# Is it worth it to master LaTeX or MS Word?

I want to know, if I submit a paper to a journal, will they change the formatting of the paper? For example, will they change the position of the tables or the figures?

If they change these, then I think it is not necessary for me to learn MS Word or LaTeX in depth. As long as I can write the paper in a readable manner, it should be fine because the journal will make the format better if the paper is accepted. But if there is a specific style and format of the papers published in the journal which must be fulfilled by the authors, then it is a different story.

My field is physics and biophysics.

• How do you produce your Tikz figures unless you know LaTeX quite in depth? Kidding a bit with that, but it's worth mentioning that LaTeX (and possibly, Word, too) can bring other benefits beside properly formatting your document or positioning of floats. – O. R. Mapper Oct 19 '15 at 20:29
• There's a long range between "mastering" and "being reasonably proficient in". Make sure to learn the things that actually make your work easier - like using proper indenting and paragraphing. Yes, I've seen papers that use spaces for indenting and endlines for paragraphs (it's usually preserved in the PDFs) - the problem isn't that they're rejected, it's that you're wasting tons of your time for no reason. The requirements of the individual journals will tend to vary - you'll want to check their individual guidelines before committing yourself. But removing formatting is usually quite easy :) – Luaan Oct 20 '15 at 7:58
• "Is it worth it to master Latex or MS Word?" Yes to the first, no to the second. – Pouya Oct 20 '15 at 10:44
• You certainly do not need to master LaTeX to be able to use the journal templates. The nice thing about LaTeX is that it tends to produce very good output by default. If you are content with the template's default, then no need to learn it in depth. But you should learn enough to at least be able to write proper equations, tables, figure captions, tell the difference between \langle and <, use references and BibTeX ... You don't need to be able to work with low-level boxes or write LaTeX packages though. – Szabolcs Oct 20 '15 at 12:20
• @Pouya: outside of some specific academia, the word LaTeX is not known, not ot mention that this is software, not to mention that this software does typesetting, not to mention that the word typesetting is also unknown. In contrast to that Microsoft Word usually rings a bell, or is widely used. – WoJ Oct 20 '15 at 21:22

This depends entirely on the journal policy, which in turn tends to depend strongly on field. For example:

• In computer science and electrical engineering, many ACM and IEEE journals expect papers to be turned in using a format that is very close to publication and then re-format little if at all. For these journals, good mastery of LaTeX is a must.
• Many biology journals, by contrast, will refuse to take nicely formatted submissions and require text to be submitted separately from figures. For these journals, you mostly just need to know how to use your citation manager.

Beyond looking at individual journals, I would say that a good rule of thumb is to consider how many equations you are likely to publish. The more mathematical your work, the more likely that your publication venue is to have embraced LaTeX and the more benefit expertise with it is likely to be to you.

• This - it will depend strongly on the journal and field. You could have a long, thriving and productive career in my field and never submit a thing in LaTeX. – Fomite Oct 19 '15 at 22:55
• That being said, LaTeX is a pretty nice skill to have all around – David says Reinstate Monica Oct 20 '15 at 3:29
• LaTeX is good for more than just equations - it is a very good tool too if you publish exactly zero equations. Ironically, LaTeX's strength is really more its labelling and referencing s well as the abundance of dedicated tools (packages) to do specific jobs. You can produce high quality schematics in LaTeX, no chance in Word, you can draw chemical molecules in LaTeX, no chance in Word, etc. - As a LaTeX user myself, I even pick it for little stuff that has little need for equations - because it is the better tool for writing documents. – DetlevCM Oct 20 '15 at 9:39
• @DetlevCM And, you know, high-quality typesetting. Something that Word et al. are notoriously bad at. – Raphael Oct 20 '15 at 15:02
• What @DetlevCM said. It's the better tool to write documents. As in, structured documents. – Tobia Tesan Oct 20 '15 at 17:39

tldr; You use what your co-authors use.

• Follow-up question: How can there be any progress then? I mostly follow this advice for writing manuscripts and are stuck with Word (but have basic knowledge of LaTeX). But luckily I didn't follow this advice when analyzing data or I would still be stuck with Excel (and probably wouldn't have my current job). – Roland Oct 20 '15 at 12:05
• That's a chicken-egg answer. Also, most people start using tools before they have any co-authors. I myself (and some friends) started using LaTeX in our third semester. – Raphael Oct 20 '15 at 16:18
• @Roland I suppose what I really wanted to say is, "you use at least what your co-authors use". Of you feel you need more, of course, you go forward and learn new things as well. Thank you. – Dmitry Savostyanov Oct 20 '15 at 19:39
• +1 for collaboration but and (it's a big but), you have to do what's appropriate for the paper as well. For example a paper that contains more than the usual amount/complexity of equations for your field, written with a co-author who doesn't use LaTeX and relies on word's "track changes" feature. Or even a paper with a lot of citations and cross-referencing. In my case the (senior) co-author in questions was amenable and I wrote the text in word with LaTeX citation/cross-referencing commands in it. (and pasted equations in as images; they didn't change). The final submission was LaTeX. – Chris H Oct 21 '15 at 8:36
• @KonradViltersten "tldr;" section is a one-line answer, written for those who can find my "Long answer" to be tldr; – Dmitry Savostyanov Oct 22 '15 at 10:13

