64

In my past experiences, I have almost never typeset my manuscripts according to the formats required by the journals to which I would like to submit. I leave my manuscripts as produced by the LaTeX article documentclass.

Recently I am wondering: Would such a behavior generally give handling editors a negative first impression?

Yes
  • 7,654
  • 8
  • 44
  • 76
  • 3
    A negative first impression, probably. But as long as the paper is professional-looking, usually not a big deal. Of course if the paper is accepted, you will have to re-do it in their format. – GEdgar Jan 12 '15 at 15:19
  • 1
    While the page limit might not be as tight in journals as in other types of venues and thus this point in favour of using the required format is not as important, you may want to specify what you mean by "format". Some parts of the styleguide that describes the "format" reach into how you structure your contents. For example, some "formats" discourage or disallow the use of footnotes, some formats require introductory paragraphs to separate a heading from a sub-heading, while others forbid exactly that, etc. – O. R. Mapper Jan 12 '15 at 15:42
  • 9
    My recent experience suggests that sometimes a paper won't even be accepted for review unless it fulfills the necessary typesetting requirements. – liori Jan 12 '15 at 21:28
  • 15
    Some journals make it explicit that this is not necessary. For example, Transactions of the AMS describes the use of their style files as something to be done after acceptance. – Nate Eldredge Jan 12 '15 at 21:31
  • 3
    I consider it a good thing if a journal has a template that they wish you to conform to, as they will usually supply one that simplifies things for you, especially if using LaTeX. I see it as taking format/typography decisions out of my hands, so it is one less thing to worry about. Just use their template, and it's as they want it. – Jool Jan 12 '15 at 18:48
  • 17
    For what it's worth: I've never bothered, and it's never mattered. – JeffE Jan 13 '15 at 01:03
  • 12
    I think @Chou deserves a lot more up-votes for this question: reading through the answers and comments below, there appears to have a remarkably strong and passionate divide in how different fields think about this question. – jakebeal Jan 13 '15 at 12:50
  • 11
    A reason not to format according to the journal is when you post your paper on a public repository or your web page (which should be strongly advised in fields where journal usually allow this). If you post a firt version in journal A's format, get your paper rejected there, and finally publish in some other journal B, it would be weird to have the preprint formatted as if it where a paper in journal A. – Benoît Kloeckner Jan 13 '15 at 14:55
  • The market for software to handle this is getting crowded, e.g. https://typeset.io/ and https://www.overleaf.com/ – Nemo Aug 02 '18 at 11:48

8 Answers8

50

I'll venture a minority opinion (I'm in mathematics, where the culture is very possibly different than in other fields).

As a referee, I have an instinctive negative reaction if I know that an author has taken the time and effort to conform to a particular journal's style. The grounds for this is that most journals employ typesetting staff for this purpose. Especially considering the very high price of many (if not all) journals, for authors to refuse to do this suggests to me a principled refusal to waste their time.

That said, I can see that many people hold the exact opposite opinion, and even that my own feelings may be a little bit silly. So I certainly don't actually hold this against authors when evaluating submissions. (Indeed, if I receive something formatted, I never know if it is the author or the editorial staff that has formatted it.)

My impression is that most (but maybe not all) mathematicians wouldn't hold it against you if you don't bother. Moreover I believe that most mathematicians in fact don't bother with such formatting guidelines.

I have never heard anyone voice @jakebeal's opinion before. Of course, counting the upvotes, he speaks for at least eight other people! You might take this as evidence that the answer to your question is dependent on what your field is.

