← Back to team overview

launchpad-dev team mailing list archive

Re: high bugs - doing a bit of a review

 

On Thu, 2010-11-18 at 18:26 +0000, Graham Binns wrote:
> > We only work on High bugs.
> 
> This isn't true for the Bugs team, at least when we do story work.

If you treat them as high, be honest and set their importance to high.
There are many bugs that many people will say are low, but since we know
a feature is not complete (maybe I mean satisfactory) without many small
issues are addressed, they are high.

I fix more low bugs than any other Lp engineer. I fix them like most of
us do, I see an opportunity and I cease it. I do not target the bug to a
milestone or even say it is in progress until I am certain I can fix the
issue without jeopardising the high bug I am working on.

I confess that I have a few low bugs scheduled because I am trying to
force an opportunity to fix them with a related high bug also in the
schedule. I think I have a 50% success rate. Even when I personally care
about working on a low bug, I have not found the trick of making the
opportunity happen.

> > I believe all high bugs should be targeted to
> > a milestone to show we are planning to work on it in the near future. If
> > a high bug is not ready to code, we are getting it ready to code, so
> > targeting to an feature/series milestone is often the right thing to do.
> > (If I could untarget from a series, I would suggest targeting to a
> > series).
> >
> 
> Does that make sense with our new continuous deployment strategy?
> AFAICT the notion of a release / milestone now only applies to DB
> changes and other things that can't be continuously deployed. 

A measure of time and sense of order is what is needed. The series is
all we need, but we cannot untarget from a series. We cannot fix a
mistake. If I assign a bug to a milestone in a series because I really
do need a bug fixed by a date, the bug will not show up in the series
report.

PS. I believe there are overlapping issues described in bugs on this
subject. There may be 5 duplicates of the root problem.

-- 
__Curtis C. Hovey_________
http://launchpad.net/

Attachment: signature.asc
Description: This is a digitally signed message part


References