# Advantages to article submission in LaTeX from the publisher's side

Many journals in my field accept both Word and LaTeX formats for article submission. From the publisher's point of view, what are the advantages of using the LaTeX format for submission?

Do publishers eventually use LaTeX to typeset the final version of the article? In that case, I can imagine that having the references ready in BibTeX and equations in TeX (among other things) can save quite some time in the typesetting process, and be less error prone. Are there any other advantages?

If a publisher accepts LaTeX manuscripts they likely use LaTeX also for the final type-setting. I am sure there are exceptions but could not point you to one.

The benefits, apart from obvious LaTeX benefits such as equations and built in standards for references, numbering of equations, figures and tables etc., lie in the way a manuscript can easily be taken from a manuscript form to a finished product.

Many journals have class files that allow you to move from manuscript to essentially "proof" mode by changing a switch in the document and "re-compiling" it. This also means that the journal can go to typesetting without moving file contents to a new format or another program (not many journals are type-set in Word).

In the case of journals that do not have class files for use by the author, moving a manuscript from a generic LaTeX format into a specified journal format is not necessarily difficult. There are probably many different approaches to this but from a LaTeX point of view all definitions of a document are there in the plain LaTeX file and it would be easy to apply a class that re-defines the plain format to something that will yield a finished layout.

With LaTeX focus is on writing the text, not formatting the manuscript. As Editor-in-Chief for a journal that uses both Word and LaTeX, I have spent many hours weeping over hopeless Word formatting (including field codes that do not work) that is both unnecessary and complicating moving the document to the type-setter. LaTeX is a text file and so does not contain anything that cannot be easily spotted and changed if need be.

So, to be fair, one can mess up with LaTeX as well, and I want to point at an overarching rule which is to always strictly adhere to any instructions for authors provided by the journal and not to send in material that is of a format that differs from what is asked for.

So the benefits of LaTeX is that the move from manuscript form to finished layout is simplified and reduces the amount of manual work for the type-setter. But, in all type-setting there is always need for manual control so LaTeX is not 100% automatic, just closer to it.

About BibTeX referencing: Most journals and I would guess type-setters want the manuscript in as few parts as possible. therefore many provide a .bst for the reference style but ask that you run BibTeX to produce the .bbl file (containing all references properly formatted with \bibitem formatting) and then paste the content into the document to provide a complete and correctly formatted reference list inside the document file itself.

• The APS, and likely the AIP, use SGML in there backend, and they convert both LaTeX and Word documents to it. (I can't find their links to the process at the moment.) I suspect the LaTeX converter was easier to write. :) – rcollyer Dec 16 '14 at 14:14
• to be fair, one can mess up with LaTeX as well +1. We all have our little LaTex tricks, and there is this lone \end{document} that was forgotten at the bottom of an included file, etc. For anything where equations are rare, a simple unformated text, with all dynamic fields disabled, can be much, much easier to deal with. – Cape Code Dec 17 '14 at 2:58
• Note that this can be a huge deal for open-access journals. One example is the Journal of Machine Learning Research. It's one of the top ML journals and (as of 2012) has run itself for 12 years on a total budget of less than \$5k (source). It has no submission fees and, really, no sources of income. It can do this largely because LaTeX allows authors to do most the type-setting in an easy and consistent manner. – Roger Fan Jan 29 '15 at 17:51

The biggest benefit: Their typesetter will love you.

Now seriously: There are three possible cases:

1. They use LaTeX for everything. Then posting a LaTeX article means: less work for them (I mean, much much much less work for them), less errors introduced, clearer proofs, typesetter more happy etc.

2. They use what you use. Then it depends on how well the LaTeX template is done. I've seen journals (mostly engineering and chemistry) that have a LaTeX template just "because people were bugging us to have one". Then choose what you prefer and what you think is easier. Or, write to them and ask what they prefer.

3. They send everything to somewhere, and all articles are completely re-done there at low costs and high quality. Yes, this seems to be the case for some journals[citation needed] and as before, it doesn't matter what you do use.

So, FWIW, LaTeX is never a mistake, nor is asking what they prefer.

As for BibTeX, it's a bit more complicated than what Peter presents. The rule is: Follow the guidelines (you have read them, right?). If BibTeX is not mentioned in the guidelines, suppose it's not supported, choose your favourite style, and either include the .bbl file or simply copy the contents of the .bbl file in place of the instruction \bibliography{mybibfile}.

• I have never really understood the idea behind "we do not support bibtex". If I were a typesetter, I'd love to have the bibtex source rather than retyping in all the references by hand. – Federico Poloni Dec 16 '14 at 17:24
• (+1 for "some jounrals retype everything by tehmselves", by the way) – Federico Poloni Dec 16 '14 at 17:25
• @FedericoPoloni The problems with bibtex are: Corrupted .bib files, authors using some tweaks that work only for this or that .bst file etc. Unless you really want to be uniform (like AMS, which strongly enforces all bibitems to contain the appropriate MR tags), you just mention your preferred style and let it be. – yo' Dec 16 '14 at 21:02

Latex is always good, but I've gotten skeptical of going out of my way to format my articles of the journals and provide source files. I've found that some journals will rekey the whole manuscript, no matter what you send them. I've picked up on that by discovering typos that were not in the originally submitted files. They also often have their own particular bibliography formatting style.

• Is this publisher or journal dependent? – Gimelist Dec 16 '14 at 20:43

I just had an article accepted to a earth science Springer journal. I asked the typesetter whether they prefer LaTeX or Word. His reply was:

It is better for us to submit the manuscripts as latex or word document to avoid font missing problem.

I'm not sure what you can make of it.

Also, it seems that the proofs were reformatted into something which is not LaTeX or LaTeX-based at all. This is weird, given that I submitted the article using Springer's LaTeX template. This could be because in this field, people hardly use LaTeX.

From the publisher's point of view, what are the advantages of using the LaTeX format for submission?

Because if they don't, their competitors will, and authors who prefer to write their papers in LaTeX will choose to submit elsewhere. See the comments here.