Anonymous
  • 26,161
  • 6
  • 74
  • 121
  • 7
    I believe that our two views are actually compatible: please note that I refer to the prescribed submission format. When a journal handles its own typesetting in detail, it typically does not prescribe a particular submission format. This is typical of biology and chemistry (and apparently mathematics, by what you say). Many major publishers in engineering and computer science, however (e.g., IEEE, ACM, AAAI, Springer), minimize their typesetting workload by specifying a particular submission format and providing appropriate templates. – jakebeal Jan 12 '15 at 22:33
  • 5
    Using a template is not that difficult as most mathematicians would think. – Greg Jan 13 '15 at 01:18
  • 7
    "Moreover I believe that most mathematicians in fact don't bother with such formatting guidelines." I think so too, and as you say, the few who do look a bit naive. – Pete L. Clark Jan 13 '15 at 03:44
  • 1
    I feel it is presumptuous to put a paper in the journal's format before the paper is accepted - it suggests that the submitter thinks the acceptance is a "done deal". Also, see the comment by Benoît Kloeckner above. – Oswald Veblen Jan 13 '15 at 18:12
  • 8
    @Greg: like most mathematicians, I have put numerous math papers into LaTeX templates for journals, and I know how hard it is. For me, it usually takes about an hour or two of time, sometimes a little more. It requires fussing with packages, citations and (taking more time) tables and line breaks. But revisions the referee suggests will also require fiddling with line breaks in many cases. So it is more efficient to wait for the paper to be accepted, when there are no more changes, and then put the paper into the format for the journal that accepted it and fix the lines breaks only one time. – Oswald Veblen Jan 13 '15 at 18:21
  • 1
    I use macros in my latex code to shorten (non-standard) things I type often. I strip these out with copy/paste before sending the code to a journal, but doing so before acceptance would be an annoying way to have to use my time. – Jessica B Jan 13 '15 at 18:46
  • 2
    (responding to the answer) I'm confused, I really don't understand your reasoning for considering an author's effort to follow a particular formatting style to be a strike against them. You say something about "refusing to do this", but what is an author who follows formatting guidelines refusing to do? – David Z Jan 13 '15 at 23:28
  • 1
    I am puzzled exactly for the same reason as @DavidZ: what is the this authors are refusing to do that elicits your negative reaction (I'm asking because I'm a very obedient person so I always at least try to conform to submission guidelines when they are provided by a journal)? – Olivier Jan 14 '15 at 09:42
  • 2
    Here is an example of the kind of instructions which I advocate ignoring: http://worldscientific.com/page/ijnt/submission-guidelines For example, they specify that footnotes should be a, b, c, but in some other journals the footnotes should be 1, 2, 3, or there should be no footnotes at all, and I think it is common for authors to worry about this only after the paper is accepted. Moreover, in my experience the detailed instructions linked at the bottom are moderately time-consuming; I had to follow them once, but only after my paper was accepted there. – Anonymous Jan 14 '15 at 17:34
  • 1
    @Anonymous I see why you would advocate ignoring the guidelines, but still, I don't see why the fact that an author chose not to ignore them would give you a negative impression of the manuscript. – David Z Jan 14 '15 at 19:23
  • 6
    Not really negative. Negative in the same sense that I might get from a grad student who insisted on wearing a suit and tie every day -- i.e., very minor, and I would consistently remind myself that it is their preference, that it could be viewed in a positive light, and that it would be unfair to hold it against them. I hope my answer has not unduly spooked anyone; I very much doubt that there would be any negative consequences for any mathematician who chose not to ignore the guidelines. – Anonymous Jan 14 '15 at 21:11
40

Speaking as both an editor and reviewer, I am definitely prejudiced against a paper that fails to follow prescribed submission formatting (which may or may not relate to the final published format). It is simply a matter of professionalism and prior correlation.

As an editor and reviewer, you see a wide range of material submitted. Some is really good, and some is really bad. I've even gotten a few that were outright insane. The vast majority of the papers that failed to follow prescribed formatting were definitely not good.

Making a good-faith effort to follow formatting requirements generally isn't hard to do, and especially when doing so just means using the journal's LaTeX package rather than the default. Neglecting it means that the author is being sloppy and unprofessional at something easy. This doesn't necessarily impugn their science, but if they don't care enough to follow professional standards on something easy, it's a good indicator that they are likely to be unprofessional in other places where it matters more as well.

