bb wrote:
I think it would be a very useful exercise for hh to install a complete texlive on a stock fedora box.
I'm sure he would be happy to, if you provided the box and some funding (!). Most developers that work on free software miss at least one of the following: * spare hardware * spare disk space * spare operating systems * spare time
Then write down _every_ step necessary to get beta cont-tmf.zip installed and working.
I guess I could conceivably try to do this, as long as I only have to do it once, but I will not be able to provide support for it afterwards. I have a fairly complex TeX setup, and for that reason I do not use texlive. I can try to write down this howto for *current* context beta and *current* pdfetex, but afterwards I have to revert my system back to 'normal operation' immediately, so there is no way for me to keep on helping people after the initial installation. Assuming someone volunteers to pick up supporting context-on-texlive afterwards, then I am willing to write down the initial page. But I do not want to waste a day fighting TeXlive for no good, so unless I hear from someone, I will do nothing.
I would think that targeting texlive as the standard platform for distributing context on linux/unix would make the most sense. It should _not_ be necessary to make anything but a trivial change to texmf.cnf.
Have you complained to the TeXlive maintainers as well? After all, they are the distributors. Perhaps they could make it just a bit easier to add stuff to an existing TeXlive install? The upgrade path for any TeXlive basically is: download an update iso image, burn it, and re-install. There is no support for partial upgrades (at least there wasn't last time I looked, which is, admittedly, a few years ago) Taco