sTeX integration in PlanetMath

[holtzermann17]

It would be great to have sTeX as an alternate input mode in PlanetMath, and eventually work towards semanticizing the PM corpus as a whole.

It would probably be easiest to do this if the PlanetMath articles were forkable as in #351, so we can explicitly create the sTeX "version". This also relates to #49 ("standard tagging and workflow") which we can use to designate the "main" version of an article.

[rspuzio]

Also, as we discussed the other weekend, once we have sTeX supported,
a way of migrating to it would be to present people editing entries with
a dialogue box which would point out semantically ambiguous TeX and
suggest some alternatives to pick from (with "none of the above" as one
possibility).

[holtzermann17]

👍

[dginev]

In my current mindset, this ticket is blocked by sTeX, rather than Planetary. A platform-friendly sTeX would be able to "gracefully degrade" down to regular LaTeX markup, and i.e. be transparent for the user.

So, in the best of all worlds, sTeX would simply be an additional package to preload in LaTeXML and would continue to return the same HTML+RDFa markup. I know that currently at KWARC there is an effort to rebase the last stage of the sTeX pipeline (OMDoc -> HTML) to pass through MMT instead of JOMDoc (which Planetary shouldn't notice in the first place).

In practice, that would mean that Planetary has an MMT server running in the background, similarly to the Virtuoso backend (and what used to be the TNTBase backend).

What I am dreaming about in the direct sTeX -> HTML+RDFa conversion however will be rather high effort, so I am just recording my thoughts here.