One exception: I am likely to give a pass to particularly aged/emeritus types who have a solid track record but are clearly not comfortable with modern word processing technology.

jakebeal
  • 187,714
  • 41
  • 655
  • 920
  • 42
    The impression I get is that people often feel the formatting requirements are unnecessarily hard to follow, and take up far too much time to do before you know the paper is going to be accepted. Given that basic Latex is very readable (and sometimes more correct than the journal format), why waste everyone's time? – Jessica B Jan 12 '15 at 17:10
  • 11
    @JessicaB In my experience, if you are using LaTeX and BibTex, that is simply not true. For pretty much every journal that cares about submission format, either the journal provides its own LaTeX package or somebody else has created one for it. Switching from basic latex to use such a package takes only a few minutes, and the package then handles all of the nit-picky formatting requirements (the macro-requirements, like section structure and manuscript length, you should be complying with in any case). – jakebeal Jan 12 '15 at 17:39
  • 53
    "It takes only a few minutes": In theory maybe, but in practice it is often more involved than that. You may find that the journal's package clashes with some other package you are using, or their margins cause your equations to overflow lines, or any number of other issues. It certainly can be a nontrivial amount of work to convert a paper into the journal's style, and I agree with @JessicaB that it seems like potentially wasted effort since the paper may be rejected. – Nate Eldredge Jan 12 '15 at 19:13
  • 3
    @NateEldredge some journals (American Institute of Physics ones for example) recompile the submission using revTeX - which has clashes with some packages. The recompiled version is what the reviewers see. In that case there won't be anything to review if you don't use the journal's packages, so if you're targetting those journals you have to do it their way. Equation wrapping can be dealt with later if necessary though. Just don't get me started on bibTeX support or lack of it (never mind bibLaTeX). On the plus side if you can compile it using revTeX it's easy to compile it without. – Chris H Jan 12 '15 at 19:51
  • @NateEldredge Different experiences, I guess --- my own experience has been that such conversions are fairly painless, but perhaps that is because I do not tend to use particularly fancy or unusual LaTeX packages. – jakebeal Jan 12 '15 at 20:08
  • 5
    @NateEldredge: "You may find that the journal's package clashes with some other package you are using" - that is even one more reason to use the journal's package right from the start, because like that, you won't spend any time and energy on meticulously designing some special element of your paper using commands from a package which, in the end, you realize you cannot use because of clashes with the journal class. – O. R. Mapper Jan 12 '15 at 20:56
  • 15
    @O.R.Mapper: As I just commented on Bill Barth's answer below, which is "the journal"? I don't think most people know "right from the start" which journal they are going to submit to. And even if you do, it could be rejected from that journal, in which case you are back in the situation of adapting your paper to work with the next journal's package. – Nate Eldredge Jan 12 '15 at 20:59
  • 2
    @NateEldredge: Indeed, academic environments may well differ there. I'm from one of the almost-only-conference fields of CS, so I admit my experiences are probably quite different from OP's who is asking about journals (myself having been "academically brought up" to never start writing before I know exactly what CfP + deadline I am aiming for, and with the knowledge in mind that a resubmission somewhere else means a 50% rewrite, anyway, as a new format will most likely give me some more pages of space, or take the same amount of space away, automatically requiring extensive restructuring). – O. R. Mapper Jan 12 '15 at 21:03
  • 7
    I agree with Nate Eldredge. I have submitted over 30 papers, never using journal-specific format, and I have never had any complaints. However, after the paper gets accepted I often have to mess with the journal's style files, which often takes an hour or more (and sometimes correspondence with the journal). Why waste my time on this before acceptance? I also agree with @Anonymous's answer: as a reviewer, I find it a bit presumptuous when a submitted paper looks like it is already ready to be uploaded to the journal's webpage. Most papers require more than cosmetic revisions anyway... – Pete L. Clark Jan 13 '15 at 03:42
