It’s nice to hear Hermes still has a relevance here and there, however, I didn’t touch it since Nov. 2006 and I don’t have any plans to get involved in tuning it further until some long-term institutional interest shows up and some personal conditions are met: I don’t want to get involved in short term projects anymore until after I get a (fitting) job in Spain and settle there.

There is no longer any active mailing list related to Hermes; when it was, all of the subscribers were passively waiting for the new release anyway.

Hermes was practically a one person project and, in my opinion, it is far from being completed to the level and coverage area I intended but funding issues and the way public research is currently organized (by forcing researchers either to jump from a short-term project to another, or to become fund raisers, aka beggars and/or liars, and by ignoring the society’s long-term priorities) put, practically, a stop to its evolution.

The ones who got interested in using it were those I listed on the Hermes site and on the Hermes @ AEI site. Perhaps lxir is the most active currently but I’m not sure. I did not maintain any link with them after 2006.

Ah, there was at one time a project called sage which showed a spotty interest in Hermes but I didn’t hear from them since 2007.

That said, Hermes is under GPL, so feel free to do whatever you wish with it, it’s just that I won’t be there to patch it for an indeterminate time, after all, one needs to know a bit of C, a bit of TeX, and to have a clear goal, to do the right thing with it.

I wish you all the success you can handle and have a happy new year, …

P.S. I just stumbled upon the arXMLiv project, which uses a different tool, LaTeXML, for converting TeX to XML+MathML, you may find this also relevant enough to your work.

Because those sources vary widely in terms of metadata quality (author, date…etc. records), some of the converted articles look a bit messy. That’s the next Hermes version’s job: to get a bit wiser in terms of recovering the metadata, and in terms of robustly recovering the internal document references/citations.

Many of the original articles are missing from the converted lot: a few of them were using drawing packages, and Hermes doesn’t support drawing yet, a few of them weren’t converted because of an unexperienced script which filters the original lot and doesn’t do an intelligent renaming of files, and about 5 of them weren’t converted because Hermes needs some (small) fixes.

The total number of articles tried is 1005 (751 articles of the month 2005/3, and 254 of 2005/4); the success ratio today is 0.438.

The good news is that this version tries to push/pop parentheses at the table cell border, or around the align marker in (ams)math structures, so that the presentation-MathML result is well-formed with less need of human intervention.

The source distribution is here, and the samples site is here, as usual.

Enjoy

The articles drawing diagrams or pictures (e.g. using the xy package or similar) were not convertible by this version of Hermes, and I don’t plan on adding support for them anytime soon.

The rest is fixable and I’m working on it. Letting me know the specific issues you may have in using Hermes for converting LaTeX to XML+MathML+Unicode increases the priority of dealing with them.

As soon as the converted articles will be posted online somewhere on the web, I’ll announce here the address of the collection.

Sources/binaries are here and examples here.

Enjoy.

begin{longtable}
caption{…}
label{…} \
the & rest \
of & it\
end{longtable}

This version managed to convert 32 articles out of 48 trials (in the order of assigned publishing number) from arxiv:math.
A 66% success rate is not impressive, but Hermes gets wiser.

Sources/binaries are here and examples here.

Let me know if Hermes annoys you in some way, maybe there’s a worthy fix to it.

As you probably know already, Hermes currently generates a mtable from a LaTeX equation long enough for TeX to split it on more than one line.
But now Hermes would like to do the following, beginning with the next version:
1. generate a single mathml row instead of a mtable for this kind of situations
2. generate a mtable only where source equations are labeled and longer than one line.

if you have strong feelings about this decision let me know a.s.a.p., Hermes will consider your opinion.

Examples of LaTeX articles converted with this version are available here.
Enjoy.

I’ll be slightly unresponsive this month.

Examples of articles converted with the current version of Hermes are at the usual place.
Enjoy.

Next version should have more fonts, should come with a Hermes reference document schema, and should show more examples like those listed here.

The source distribution and the windows binaries are available for download from the usual place.

I had to fix some of the most nagging bugs which appeared in 0.9.5 or persisted from the old versions, they really got on my nerves this time .

My plan for the next one is to rearrange the font mappings to make them, or their patching, less error prone.

Your requests/comments/questions are useful in tuning the next release.

The converted and rendered articles are here.

enjoy.

It is a step closer to a Hermes document schema.

If you have requests, comments, questions, suggestions, please use this place, so other readers can use the info here to get a more detailed view of what Hermes is about.

A more radical version of Hermes is in the works, 0.9.4 was the last one developed within the ‘deadline’ mindset.

My plan is to have a Hermes which generates a clean and validable document model, complete with the metadata necessary to store this document in the institutional repositories (digital libraries).

Enjoy.