On Aug 2, 4:58 pm, "Dr. David Kirkby" <david.kir...@onetel.net> wrote: > .. if the difference between experimental > and optional could be clarified.
Personally. if I could define it, I would use these definitions: 1. standard: that's well tested, included in each distribution, essential, and the responsibility of the Sage project that they work. 2. optional: that's some addition to sage that is supported by the sage project (more or less), but if it builds on all systems or not is not an issue. It should follow some defined standards, everybody who wants to improve one of it has to go through trac+review to do so and there should be a matrix on which systems one of these packages is expected to work. If requested, some of them may be promoted to "standard", as it already happened more than once. 3. experimental (which I would rename to "contributed"): this is a set of packages a user has created, it is not supported by the sage project, but only by one or more maintainers who are the only ones who are responsible for it. they might not work on all systems, they might not work at all, but most importantly, they bypass the trac+review process. The reason why I wish there is this 3rd category is that for example the openopt package that i've created, it is up for review (#7708) for more than 6 months. That's longer than the release cycle. If it won't work somewhere, I would try to fix it and take the responsibility if it is wrong or faulty. I don't see the benefit of still providing a really outdated version of it (officially) while not pushing this updated version. H -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
