I started analyzing the feedback reports we have received so far. There are 170 reports, of which I think 138 are valid (haven't checked what all the others are, but there are some incomplete versions I sent when developing feedback).
Of the 138 valid reports we had 84 that reported seconds since last failure greater than 0. I am not sure why we have so many zeros, yet. Doing some calculations with the non-zero values (yeah, this skews the numbers in our favor) yields stuff like: longest time without failures is almost 12 days, average about 7 hours, but most people see the feedback dialog every four minutes or more often. We have 14 tracebacks that have happened at least two times, the top traceback has been reported 775 times (please note that a report can actually include the same traceback several times). The three most often reported tracebacks all happened during OnIdle. I haven't yet filed bugs on these issues (need to search bugzilla first etc.) You can see the results here: http://wiki.osafoundation.org/bin/view/Journal/FeedbackAnalysis20060905 This is currently all manual, but once we have some ideas what kind of data is useful to us we can automate the most frequent stuff. -- Heikki Toivonen
signature.asc
Description: OpenPGP digital signature
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Open Source Applications Foundation "chandler-dev" mailing list http://lists.osafoundation.org/mailman/listinfo/chandler-dev
