On Sat, Nov 2, 2013 at 3:28 AM, libreoffice-ertai.nl <[email protected]> wrote: > Hi All, > > I think we should just collapse to a 3.x series and have that tell you that > it is too old.
That sounds fine for 3.x builds. Per the ReleasePlan page, 4.0 will reach EOL before the end of this month, so we should have a solution that can include those builds as well. > > But what I also noticed was that we are asing for: > - Version the bug appeared > - Latest known-working version > > It seems to me that the 1st one is always the 2nd version + 1. If somebody > looks at it. Sure, that's the ideal state it should be in after triage and bibisecting*, but I think that most users won't fill in the Latest-known-working-version field, and for those who do, there may be a gap of 2 or more versions, depending upon when they upgraded LibreOffice. > Shouldn't one of those be the version people are working in, as that needs > to be checked by BSA to see if it needs to give this warning. > Personally, I'd be fine if someone said "I see this bug in 4.x but I remember that it worked in 3.3". Are we going to have to test to confirm that? Sure. Do we have to do a bit of sleuthing to track down when the bug was introduced? Sure. But this additional information gives us at least a hint that this is a regression, which is useful. So Version-bug-appeared should be pre-EOL, but Latest-known-working-version can be EOL. --R * technically speaking, bibisecting will let us drill-down more precisely than just our released builds, but we all get the general point _______________________________________________ List Name: Libreoffice-qa mailing list Mail address: [email protected] Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette List archive: http://lists.freedesktop.org/archives/libreoffice-qa/