Somewhat echoing other comments and answers: no, don't think of "mastering" any particular type-setting software, but, on the other hand, yes, it is probably worthwhile to become sufficiently proficient in whatever you choose, so that you do not spend much cognitive energy on typesetting per se, as opposed to content. The particular choice of typesetting system depends on your context, of course, ...

(I myself was coerced to learn some details of (La)TeX in order to satisfy the late-adapter/adopter-incompetence of some publishers in the early 1990s. That is, I "needed" to figure out how to do really-low-level things like move the location of the page-number... This was and is a stupid way to spend time, but maybe unavoidable in some contexts... Similarly, had to have everything sufficiently controlled in pointlessly low-level ways to satisfy the whims of English-major-B.A. editors... so had to retreat from LaTeX to plain TeX to have access to silly things... My advice: do not do any such thing, anymore. Even the heel-draggers have caught up.)

Again, an implicit question somewhat in reaction to "do I have to master...?" is "do I have to have a clue...?", and the answer is "yes". Do not let inability to typeset your work be a noticeable bottleneck.

Another sleeper question: "when you're a novice, should you conform to the formatting and typesetting conventions of the ambient academic culture... or is it ok to get creative...?" :) Well, I tell my own students that conformity is certainly not a high virtue, but it is obviously a convenient virtue, e.g., if one wants to avoid routine dismissal as a crank/crackpot. This is not entirely ridiculous, in fact, I think, since demonstration of awareness of the rituals of a (inevitably, social) group is a positive signal to that group.

• If you ever have the formatting issue now, fancyhdr will help. – DetlevCM Oct 20 '15 at 9:40
• Word is not a type-setting software. – Raphael Oct 20 '15 at 15:03
• @Raphael, isn't "Word" just WYSIWYG typesetting software? There are WYSIWYG versions of TeX around, too. – paul garrett Oct 20 '15 at 16:14
• Well, the answer certainly depends on what you mean by "typesetting". Word (and other word processors) sets letters on "paper", yes. Does it handle any of hyphenation, widows and orphans, (pleasing) justification, kerning (using font metrics), ligatures, or others? No, at least not to my knowledge. – Raphael Oct 20 '15 at 16:17
• @Raphael, I didn't pretend "Word" and such were good typesetting softwares! :) – paul garrett Oct 20 '15 at 17:15

I think it depends on what kind of "depth" you are talking about.

You'll certainly want to learn at least one of those systems, to a level of competence that allows you to produce documents that are of comparable quality to your field's standards, and that meet the specifications that journals in your field require.

This does not necessarily require "mastery", just comfort with basic features. Indeed, knowing too much about the system and how to tune it to create very specific and/or unusual effects is probably not helpful. It's much more important to be standard than to be fancy. When very specific formatting or effects are required, journals will often either provide templates and/or style files, or simply do their own formatting.

Some features of these applications are not so much aimed at producing particular effects, but at making it more convenient for you to use them (e.g. citation managers, packages for formatting particular kinds of diagrams, etc). These are things you can pick up to the extent you find they are helpful.

You should be aiming to produce a typeset manuscript of the highest possible quality, regardless of whether the formatting is going to be changed by a journal. Though I can't give a really convincing argument for this, it is something that I passionately believe in, and I hope that the following list of half-reasons might go some way to convincing you.

• It will make you feel good about the research you've done. There are few things more satisfying than putting the final touches to a beautifully typeset piece of work. You will feel proud of the document you've created - and this pride will carry over to the research that you're writing about.
• The formatting and typesetting is an important part of your paper. Nowhere near as important as the scientific content, of course, but when somebody reads your paper, the way the words sit on the page is an important part of the way the information reaches them off the page. Since the primary goal of a research paper is dissemination of information, you should not disregard the matter of how easy the paper is to read. Of course, if the journal messes up the formatting completely then this could become irrelevant, but a good journal typesetting policy should emphasize exactly the same things that are important to you - clarity, elegance and readability - so if you take these things into account, you can be sure that you are producing something as close as possible to what the journal will eventually be printing. If you leave the whole thing as a 'clean-up' exercise to the journal, then you are giving up all control over an important part of the message of your paper.
• A rigorous approach to typesetting encourages a rigorous scientific approach. This is mainly relevant if you've got into the (excellent) habit of writing up your research as you go. Learning the best ways to break documents up into sections and recombine and rework them together (part of the 'mastery' of any typesetting system) is an excellent crucible for organizing your research in a truly rigorous and flexible manner. One of the best things about a system like LaTeX (and even MS Word) is that it encourages creating a fully granular, extensible document with different sections that can be moved around if they need to be or improved upon at some stage. This is exactly the attitude that you want to have with your research. If you neglect to master your typesetting system then you run the risk of typing out one large sequential and inflexible document, and this can prejudice your view of your own research in a bad way.

