On Dec 9, 12:43am, Marek Rouchal wrote:
> Subject: latex2html: Re: New developer's site URL!
> On Tue, 8 Dec 1998, Bruce R Miller wrote:
> > So, how does one become a developer of l2h?
[...]
Thanks for your vote of confidence; I wanted to reflect a bit before replying.
> Now for the organisational part.
(The tricky part!)
> [...] Let me start with the cons ;-)
>
> * you could possibly destroy other people's work
> * the repository could become a mess if too many people are working there
> * you can get the latest developments via the Web, too
> * you can post your contribution on this mailing list and one of the
> developers includes it for you
We want to avoid conflicts in code & API's, but we also want to avoid conflict
in style & philosophy. I would very much like to contribute to the general
tool, but I have a bias towards getting a more flexible platform for
translation to XML, MathML (and, ideally, OpenMath). Also, IMHO (maybe not as
humble as it should be), to get there, and even just to keep l2h maintainable,
it would be good to see more modularization and cleanup in the code, and Perl5
modernizations; which I'm willing to do, but...
A lot of the changes I can envision wanting to make would be too obnoxious to
pass thru any of the overworked official developers. And yet, some are also
likely to be too extensive to make without some sort of reality check by the
current folks. Such changes might not `destroy other people's work' in the
literal sense, but may destroy the design & spirit of it.
Another problem area is testing: since l2h maps so many different inputs to
so many different outputs, and since (as I've seen) seemingly minor changes
can have rather drastic side effects -- and many effects you only see `by
eye'-- having a good test suite is very important! How do you do it Ross?
So, I think I have some good things to offer, and would like to contribute, but
I would like to see how I fit in, and how I should run ideas past people before
committing a change (and how to thoroughly test them afterwards :>)
--
--
[EMAIL PROTECTED]
http://math.nist.gov/~BMiller/
