Bug#565219: qa.debian.org: bug history graphs are incorrect
On Sat, Jul 17, 2010 at 08:00:42PM +0200, Francesco Poli wrote:
> On Fri, 12 Feb 2010 20:42:35 +0100 Mike Hommey wrote:
>
> > On Fri, Feb 12, 2010 at 08:27:53PM +0100, Francesco Poli wrote:
> > > On Fri, 12 Feb 2010 19:53:48 +0100 Mike Hommey wrote:
> [...]
> > > > New graphs are available on http://merkel.debian.org/~glandium/bts/
> > >
> > > They look nice!
> > > They even look correct and in sync with the PTS, at least from some
> > > random checks!
>
> I keep taking a look at those unofficial graphs, from time to time.
>
> On a second thought, there's something that I don't especially like in
> them, from an esthetic point of view.
>
> If I compare them with the current official graphs, I notice a number
> of cosmetic differences: many of them look like an enhancement to me,
> but two of them seem to be a regression.
>
> The first change I don't like is that the background grid is now made
> up of dotted lines, rather than of solid lines.
> I would prefer solid lines: I think they make the graph more readable.
>
> The second change I don't appreciate too much is that the background
> grid seems to have horizontal lines for integral values of the vertical
> axis (which makes sense), but also for half steps (which does not make
> sense to me: the number of bugs is always an integer, may a package
> have 14.5 bugs? I guess it cannot...).
> I would prefer having fewer horizontal lines (at most one for each
> integral value of the vertical axis), since having too many horizontal
> lines makes the graph hard to read.
>
> I hope that constructive criticism may be helpful... :-)
I wouldn't mind someone searching for the appropriate rrd flags to make
that happen ;)
Please note that those on people.d.o already show for half steps, but
those on merkel show more. See
http://people.debian.org/~glandium/bts/n/nss.png
vs.
http://merkel.debian.org/~glandium/bts/n/nss.png
for example
Mike
Reply to: