User:NeilK/Worklog/2011-02-22 to 2011-02-28

From Wikimedia Foundation Governance Wiki
Revision as of 03:51, 26 February 2011 by NeilK (talk | contribs) (summary of my week...)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Did not keep daily notes, here is just a summary:

- Continued to work on various UploadWizard issues, especially with Kaldari.

- Met with Mark Hershberger, Alolita Sharma, Ryan Kaldari re: bugmeister role, how this relates to WMF projects and those that aren't official products. MH was changing our blocker bug for UploadWizard 1.0 which we felt was a bit of an overreach.

MH asserts Rob Lanphier wants Bugmeister to essentially act as development manager for all the bugs, actively follow everything. (!!) Seems a bit impractical to me, instead Bugmeister should just attempt to ensure that issues aren't forgotten, not to manage their development. Setting aside that question, main question is how do we want Bugmeister to act with actively managed projects. No need for him to be "product manager", we have one. Chose to use Bugzilla assignment to some milestone to 'signal' whether bugs are being actively triaged. Those of us working on UploadWizard think we are using Priority and Severity correctly, and want both, so Mark Hershberger said he may create a custom field for his own use.

- Got TrevorParscal to help with my issues with the Javascript PEG parser. He spotted the cause of the infinite loop bug. Pretty soon after that committed the library, although it is not yet hooked up to UploadWizard or RL.

- Attended preso on Korean Wikipedia from visiting member RyuCh

- Cataloged all bugs with UploadWizard to figure out how long it will take to get to 1.0 (people are understandably anxious to see it released). Estimate looks like 20 to 40 days. Large uncertainty at this moment because I spent a lot of time in Dec-Jan fixing underlying architectural issues without closing related bugs, and Kaldari is finding most assigned bugs to be already fixed before he even works on them. So many bugs may be zero effort. Also, I strongly suspect all "error" related bugs can be solved at once, but this may not be the case.