As I say, these are all fairly 'touchy-feely' arguments, and none is convincing on its own, but I strongly believe that together they give a good reason for taking typesetting as seriously as you would if you were preparing a manuscript for direct inclusion into a journal.

The question remains: what typesetting system should you use. As you can probably expect, I recommend mastering LaTeX, as it will give you the largest formatting reward for a relatively small effort. MS Word mastery is not a bad skill to have, but as a scientist I think you will appreciate the direct control over what is happening in LaTeX (and the knowledge of what is going on) over the sometimes cryptic "let's move this diagram over here for no reason" attitude of MS Word. It's more effort to learn, certainly, but you'll soon realize that it was worth it. Sooner than you think, you'll reach a point of fluency in LaTeX where the time you spend formatting your article will be, while considerable, only a small fraction of your research time. That is the ideal situation to be in.

• +1, although "let's move this diagram over here for no reason" still happens quite frequently in my LaTeX documents. – Mangara Oct 21 '15 at 0:31
• @Mangara: LaTeX figure placement is well-documented and customizable to some extent (though the algorithms aren't easy), unlike Word. But it's indeed somewhat advance LaTeX-fu. – Blaisorblade Oct 21 '15 at 2:05
• In your last bullet, are you claiming that one can't cut-and-paste a section to another position of the document in Word? – Federico Poloni Oct 21 '15 at 13:40
• @FedericoPoloni No. That bullet is not anti-Word at all. Rather, it is answering the OP's question: Should one seek to master one's typesetting program (be it Word or LaTeX)? Word has perfectly good mechanisms for moving text around and structuring documents, as I said in that bullet point. You might disagree that cutting and pasting in Word counts as 'mastery', but I think more advanced means of document (re)structuring in Word do come under that category. – John Gowers Oct 21 '15 at 15:05
• @FedericoPoloni Did you read my comment? I said nothing about Word vs. LaTeX. Indeed, this was one of the positive things I said about Word - it has perfectly good mechanisms for restructuring. – John Gowers Oct 21 '15 at 21:44

First of all, the publisher will almost certainly change the format of your paper. The extent of the changes, however, vary with publisher. Some publishers prefer LaTeX because the manuscript is close to final form. Other publishers, however, prefer MS Word because their workflow has been built-up around Word documents.

You should become proficient in both MS Word and LaTeX. Here's why.

Reasons to learn LaTeX:

1. You're a physicist; that should be reason enough.
2. The preprint of your paper (for your personal website, arXiv, etc.) won't look like a manuscript draft.
3. You can make great looking CVs, letters, dissertations, grant proposals, etc.
4. LaTeX defaults are nice, Word defaults are pretty horrible (tables, margins, figure captions, etc.)

Reasons to learn Word:

1. Many of your colleagues will likely use it, thus you will have to as well.
2. Some journals may prefer submission in Word.
3. The better you know Word, the less you'll have to fight with it when it starts auto-formatting everything and moving your figures around.

If you want to keep your manuscript simple while maintaining flexibility in the output, consider writing in Markdown, then converting to whatever format you like using pandoc.

From your question history and history here, it sounds like you're just starting out in research. Are you planning to write a thesis at some point?

Even for a Masters thesis there are some advantages to using LaTeX, for a science PhD thesis it's indispensible for the following reasons:

• Equations (even simple ones will soon become much easier in LaTeX)
• Cross-referencing of figures and sections within a document: learn a couple of commands and it just works.
• Citations, especially if your field uses a numeric citation style (would you want to renumber 200 citations, some used several times, on the day you send it to the printers?)

There are tools to do most of this in word. Some of them are OK, some are free, but most are rubbish and/or expensive compared to the stuff that's built-in for free in LaTeX.

The relevance of this to paper writing is that learning to use LaTeX specifically for writing a thesis is too hard, too late. Write a few papers in it, maybe the odd internal report as well, and it will be easy by the time you're writing your thesis.

Finally don't miss the very helpful tex.stackexchange.com, where there are plenty of getting-started questions.

It's good to know LaTeX, that will help you a lot, and is arguably essential in science. I agree though, that if you absolutely master Word, that would allow you to make similar looking papers, and the truth of the matter is that Word is extremely powerful, is used to its fullest. Take that as you will.

