> On Jan 10, 2015, at 12:09 PM, Gregory P. Smith <g...@krypto.org> wrote: > > I agree with MAL, it is more beneficial to trust people and give out commit > access early.
For comparison, I think that is the norm for paid work. At companies I've worked for, new programmers are given check-in rights on the first day. To have become the Chief Data Scientist at Continuum, Davin had to impress the likes of Travis Oliphant and Peter Wang. I've certainly been impressed with how carefully and thoroughly he researches and thinks through everything he does. Dr. Potts brings a rich skill set and has volunteered to put substantial time into a complex module with many outstanding bugs and that has long been in need of some love. He's already spent time researching past discussions on the multiprocessing module, reading all of the 180+ open bugs to assess what needs to be done, and preparing two patches currently open on the tracker. We should at least grant him developer privileges on the tracker to it much easier for him to move the tracker items forward. As far as I can tell, no other developer is showing active interest in those issues (though Antoine did just commit one of Davin's patches). Commit rights are a separate issue, but I wouldn't see the point in saying no there either. The final step of committing your work is easy part. The hard part is forming a coherent view of the package as a whole, triaging the open bugs, preparing patches, and engaging in discussion with interested parties (if any). Raymond _______________________________________________ python-committers mailing list python-committers@python.org https://mail.python.org/mailman/listinfo/python-committers
