort of emulate it. It just lowers the barrier to do that. They could read the body and be like oh those are great ideas, but now they have the code to actually go and quickly implement it themselves. Final thing is references, and oh my goodness references are so important, so I really encourage you to do these well. Well the list of references at the end. The truth is that references should be sort of interspersed throughout the paper, and they sometimes have a lot of density. For example in the introduction you might have a lot of references, you're sort of setting the stage and the context, but throughout your paper if you're using something that somebody else has come up with you should have a reference for it. And it's extremely important for a reader here because a reader isn't just going to read your paper, they're sourcing and identification of references that you've done now can guide them on their journey so they can learn new things at the appropriate spot. So definitely really work on having a nice full-fledged level of references to help support your paper. When it comes to actually writing out your reference section, just a few tips. First tip is to have a consistent style. I actually don't care which style it is, although some publishers care about what style it is, I just want it to be consistent. I'm a huge fan of using reference managers. There's many of them out there like Zotero and Mendeley. This kind of software allows you to manage all of your references, keep track of them, search them, and then actually incorporate them into your papers really really easily. It's important that you actually refer to your references in the paper itself. That is, you don't want just a list of papers that you've personally read. The point is that when you're in the paper and you want to signal to the reader that there's more information here or this result came from somebody else, what is the thing that you're actually referring to, that is now going to be down in this list of papers at the end. And as a reader of a paper, references are extremely important because you can figure out where you're supposed to go next, what is the next step in your learning journey. The fact that someone sort of identified the relevant references in their field is extremely important. And then final thing, not to be too pedantic about this, is that you should actually read your references. Sometimes we can sort of find a quick abstract and we only read the abstract and we just throw it in to sort of build out our reference list. Don't do that. If you spend the legwork to really read through your references, understand them, build your context, it's going to improve your understanding, it's going to improve your paper substantially. So do indeed read your references. All right, so I think we've now talked through the major structural components of a paper. Of course, the exact standards are going to depend a little bit on your discipline, but this is just some general principles that we've talked about in this video. I do want to thank Overleaf for once again sponsoring today's video. Overleaf is a wonderful cloud-based LaTeX editor that's worked with me for my entire playlist on LaTeX. And actually, the free version of Overleaf is totally great just by itself. Anything you realistically want to do for writing up your paper, you can do entirely in the free version. Nevertheless, I am a fan of the paid version. You can get a promo link down in the description. And the reason is the paid version adds a bunch of quality of life features on top of all the basics of the free version. Things like being able to collaborate and leaving comments, tracking the changes, and seeing sort of version history. Even things like being able to integrate with the reference managers I talked about earlier. So check out the link down in the description. And with that said, thank you so much for watching. If you have any tips or tricks about writing papers, leave those down in the comments and we'll do some more math in the next video.