[Chandler-dev] Re: instructions for dogfooders

[Katie Capps Parlante]

Hi Andre,

Aha! I didn't understand that you meant a mailing list for dogfooders.

I agree with you that a community of technically skilled 
users/dogfooders helping out other users/dogfooders would be a real 
asset to the project. It is indeed frustrating to have a bug parked at 
"works for me" when the bug still occurs for the person dogfooding -- I 
can see how a forum for discussing the bug would be helpful. You bring 
up a good point that we have a window of a few months to think through 
the mechanics about how we'll support early adopters.

Anyone else have 2c on whether now is the time to start up such a list, 
in advance of Preview?

Anything else that would be helpful in establishing such a community? We 
weren't really ready for that in the past, but Chandler is now stable 
enough that we can start pushing on that as a goal.

Is "works for me" the right resolution for bugs the developers can't 
reproduce but we know happen for end users? If not, how should we handle 
these bugs? (Mail bugs in particular seem prone to this kind of situation).

Cheers,
Katie

Andre Mueninghoff wrote:
> Hi Katie,
> 
> Thanks for your interest. A little of both, I think, if I understand
> your example options.
> 
> My goal is possibly the same as yours, that is, to maximize OSAF
> developer productivity towards new features and functions, as compared
> to following up on bug reports that have questionable reproducibility
> instructions at best, or have a root cause in a misconfigured user
> environment at worst. It seems to me there might be/are probably others
> like me who encounter issues that occur more than once, but are unable
> to reproduce these issues consistently such to be able to craft a
> coherent bug report for further investigation by a developer. I think
> there may be an opportunity to provide a bit of structure to the
> community of "some-what technical, but non-developer" dogfooders through
> perhaps a list. It could happen, for example, that more
> technically-skilled dogfooders step in are assist others in
> distinguishing between known issues (a.k.a. duplicate bug prevention),
> user environment issues, and user expectation mismatches (expectations
> speeding ahead of the release plan). Dogfooders might be able to assist
> each other in verifying instructions for reproducing issues, all prior
> to intial triage <grin> by the developer(s) covering the relevant
> component. It seems to me that developers and dogfooders share a common
> lack of interest in bug research that results in a "works for me" tag
> though the dogfooder continues to experience the issue. The opportunity
> for OSAF as preview approaches is to shape and manage actively the
> expectations of the eventual beta users by working out in advance the
> mechanics of the support of early adopters. Hope that makes sense and is
> helpful.
> 
> Please let me know if you have any questions.
> Thanks, Andre