• "if you absolutely master Word, that would allow you to make similar looking papers": there is by no means any way how Word can even be compared to LaTeX. – gented Oct 19 '15 at 22:00
• @GennaroTedesco Of course they can be compared. We show that LaTeX users were slower than Word users, wrote less text in the same amount of time, and produced more typesetting, orthographical, grammatical, and formatting errors.December 19, 2014 study by Markus Knauff and Jelica Nejasmic Our study shows that each document preparation system has unique advantages and disadvantages, and there might be no “best” tool for all aspects of a highly complex task such as producing diverse scientific publications. journals.plos.org/plosone/article?id=10.1371/… – Dan Oct 19 '15 at 22:26
• @Dan You quote a study with 40 participants on academia.stackexchange.com? No more questions, your honor... – Robert Oct 19 '15 at 22:42
• "is arguably essential in science" Starting to sound like a broken record here, but not in all parts of science. – Fomite Oct 19 '15 at 22:54
• @Robert It was literally the first search result, and it supports jakebeal's response. I figured a 40 participant study was better than any one person's opinion. – Dan Oct 19 '15 at 23:02

In many physics subfields, colleagues will expect you to put your preprints on the arXiv. You may also post them on your website or an institutional website. Those will appear in the way that you format them, without any help from a professional editor.

You're a physicist? Then it's worth it to spend the time to learn LaTeX to a reasonable level. There are a bunch of templates for doing all sorts of neat diagrams about quantum states, those silly operator brackets, and what-have-you. (I'm obviously not a physicist...) None of that in MS-Word - as far as I know.

As for the MS-Word option - these days you have LibreOffice Writer, which is quite powerful, stable, feature-rich and cross-platform. For some cases/issues it has not fully caught up with MS-Word, but it's quite sufficient as your default.

Now, if you're lazy but you want LaTeX-style math at least you could use one of the MS-Word/LibreOffice plugins or extensions or OLE apps which allow this. For MS-Word for example you have Design Science Equation Editor.

Finally, if you're not lazy and actually practice seriously with at least one of these, you'll find it a lot easier to use the other (especially LaTeX -> LibreOffice), because that will get you used to defining themes and styles and sticking to them rather than the "make-this-line-bold, now Enter Enter Enter to get more space" you see people doing.

learn MS Word or LaTeX in depth

Not sure how much understanding is "in depth". If your paper doesn't have any equations, then 1 day is already too much for you to learn Latex from scratch.

Learning Latex is just like learning to type with 10 fingers, or learning to use vim. At first, it seems to take more time, but when you get familiar, you will work in a much more efficient way. For example, maintaining references, links etc in Word is a pain, but in Latex it is just a piece of cake.

However, if you decide to type with 2 fingers in your whole life, you will still be fine. Word is much easier to learn, so it is more popular outside academia and the fields where computer skills are not important, e.g. Humanity and Social Sciences.

• I realized positioning figures and tables at the place I want is harder than Word, and when Latex does it automatically, I don;t like the result. So, I don't think learning Latex is like learning to type with 10 fingers. – MOON Oct 21 '15 at 7:58
• if you want the figure to stay where it is, simply put \begin{figure}[htp]. That's it. – qsp Oct 21 '15 at 8:05
• Didn't work well for me. – MOON Oct 21 '15 at 8:07
• Not sure what you want. You can ask a question here: tex.stackexchange.com – qsp Oct 21 '15 at 8:12
• then 1 day is already too much for you to learn Latex from scratch — Presumably you mean "then you can learn latex from scratch in considerably less than 1 day". – JeffE Oct 21 '15 at 9:13

"Is it worth it to master LaTeX or MS Word?"

Just wanted to echo Dmirty's sentiment, but also mention that LaTeX proficiency is a bit of a niche skill. If you are good at it, it can potentially open other doors for you; I found my current job thanks in part to this. If you like LaTeX and find that you are good at it, it is worth your time to master it since these skills may pay off in other ways. However I cannot really say the same about MS Word, its such a ubiquitous program that you are expected to have some proficiency with it in academia.

• Word is definitely not something you're expected to be proficient with. In my experience, it's legacy software used by administrators. Documents produced with it generally don't display correctly for non-administrators. Word itself doesn't run at all on Linux, and the Mac version is extremely slow. I've heard rumors that Windows can be used outside administrator offices and gaming PCs, but it's been a while since I saw any evidence for it. – Jouni Sirén Oct 20 '15 at 21:41
• a bit of a niche skill — Unless you work in a field that uses it for everything, like math or physics or theoretical CS. I only use Word for administrative tasks. – JeffE Oct 21 '15 at 9:15
• @JeffE Of course, far more people work in administrative tasks than in math/physics/CS. – Ypnypn Oct 21 '15 at 20:24