  • 21
    I am tempted to downvote this answer. I think it is ridiculous (and unprofessional) to be prejudiced against a well-formatted, readable paper just because it does not follow the journal's formatting requirements (unless the journal insists on that in the submission guidelines, which I have never seen for initial submission). And I agree with the comments above: reformatting a paper has never taken me just a few minutes. – Sasho Nikolov Jan 13 '15 at 05:50
  • 3
    I've never found reformatting a paper a burdensome task. It doesn't take a few minutes, but in less than an hour can be done, and given that reformatting is a rare event, I think it can be considered acceptable. What I really find annoying is that several publishers do not maintain an updated LaTeX system, forcing the authors to employ obsolete packages. – Massimo Ortolano Jan 13 '15 at 08:48
  • 4
    @NateEldredge: Addendum, as I don't like to leave my claims unsourced: I am used to conferences with statements such as "Submission formatting and templates: All papers must be formatted according to ACM proceedings style.", "Files submitted (...) that do not use the submission template will not be considered for review." or "All papers must conform, at time of submission, to the IEEE Formatting Guidelines.". – O. R. Mapper Jan 13 '15 at 10:23
  • 1
    @O.R.Mapper conferences are a different game altogether. A major reason many of them insist on a formatted initial submission is that they have strict page limits too. But the question is about journals. – Sasho Nikolov Jan 13 '15 at 16:37
  • 2
    @SashoNikolov Many journals have strict page limits as well... – jakebeal Jan 13 '15 at 16:40
  • 2
    I cannot think of a journal in which I have published that had submission guidelines. Most of the journals have LaTeX classes, but all that I ever submit is a PDF I have created - I submit the LaTeX only if the paper is accepted. I believe the journals would accept a scanned typewritten document, or even a mailed typewritten document in some cases! – Oswald Veblen Jan 13 '15 at 18:11
  • 1
    @OswaldVeblen Here is an example of a journal with strict submission guidelines. This is quite common in IEEE, ACM, and other computer science / engineering journals. – jakebeal Jan 13 '15 at 18:21
  • 3
    @jakebeal: fortunately for me, I don't have to deal with those journals (in mathematics, we seem to always do these things after acceptance). – Oswald Veblen Jan 13 '15 at 18:22
  • 3
    @OswaldVeblen There is clearly a strong divide between field cultures here... – jakebeal Jan 13 '15 at 18:23
  • 4
    Certainly. In the format of this site, the only way I seem to have to point out when an answer is very different from my experience is via comments. I think it would be possible to clarify the answer by specifying which field(s) you are referring to. @jakebeal – Oswald Veblen Jan 13 '15 at 18:26
  • 1
    While the answer makes sense for mathematics, physics and the like, it excludes the majority of academic papers and manuscripts by focusing exclusively on LaTeX. I think most of the papers are written in Word, as only a minority of all academic fields use LaTeX. In Word, formatting can take much longer time to change. Half a day is not unusual in my experience... – Richard Hardy Mar 30 '21 at 09:47
  • @RichardHardy I believe this still holds for other fields as well; my experience with biology submissions (which are often in Word) is that the submission guidelines tend to be very loose regarding typesetting. If they do ask for something, though, they still generally care, e.g., if they ask for double-spaced lines and you decide to ignore it and single-space, it may not even make it to the editor before the pre-review QC checkers throw it back to you. – jakebeal Mar 30 '21 at 10:41
  • @jakebeal, I did not formulate my comment carefully enough. I only meant that the part regarding how much work it is to tailor a paper to the journal's format may not apply all that widely. The rest of the answer I have no problem with. – Richard Hardy Mar 30 '21 at 11:09
19

