On Sat, 2006-07-01 at 12:54 -0700, Elijah Newren wrote: > Shorter version this time... > > On 1/6/06, Ryan Lortie <desrt desrt ca> wrote: > > Hello Release Team. > > > > For next release cycle, I have a simple suggestion to improve the > > release process: > > > <snip> > > When you can go on Bugzilla right now and see "Blocker bugs: > > GNOME 2.12.x (these must be fixed before/in the specified GNOME > > version)" with a list of open bugs under it something is wrong. > > So, I talked to Olav and I don't think we have enough time to send out > the occasional blocker summary emails. So we do need a new volunteer > to do it. Can we get you to do this Ryan? It'd be great if you could > go over the list and send out an summary/update every couple weeks. Ok. As discussed on IRC I'm emailing a preliminary version of the list to this mailing list. I've removed most of the "blocker" bugs that were really obviously not blockers by removing the blocker designation or convincing the person who set it to remove it. Here's what's left. A lot of these are not legitimate blockers in my opinion but I'm not interested in stepping on everyone's toes. #122688: modal dialog popup + drag in progress = mouse freeze If the user is dragging an icon or rubberbanding in nautilus at the time that a modal window pops up then the entire desktop freezes. The only way to unfreeze the desktop is to kill off nautilus. mclasen says that GTK has a mechanism to recognise and avoid this situation but nautilus doesn't use it. Huge user impact and no reason it can't be fixed. #125364: vte screens are black when initially shown Low user impact and easy to fix (just refresh the screen). Probably should not be on the list. #133815: keys refuse to "unbind" #135476: Setting a key combination doesn't work multimedia keys Evil twins. #133815 is very evil because it causes you to lose the use of a key on your keyboard until you restart the session (should be a blocker). #135476 is less evil since it merely restricts you from using, for example alt+p for "play". Luis "tentatively mark"ed it as a 2.12 blocker but it doesn't seem bad enough to block a release for. #157941: Help browser content is inaccessible [REGRESSION] The screenreading software can't "see" the text in the help browser. You can make a reasonable argument that if the help isn't accessible then the desktop itself is not accessible. Probably a good blocker. #170207: Slow, incorrect rendering of this pdf file Some (rare) files render very slowly in evince. Luis marked it as a 2.12 blocker because "it is going to look terrible if 'our new awesome pdf reader! look how awesome it is!' in 2.12 is slow like a turtle on downers". The bug is in fact a problem with libpoppler (not Gnome) but remains open in the Gnome bugzilla due to its direct impact on evince. This is probably not a show-stopper. #302096: Crash while trying to burn a CD Nautilus crashes when trying to burn a CD. This happens an awful lot (from the number of dups) but nobody seems to be able to get a good stack trace. This is a great candidate for a blocker because it's serious and common. There seems to be very little traction, on the problem, though. #303280: evolution-plugin email receipt A crash that was fixed but then not fixed....? In any case it's caused by a very infrequently used feature. I'd have removed this from the blocker list myself if it wasn't a crasher. #310153: crashes on attempt to use in libgnomeui This one is a keeper. Somehow an invalid URI is getting into the bookmarks file and causing libgnomeui to crash on reading it. At the very least we can work around this problem very easily and hopefully it will be easy to fix properly. #312171: hal/non-hal incoherence: gnome_vfs_drive_get_activation_uri () This one is probably not a keeper. I'm not sure why it was added to the blockers list but I'm not really sure I understand the consequences of it and I don't want to step on toes by removing it. #313047: Evolution Mail Account Settings need some UI love Oh boy do they ever. Set as a show-stopper by a Novell employee but I disagree. As bad as it is, it's not stop-the-release bad. #314137: 'open folder' icon never reverts to closed after window is closed Luis marked this as a 2.12 show-stopper because of its high visibility. Even if it is highly visible, the user impact is very small. Not sure if this is a good candidate for actually blocking a release (it certainly didn't last time...). #314774: [patch] xrdb called needlessly multiple times Fixing this bug would improve our login time. Slightly slow login time is no reason to stop the show, though, is it? #317312: [CAN-2005-0023] gnome-pty-helper writes arbitrary utmp records We have an open security advisory and nobody has even touched it. Even though the results of this attack are minimal (faked log files) this needs to be looked into and fixed ASAP. #318951: Links breaks the spatialness of Nautilus If you have a symlink to a folder and you open it it opens in a separate spatial window from if you open the folder itself. This really does confuse the user rather substantially and ruins their idea of folders as objects. It should be reasonably trivial to fix this. I think this is a reasonable blocker. #319308: Accessibility: Folder name does not exist in canvas Marked as a blocker, it really isn't. It has a patch attached and needs a bit of attention (probably a quick close). #319549: gnome.ui.icon_lookup() segfaults on weird filenames This bug causes segfaults when rendering icons for odd filenames. I've tried to trigger this by creating some of these weird filenames and wandering into a directory containing them with nautilus. This does *not* trigger the crash. As such, I'm guessing the user impact of this problem is probably small. Probably not a good blocker. It has a patch that claims to fix the problem so it should get some attention. #320217: recoursive copy & replace fails and all files are deleted 'gnomevfs-copy foo foo' erases foo. This is a disaster but it looks like it's going to be fixed. An appropriate blocker. #324406: "Open Location" autocompletion truncates filenames Autocompletion in the nautilus location dialog truncates filenames due to utf8 vs. byte length confusion. Not a blocker but I've attached a patch to fix it and it needs attention. #325737: Nautilus crashes after adding files to burn area Nautilus crashes when you drag files into the burn folder on Solaris. This is obviously a big problem for Solaris users and a fix is attached. An appropriate blocker. #325861: gnome_vfs_async_xfer reports wrong GnomeVFSXferProgressInfo in the callback for http method I don't really understand this problem but it's well on it affects end-user applications in a visible way and is well on its way to being fixed. It was also marked as a blocker by people who know what they're talking about :) #325988: text files listed as application/octet-stream on non local system This is a fairly significant problem for people who are browsing files on a SMB (or some other) remote filesystem. Should stay as a blocker. #326323: Crashes on login with a11y enabled Gnome-session crashes on startup if a11y stuff is enabled. This is a no-brainer blocker. If some release-team or "senior-ish" bug people could go over some of the bugs that I've cast doubts on in this email and remove them from the blocker list it would be appreciated. At that point I'd be able to take the ones that remain and write an email to desktop-devel-list with the list along with some information about what "Gnome target" means (and when to use it). Cheers.
Attachment:
signature.asc
Description: This is a digitally signed message part