[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: bug statuses for specific information needed?
From: |
Philip Nienhuis |
Subject: |
Re: bug statuses for specific information needed? |
Date: |
Thu, 20 Feb 2014 13:26:33 -0800 (PST) |
Mike Miller wrote
> Looking for comments, tell me if I'm overthinking or this isn't needed.
>
> Since I've been doing a lot of bug triage lately, there are many
> Octave bugs that tend to fall into two very specific (similar)
> statuses that are not quite "need info" and not quite "confirmed".
>
> * Bugs that need input from someone with specific mathematical or
> domain-specific knowledge. Not necessarily that they will be able to
> work on or fix the bug, but needs someone with expertise in
> constrained optimization, or matrix factorization, or even someone who
> knows their way around the GLPK library for example. Would it help at
> all to be able to set the bug status to something like "needs expert
> input"? Or would it confuse and not help bugs get fixed any faster?
+1 for "Needs expert input"
"Need info" could be renamed "Need more info".
> * Bugs that need someone with access to Matlab to run a test case.
> This often happens with the help of the bug reporter, but sometimes
> they don't have Matlab themselves, or don't have the necessary
> toolboxes. Would it help at all to have a bug status like "run this in
> Matlab for me"?
I'd feel that as some sort of defeat for us.
> Or is this too specific and unneeded? Or is this
> better solved through some other medium, like specific subject tags on
> the mailing list?
AFAICS such requests have always, often sooner than later, been picked up by
someone with access to ML.
The existing Item Group tag "Matlab compatibility" would do I think, and the
maintainers mailing list is a very responsive medium in this respect.
A bit OT, I think in the "Item group" list the tag "Crash" often seems
misunderstood.
AFAIU the core devs seem to interpret a crash as an occasion where Octave
itself stops working and is terminated.
Mere mortal users / bug reporters interpret errors that are catched with
some verbosity (error messages, traceback) as a crash as well.
Would it be a good idea to add a tag "Execution errors" or something like
that?
Philip
--
View this message in context:
http://octave.1599824.n4.nabble.com/bug-statuses-for-specific-information-needed-tp4662097p4662104.html
Sent from the Octave - Maintainers mailing list archive at Nabble.com.