Beware of what the journal styles put into your paper. For instance, Elsevier's style file elsart (recommended for instance by this journal in my field) contains a footer with the words "preprint submitted to Elsevier". If I haven't submitted it yet, I don't want to write a false statement on an e-print on arXiv. The recommended style file for this other journal in the same field inserts the text "Copyright © 2010 John Wiley & Sons, Ltd." on the manuscript, which is outright false and borderline criminal in my view.

So you'd better not use documents created with these classes for anything beyond journal submissions (e.g., preprints, which at least in maths are basically a necessary step, or sending manuscripts to a colleague). This means that if you use them you need to prepare at least a second version.

My experience is that changing LaTeX format can be time-consuming, due to various package incompatibilities. So I have started submitting papers using the style which I am already using for the preprint (and for my internal notes --- I typically start to write down a manuscript much before deciding to which journal I am sending it). No one has ever complained (editor, referees...). More recently, Elsevier explicitly authorized this practice on many journals by launching an initiative called Your paper Your way.

So my suggestion is just forget about journal styles unless someone insists on them.

Federico Poloni
  • 46,039
  • 18
  • 129
  • 194
  • 4
    What a great answer. The "Your paper Your way" thing is nice: the evil giant really does throw us a bone every now and then, it seems. – Pete L. Clark Jan 17 '15 at 18:12
  • 8
    Unfortunately, I have an update to "no one has ever complained". A journal has recently written back to me that I had made the terrible mistake of putting the references before the appendix (instead of after), that they couldn't send a manuscript in this state to the referees and that I should change that. All this one month after the submission. And their guide for authors did not mention appendices at all. Just to name and shame, the journal is APNUM (Elsevier). :( – Federico Poloni Aug 21 '15 at 07:42
13

What reason might there be for not using the journal's format? Here is the one I have heard of.

If your papers are always accepted by the first journal you send them to, then you might as well format them for the journal. (But it probably means you are aiming too low!)

On the other hand: if you sometimes have a paper rejected by one journal, then send to another (and another, and another...), why should you have to change the formatting for each one? In cases like this, the author would prefer to do the formatting once, for the accepting journal. Sensible journals would allow this.

GEdgar
  • 18,325
  • 8
  • 47
  • 66
12

Not surprisingly, the best way is to follow the instructions. But, if you for some reason do not then keeping a manuscript very simple is the best second approach. Simple, generic, typography, 1.5-2 line spacing, figures and tables separate from the text works in most cases. As an editor and reviewer I find evidence of special formatting most annoying. It distracts the reading and is also completely unnecessary since the journal will likely reformat the article during type-setting.

There are some aspects where you should try to put some effort in: try to make sure you adhere to the journal's standard fro referencing and make sure your reference format is correct. It is really annoying when references are not complete or the reference list is haphazardly put together. The key is in the details so make sure you get the details correct. You do not want the editor or reviewers to get annoyed by inconsistencies in details, make sure they can read a well prepared manuscript with little effort and disturbing elements.

So, if you cannot follow the exact instructions, just keep it very simple and avoid inconsistencies.

Peter Jansson
  • 73,217
  • 12
  • 206
  • 344
  • 7
    I am aware of the habit of separating table and figures from the main text in some fields, but doesn't this make the paper harder to read? Using any docuent preparation tool that will cleanly position the figures and tables and wrap the text around would be best, wouldn't it? – Benoît Kloeckner Jan 13 '15 at 14:51
10

Sadly, many journal submission guidelines still request an archaic format in which the figures are all placed at the end of the manuscript. Often the figure captions are themselves separated from the figures. This may have made sense in the days of hard copy submissions, but is pointless -- especially at the review stage -- given current technology.

As a reviewer I find this format intensely annoying and it is possible that my review quality suffers as well; I sometimes read through several figure references before flipping back to look at several figures at once. My personal opinion is that authors' highest priority should be to submit their work in a format that minimizes effort on the part of the referees, and if this conflicts with house style requirements, the author has every reason to ignore those requirements.

This is not to promise that every journal will tolerate such a decision on the authors' part, but I have never seen a harsher consequence than a relatively polite request to reformat.

Corvus
  • 18,086
  • 6
  • 63
  • 82
8

I think there's a balance to be struck here. Some reviewers will be annoyed if there's no room on a printed version to scribble their notes about the paper. This means that typical IEEE/ACM two-column, single-spaced formatting might annoy some people. The argument from them pretty weak if your formatting conforms to the journal style rather than being some tight format that you made up.

On the other hand, using the journal style generally gives you a reasonably readable presentation designed, more or less, that way. It was probably also designed in the age of print to get the most words on the page, so it's not perfect.

