# breqn in journals

I read the breqn manual, and I love the ideas in it (semantic markup for equations, automagic line breaks in math displays).

I would like to use it in my work (I am a research mathematician, so I mean mostly producing journal papers). However, I am a little bit worried about compatibility issues. Does anyone have experience with it in a similar context? Namely, what I would like to know is:

1. do math journals support it out of the box, or do I risk getting a nice message from them on the lines of "could you please re-format each and every formula in your paper before publication?"
2. are there any outstanding incompatibilities with commonly-used packages?
3. has the API changed a lot recently? I.e., will my collaborators with a slightly outdated TeX version still be able to compile the documents?

I hope that the question is in-topic here.

-
What is really needed is a way to convert the breqn broken equations to more standard amsmath stuff such as align, just before sending the file to a journal. I need to think on how this could be done (breqn has been on my todo list for at least a year now). [likely future contributor to breqn here] –  Bruno Le Floch Jun 9 '12 at 10:04
@BrunoLeFloch -- while conversion of breqn broken equations to amsmath structures is a great idea (and i hope you do come up with some workable thoughts as to how it might be done at least semi-automatically), to the best of my knowledge breqn is not rejected by ams journals. (after all, its initial development was here.) copyeditors may mark some of its results for adjustment, but i suspect that some must get through, although i have no concrete evidence either way. breqn is definitely not on our "reject" list. not perfect, but useful. –  barbara beeton Jun 9 '12 at 13:37
@barbarabeeton thanks for the information. This raises another natural question though: which packages are on your "reject" list? Shall I ask this as a separate question? –  Federico Poloni Jun 9 '12 at 21:41
@barbarabeeton: Reasons I don't use breqn in my documents is that (1) I fear that different versions on my collaborator's installations change may the breaking of equation (I work with people still using eqnarray) (2) it occasionnally fails to find an appropriate line-break, and there is no way to force/suggest line-breaks (useful for copy-editors). Both problems could be solved by providing an option for breqn to write a file with info about the line-breaks, and an option to use those line-breaks (I know this is a difficult problem). –  Bruno Le Floch Jun 10 '12 at 9:16
@FedericoPoloni -- the list of "reject packages" is too long to give in full, but here are some categories: anything that changes page dimensions or alignment; anything that changes the main text font; fonts that come only in metafont form if a comparable outline font is available; fancyhdr (conflicts with ams journal specs); caption packages (incompatible with ams caption style); anything unknown at ctan (unless provided by the author, and it doesn't conflict with ams style); makeidx (in ams classes); comment (all "private" comments must be removed, or we remove them here); obsolete packages. –  barbara beeton Jun 10 '12 at 13:22

I believe that breqn is not perfect and if you want to achieve absolute beautiful mathematics, then you (a human being) should decide where an equation should and where it should not break.

1. I think, you would get a nice message from them on the lines of "could you please re-format each and every formula in your paper before publication?"
2. The only one I am aware of is that when \TeXXeTstate=1 (enabling bidirectional typesetting in etex-based engines), with breqn you get absolutely different results (because the internal data structures built by TeX--XeT differ from those built by TeX, the typesetting of a document by TeX--XeT may therefore differ from that performed by TeX). As an example:

\documentclass{article}
\usepackage{breqn}
\begin{document}
\begin{dmath}[label={sna74}]
\frac{1}{6} \left(\sigma(k,h,0) +\frac{3(h-1)}{h}\right)
+\frac{1}{6} \left(\sigma(h,k,0) +\frac{3(k-1)}{k}\right)
=\frac{1}{6} \left(\frac{h}{k} +\frac{k}{h} +\frac{1}{hk}\right)
+\frac{1}{2} -\frac{1}{2h} -\frac{1}{2k},
\end{dmath}
\end{document}


produces:

but

\documentclass{article}
\usepackage{breqn}
\TeXXeTstate=1
\begin{document}
\begin{dmath}[label={sna74}]
\frac{1}{6} \left(\sigma(k,h,0) +\frac{3(h-1)}{h}\right)
+\frac{1}{6} \left(\sigma(h,k,0) +\frac{3(k-1)}{k}\right)
=\frac{1}{6} \left(\frac{h}{k} +\frac{k}{h} +\frac{1}{hk}\right)
+\frac{1}{2} -\frac{1}{2h} -\frac{1}{2k},
\end{dmath}
\end{document}


produces:

3. I think your collaborators will be able to compile your documents with a slightly outdated breqn package as long as they use PDFTeX.

-