On 7/8/07, Ralf Hemmecke <[EMAIL PROTECTED]> wrote: > > Some ideas: > > 1. Assuimg Axiom becomes part of SAGE, all Axiom patches are > > to be voted on democratically among the Axiom deveopers. > > (Majority rules.) > > > Any comments on any of this? > > Do you work in sage like that "majority rules"? I know that I simply > feel unable and therefore not very well if I have to vote on a patch > that I don't understand.
No. The way it works in SAGE is that you submit a patch. If it is your own work (eg, not a bug fix requested by William), he will send it to a "referee" http://www.sagemath.org/jsage/editors.html who checks it over, offers suggestions, etc. The goal is to have a refereeing job done within a week. All new code must have preworked examples in the docstring which are automatically run using a command "sage -t filename". These examples must be representative of typical useage and must work (ie, must "pass all doctests"). Basicaly the referee checks these and perhaps perhaps makes suggestions for small improvements to the code. Once the patch is approved by the referee, the referee then sends it back to William and it is included. If it is not your own work (eg, a requested new feature), then some times a long discussion and voting takes place on sage-devel, after a "call for comments"-type email is posted. Of course, if you don't know or care about the feature then don't vote. However, William can over-rule votes. I can't remember the last time that happened. Something about a suggested implementation being inconsistent with a feature or goal of SAGE I think. I really think it was a case of William just knowing the system better than anyone else and making the proper decision, but I can't remember the details. > > I know that Waldek has done a lot of good work for Axiom, but I have not > looked at even one of his patches. I somehow don't think that voting by > majority is a good strategy. Does a referee-style system, perhaps in addition to a "call for comments" post to an email list, seem reasonable? > > I have seen that LYX developers need at least two OKs until a patch is > accepted to trunk. That I find quite OK, but it is all a question of > what policy should be accepted and how many developers are actually > watching and contributing. > > Ralf > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