The worst thing you can do from a reviewer's perspective is to come up with your own formatting that looks slapdash, unprofessional, sloppy, or is hard to read. Almost no matter what you do, someone will be grumpy, but if you use the journal template, you'll engender the least complaints.

Bill Barth
  • 48,733
  • 6
  • 112
  • 194
  • Thanks, I missed some necessary information. If possible, please be referred to the edited question :) – Yes Jan 12 '15 at 15:23
  • 3
    If you're doing it LaTeX already, just use the journal's template from the beginning. – Bill Barth Jan 12 '15 at 15:25
  • 13
    That assumes that you know from the beginning which journal you are going to submit to. In many cases that decision isn't made until the paper is already finished. (Or, maybe you did know which journal you would submit to, but it was rejected from that one, and now you have to convert it to use a new journal's format...) – Nate Eldredge Jan 12 '15 at 20:57
  • 3
    @NateEldredge. Yes, that's true, though typically applying a new template isn't that hard. I still recommend that you do the work to put your article in a journal's style before you submit. This is especially true if there are upfront page limits. If your field is like mine, almost everyone uses an IEEE-like or ACM-like two-column format even if they aren't them. – Bill Barth Jan 12 '15 at 21:21
  • 3
    @Bill Barth: as I wrote somewhere else here, I would view it as presumptuous to submit a paper already in the journal's format. I just submit a PDF made in plain LaTeX. If there are tight page limits, or other special circumstances, that is different. But I am more familiar with journal submissions where the page count is very flexible. – Oswald Veblen Jan 13 '15 at 18:14
  • 3
    @OswaldVeblen, why? So many journals and conferences want submission in their format upfront, it seems very punitive to associate proactive formatting with presumption on the author's part. – Bill Barth Jan 13 '15 at 19:06
  • 2
    @Bill Barth: it turns out that is a very discipline specific thing. In math, I cannot remember any journal I have submitted to that requires a special format before submission - see the comments below jakebeal's answer. (And, if I use the journal's format from the beginning, what do I post on the arXiv? A format that names a journal that may reject my paper?) – Oswald Veblen Jan 13 '15 at 19:09
  • 2
    @OswaldVeblen, it's clear from this thread that requirements are very discipline-specific. Given that, I would hope that those who have a bias against pre-formatted articles would learn to be a little more open-minded. As to arXiv, you can do what you like within the journal's guidelines, but you might just eliminate the journal name from the template. IEEE and ACM templates, as well as Wiley and others, are generally generic to the publisher and the specifics are filled in as a variable in the template. – Bill Barth Jan 13 '15 at 19:41
5

If you are writing your paper in LaTeX, then you should use the LaTeX template provided by the journal (I have yet to submit to a journal that does not provide a LaTeX template). This will take you a few extra minutes.

The LaTeX output is what both the editor and the reviewers will see. Why would you not want to ensure this looks professional? Are you submitting to so many conferences you do not have an extra 20 or 30 minutes to polish your submission?

I don't check for adherence to journal standards during reviews, but I've gotten several papers where the author(s) didn't see what the LaTeX file would look like in PDF form. These papers almost always have other issues, and reading through a poorly formatted PDF discourages me from giving the paper the extra attention it might need.

jakebeal
  • 187,714
  • 41
  • 655
  • 920
sevensevens
  • 7,461
  • 20
  • 39
  • This seems an unrelated problem. If someone is bad at LaTeX document preparation, they will create a poor document no matter what the style is. If anything, hastily converting to the journal style file before submission can only make things worse. – Federico Poloni Jan 16 '15 at 16:59
  • 2
    As you've likely spent months researching for this paper, I don't understand why you would not take an extra 30 minutes to ensure your submission will look professional – sevensevens Jan 17 '15 at 16:14
  • 3
    I don't understand why a journal would take an extra 30 minutes to impose pointless, nonuniform, and occasionally self-contradictory formatting requirements on submissions pre-review. – Paul Aug 21 '15 at 00:26