On Sat, Aug 17, 2013 at 5:33 PM, rupert THURNER rupert.thurner@gmail.com wrote: moved up from below because I'm answering your points in the context of gitblit:

On Sat, Aug 17, 2013 at 12:47 PM, Faidon Liambotis faidon@wikimedia.org wrote: let me give you an example as well. git.wikimedia.org broke, and you, faidon, did _absolutely nothing_ to give good feedback to upstream to improve the gitblit software.

• upstream bug: https://code.google.com/p/gitblit/issues/detail?id=294

I don't think filing that was useful or appropriate. That's akin to a user of [[Diplopedia]] or portlandwiki.org reporting a site outage at WMF bugzilla.

hi faidon, i do not think you personally and WMF are particularly helpful in accepting contributions. because you:

• do not communicate openly the problems

What do you call https://bugzilla.wikimedia.org/show_bug.cgi?id=49371 ?

• do not report upstream publically

and https://code.google.com/p/gitblit/issues/detail?id=274 ?

• do not ask for help, and even if it gets offered you just ignore it

with quite some arrogance

Help is *very* often welcome and accepted. (sometimes takes a while to get outside submissions reviewed and sometimes not. there are certainly people that can help if something's gotten stuck. e.g. Sumana)

As Ken said, getting the site working again is often a higher priority than making stack traces. This is especially true because it wasn't a new issue, time had already been put into investigating it and tweaking various parameters as well as working with upstream and it wasn't hard to reproduce the issue. (i.e. the stack trace could be made at other times if needed and could wait for it to be made by someone already investigating the problem; I don't know offhand how involved Faidon's been with the gitblit investigations but I know Chad was very (more?) involved with it.)

-Jeremy