More lies^wstats on the krazy results

Found a bit of time to continue the number crunching saga. Two new datasets are added to the KDE SC Stats topic center. The first data set is about the number of krazy2 issues per module (e.g. kdelibs, kdeedu, etc.):

We see two modules that have clearly more issues than the others, kdelibs and kdepim. But, that was to be expected because when we look at the lines of code for these modules:

we notice that kdelibs and kdepim clearly have more lines of code than the other modules as well. Actually, the two graphs have quite some similarities which is not completely surprising. One of the things that draws the attention is that we can see that although kdebase-workspace, kdebindings and kdeedu have an almost equal evolving lines of code count, kdeabse-workspace has a significant higher number of issues. Well, as we said before, these raw numbers do not tell that much, so lets have a look at the number of issues versus lines of code:

In the above image we see how the each module contributes (or not) to the downwards trend for issues/loc I showed last time. In general, most of the modules have a somewhat downwards trend though not very convincing. The two modules contributing most to the current number of issues per KLOC are kdeartwork and kdewebdev (look at the most right bars of each module). Clear winners in getting te number of issues down are kdeaccessibility and kdeplasma-addons. Furthermore, props for kdeedu and kdeutils who got the number of issues down under 2 per KLOC for KDE SC 4.4.0. Soon to follow, stats per krazy check.