>From: Frank Schönheit - Sun Microsystems Germany <[EMAIL PROTECTED]> >Sent: Mar 19, 2007 1:16 AM >To: [email protected] >Subject: Re: [qa-dev] Base Results > >Hi Andre, > >> James Mckenzie schrieb: >>> Finished Base tests. Issue 49204 is still with us? >> >> Of course it is. And a good example why it is hard to encourage people >> to do more than one TCM test-run. We do always the same tests - and find >> the same errors again and again. >> >> <ironic> >> Maybe we should implemen templates in TCM? "Click here to set known >> issues as failed". >> </ironic> > >Hmm. Didn't even know this issue comes from TCM. Anyway. > It does. There is a test to look at the alignment functionality in a table. I don't know why this is implemented here and not in a report. In my mind, it does not make much sense to align a field in a table in a certain fashion, but in a report it very much does. I will look at the alignments available in the report feature.
>Please look at issues targeted "2.x", with owner "oj" or "fs", and >you'll see that the Base team has a lot of other things to do, before >fixing a "P4 OOoLater" task. I will do so this evening. > >As much as I understand that it's frustrating to report/QA bugs which >don't get attention for years, that's, as said in another thread here, a >matter of resources and priorities. > I am well aware of this limitation. I was questioning if it even was worth looking into as I don't think this feature is even used much in the real world. >And, in the hope this doesn't sound too harsh: I would not be willing to >spend any (non-negligible) amount of time into an issue just to satisfy >TCM (or any other QA process, for that matter). That is, if the bug does >have a very limited impact only *in real life* (as indicated by the >priority), then I'd consider it a waste of precious resources to fix it >before the more important issues. You will not get an argument out of me on this. > >I know that in testtool, as well as in the automated UNO API test >scripts, there are mechanisms to log such "this is a known bug with >number #i12345#" events, and to *not* count them as test error. I'd say >TCM needs something similar - simply because QA resources are equally >precious, and should not be wasted, too. > TCM needs more than just a small space for an issue as you have to go look up the issue to see if it affects the platform you are testing on and if it has been fixed or not. This would help greatly in the TCM QA effort. Maybe another issue to file? James McKenzie --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
