launchpad-dev team mailing list archive

[Martin Pool <mbp@xxxxxxxxxxxxx>]

2010/1/13 Jonathan Lange <jml@xxxxxxxxxxxxx>:
> Are you suggesting not implementing bug heat then, or are you
> suggesting having an even simpler algorithm than the one already
> given?

Not implementing it for now.

I think the currently specified algorithm is so greatly reduced from
the ubuntu gravity algorithm (for good reasons) that we might as well
go all the way and just make it be the unified-affects-count.  That
avoids having a distinction-without-a-difference in the UI, and
perhaps would free up resources to work on affects-count bugs, which
we know would be popular with users.

I say 'we might as well go all the way' because I'm not convinced that
having 4 dupes with one user each is meaningfully different to one bug
with 4 affected users or 4 subscribed users.

Making the process of marking 'affects me' more rewarding, for
instance by having a list of bugs that affect me, would create better
input data.

If the point of the general heat concept is to help you assess your
new/untriaged bugs, there are things we could do to address that head
on: for instance a directly linked page of new bugs by affected-count.

Later on perhaps Launchpad could add per-project tag-based scoring tools.

I think many of us would like to see Launchpad get existing features
more polished before adding new ones and this is a chance to do that.

I am aware that bug heat has been a hot concept for years, and that
there have been offline discussions about it that I haven't seen.
Also that there is a roadmap saying this will be done and those things
have a certain inertia.  Ultimately the bugs team must do what they
judge serves their users best.  This is just how it seems to me.

-- 
Martin <http://launchpad.net/~mbp/>