On Wed, Aug 3, 2011 at 7:13 AM, Simon King <simon.k...@uni-jena.de> wrote: > Hi all! > > On 1 Aug., 17:18, Jason Grout <jason-s...@creativetrax.com> wrote: >> On 8/1/11 7:19 AM, Simon King wrote: >> >> > sage-newbie, sage-solaris, sage-flame, sage-marketing and sage-edu >> > together made up for 133 posts in six months. Moving all of it to sage- >> > devel would mean an increment of about 4%. If part of it was moved to >> > sage-support instead, the increment would be even less. >> >> sage-flame is an important separate release valve for anything on sage >> lists. I certainly don't think it should be merged with anything else. >> >> +1 for disbanding sage-marketing and merging that with sage-devel. >> +1 >> for disbanding sage-newbie (encourage sage-support instead) and merging >> sage-solaris with sage-devel, and maybe even merging sage-windows with >> sage-devel. > > Then I ask directly: > > 1. Is it technically possible to merge two existing Google groups?
I don't know... I really doubt it though, since Google groups don't have a lot of features. > 2. Do we want less Google groups for Sage? I don't. > > To summarise what I've been arguing above: > > I. > New contributors are likely to post on the big lists. Having many > small lists can lead to one of the following situations: > > (a) Experienced people are busy on the small lists, thus, read sage- > support less frequently, thus, the big lists are less responsive. This is not my experience. If not for the small lists, some of these experienced people would just read *no lists*. If they make the time to read the big lists, they will just read all of the lists of interest to them anyways. For example, for several months late last year *I* did not read sage-devel, since I didn't have time (and there was unfriendly discussion about certain topics, so I didn't enjoy reading it). I did still read some specialized groups like sage-nt, uwsage, etc. Now it's the summer and I do have more time for email, so now I read all the groups. > (b) Experienced people ask to move the question to one of the smaller > lists, thus, the question remains unanswered for longer than > necessary. This assumes (a), which in my experience is not necessarily true. > Both is frustrating for novices. > > The situation four years ago was much different. Both sage-devel and > sage-support were very responsive, and that has been one major reason > for choosing Sage as a platform for my computational projects. I don't agree that sage-support and sage-devel have become overally noticeable less *responsive* than they were four years ago, at least since I've been watching them closely during the last month or two, though there is less traffic (as I've observed). I think there are more subscribers to both lists today then there were in 2007. > II. > Having a combined list for sage-nt, sage-algebra and sage-combinat- > devel would (according to figures from the past 6 months) result in a > list with little more than 7 posts per day. That should be small > enough so that all people interested in abstract nonsense could easily > follow - actually more easily than with three lists. In particular > when interested people are only subscribed to two of them. As a specialist, and having read both lists recently, I have concerns about merging sage-combinat-devel with sage-nt. The discussion on sage-combinat-devel is much, much different than sage-nt, and to some extent the lists are mutually incomprehensible. Though I skim it lately (I'm reading all sage lists this summer), I usually understand little on sage-combinat-devel, and I can't imagine sage-combinat-devel readers wanting to read a lot about subtle issues involving elliptic curve point counting, say. > III. > Moving sage-solaris, sage-flame, sage-marketing and sage-edu to sage- > devel would yield an increment of (in average) less than one post per > day on sage-devel. I believe nobody can reasonably say that such a > little increment is a "flood with tons of technical details". A single > additional post per day could easily be filtered manually (i.e., > ignored), if necessary. I'm concerned about moving sage-flame to sage-devel. Many posts to sage-flame are *offensive*. This is like moving a 1am adult channel to a prime time network. Sometimes people post one word messages to sage-flame such as "F___!", just to let off steam. My impression with sage-marketing is that often the discussion is something that doesn't feel right for having thousands of readers. (?) > IV. > sage-marketing and sage-edu concern topics which are perhaps not in > the centre of everybody's research interests. However, most are at a > university and could occasionally contribute to sage-edu, and I guess > that most people would at least have an opinion on marketing and could > occasionally contribute to sage-marketing (ok, that could be "painting > a bike shed"). Hence, why hiding these two topics from a larger > audience, putting them into a small list? Having sage-edu separate is valuable in a somewhat similar way that having both http://meta.stackoverflow.com/ and http://mathoverflow.net/ is valuable. Many discussions on http://mathoverflow.net/ are quickly closed to keep the topic about research-level mathematics... which really helps foster a highly impressive community of on-topic research mathematics discussions. Building a community like at http://mathoverflow.net would perhaps be very difficult to build if it mixed undergrad and research level mathematics. (Please don't misunderstand my analogy -- I am not claiming sage-devel is about research math.) > > V. > sage-windows discusses a topic that is said to be a major goal of > Sage. Hence, why not underlining its importance? Why not raising > alertness by exposing it to the "big audience"? Merging the sage-windows list with sage-devel is not the only way to raise alertness. Another approach would be for somebody to post periodic updates to sage-devel about windows porting work, along with a note that says: "get involved by: * subscribing to sage-windows; * downloading and building sage following the directions at http://xxx, etc." Also, there are quite a lot of people who are highly involved with Sage development work, and who read sage-devel, but have (unfortunately) zero interest in porting Sage to Windows. They use OS X or Linux, and don't want to think about the world outside that. But I'm not in favor of barraging them with messages about porting Sage to Windows (or Solaris or anything else), since I greatly value their contributions to the platform-independent parts of Sage. -- William -- William Stein Professor of Mathematics University of Washington http://wstein.org -- 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
