Re: RC-stats paragraph on DPN
On 13/11/12 at 13:31 +0100, Cédric Boutillier wrote:
> On Mon, Nov 12, 2012 at 03:21:07PM +0100, Francesca Ciceri wrote:
> > Hi there,
> > I wanted to ask your opinion about the RC-stats paragraph on DPN:
> > historically we relied on Alexander weekly blog post on the number of RC
> > bugs and related stats. When he stopped to publish it, we decided to use
> > directly the udd queries to give our readers those numbers. Now,
> > fortunately, Richard Hartmann started to send a weekly blogpost similar
> > to Alexander's one, so we decided to switch back to it.
> > As you probably know, DPN is frozen on Friday night and published on Monday
> > morning. This means that sometimes the number of RC-bugs on the paragraph
> > are not precise. So, here the question: do we prefer to have a link to
> > a blogpost (Richard's one) providing a detailed analysis of the numbers
> > (which are slightly inaccurate at the moment of DPN publication) or we
> > prefer to use directly udd queries (ie: means more accurate numbers, as
> > we can add them last minute, but not detailed analysis of them)?
> > We could also ask Richard to publish his blogpost on Monday morning-
> > What do you think about it?
> I think that we should rely on Richard's blog post, which provides
> detailed statistics. I believe there is more value in the detailed
> statistics, for data a few days old, than in the two numbers we collect
> from udd, which anyway risk to differ from the value the readers could
> get from the same query, because of the volatility of the number of
> bugs. We could mention explicitly that the numbers from Richard's post
> give the status for (the end of) a particular week, a point the readers
> to the UDD link if they want current values.
If that's helpful, you can generate your own data with the same queries as
Richard's from http://udd.debian.org/cgi-bin/rcblog.cgi