- From: Olivier Thereaux <ot@w3.org>
- Date: Tue, 29 Oct 2002 09:20:15 +0900
- To: QA Dev <public-qa-dev@w3.org>
A quick answer, I can investigate further if needed. First, I think providing packaged versions of our software is a good thing. It does not mean we have to spend a lot of work packaging the validator for each architecture and each packaging system. What we can do and this seems to be the current practice at W3C: - we provide the tarballs as the "official" released package, - depending on availability in-house, we can also provide a few packages. We used to have a redhat fan at W3C (Daniel Veillard, rpmfind's father), so for a long time we had rpm packages (etc.). - we welcome packages created by external contributors, and we either link to them or host them, with proper credit and warning. On Monday, Oct 28, 2002, at 16:42 Asia/Tokyo, Terje Bless wrote: > Initially, my thought was that we > would only support Red Hat (8.0) with the RPMs (the tarball is of > course > more or less platform independant). But that might be construed as the > W3C > endorsing a particular vendor... I understand some people may think so, but I wouldn't pay it much attention : most people would be happy to have rpm, and enemies of rpms cannot complain very long if we tell them we would be glad if they create their own package. > My feeling is that we're supporting Red Hat with a binary installation > only > because that is what we are able to support -- implying among other > things > that if a Debian or *BSD packager should volunteer they'd be gratefully > accepted, of course -- and that this will not be a problem in practice. Exactly. -- Olivier
Received on Monday, 28 October 2002 19:20:18 UTC