Re: Windows 21.4.5 Ready for release.

William> Thanks Andy, Three questions: The main reason for version skew within the XEmacs package system (such as that you picked up for PSGML) is that we initialize the "available" database with a fairly recent issue of package-get-base. However, you need to actually update the package list from Tools | Packages | Update Package Index before the version numbers will be accurate. (We don't worry about it too much because the purpose is to initialize the list of packages, the version numbers just come along for the ride so to speak. We do try to make sure the package-get-base is pretty fresh at release time, but The Package Czar and I are not very coordinated right now for reasons beyond our control.) Andy, are you picking up test versions of some of the packages? I don't think so, but that is a theoretical possibility, too. William> 1. Is there a mapping table somewhere? [Somewhat OT: William> "WHY are there different numbering systems?"] APA> For XEmacs Release numbering, see APA> http://www.xemacs.org/Releases/Public-21.2/numbering.html This is irrelevant to William's question. APA> Package numbering has to deviate from external package APA> releases since it counts XEmacs package releases, not APA> external releases. We may fix bugs locally, add features APA> before external package is released, etc. This is true as far as it goes. Still, what we should do is use the .rpm/.deb convention of having an upstream version number and an XEmacs package release. However, many Emacs people are anti-version number in principle (version numbers hide the feature set, which is what you really want to condition on) and many packages simply don't _have_ version numbers. In other cases packages often contain a number of separately maintained files each with their own version number. What should we use for the XEmacs package version? Nobody has volunteered to work it out, and the current system works well enough. These considerations, plus the fact that many packages are externally maintained, also make maintaining a mapping table a bit difficult. It's one of those many good ideas we don't have a volunteer for.... In cases where the upstream maintainer and the XEmacs maintainer are the same, the version numbers typically are the same (eg, VM). We're trying to bring Paul on board for JDE, but he's waiting for some changes that will reduce the burden on him. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Don't ask how you can "do" free software business; ask what your business can "do for" free software.