+1 to a request for Jakarta-Math-Discuss list. I'd like to be Apache-Math-Discuss, but that might cause issues, cf the Python mail list request a while back.
Sorry, but I am still -0 to splitting discussion or development of [math] from j-c.
Why I like this: It is not tied to the commons-math code, but to the community that the commons-math code wants to join. I expect [math] to be using it for math discussion with commons-dev for Java related dev issues (such as maven, site etc) and releases. Average Java joes would still come to commons-user for [math] questions, so commons-math would have to be covering that too.
I said +1 in that I'm prepared to go onto infrastructure and ask for this etc, get the PMC to ask for it or whatever.
Do we have jelly-user-discuss, digester-user-discuss, collections-user-discuss etc.? I agree with Martin that j-c is one community and I see no compelling reason to split [math] off and some good reasons not to. As a user of multiple commons components, I monitor both commons-dev and commons-user. As a commons committer, I try to help out on both of these as well. There are *lots* of benefits to having the additional eyeballs in the j-c community that I, at least, am not willing to give up. I also strongly disagree with the implication that we can separate discussion of "Java-related dev issues" from math-related ones. Commons-math is a Java math library. Splitting the discussion in this way is like saying that [digester] or [jelly] should discuss XML-related issues on "digester-user-discuss" or "jelly-user-discuss" or that Stephen's recent post about Tree structures in [collections] should be on "collections-user-discuss." This is not good for the community, IMHO. It also threatens to move us away from the code-centric model that makes Apache work (again, IMHO).
Phil
Hen
On Sat, 14 Aug 2004, Stephen Colebourne wrote:
I argued for Apache Math as I believe its this best goal. Remember that size is not what defines a TLP, community is. However, as none of the main [math] developers want this at present we need to find a second solution.
I am more convinced of the need now. The proposed new list is in fact not really a "user" list in the classic definition that we have. It is much more of a discussion list.
The only question I have is whether votes, and actual code discussions will occur on this list or not. (And I ask that from a supervision point of view - HttpClient lost supervision/review when they created a new list)
So, I reckon that commons-math-user (or commons-math-discuss?) could be the best solution to the problem. Consider me +0.5 so long as votes and true code discussions remain on commons-dev.
Stephen
Henri Yandell wrote:
I'd also suggest that 3 months after creating it, Mark would have to justify the creation by showing that community has begun to grow.
I think math could accept such an agreement. What would be a justification, "list activity" above a certain threshold?
Existence of community. No idea how you measure this :) Noise is one, number of active members would be a better one, but you have to measure active as people making noise.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
