SPDesktop now looks in good shape, which is good as that initially looked like the major show stopper. (Fixes thanks to rwst with some GC work from mental.)
Inkboard --------
A couple of people have reported either difficulty in connecting: bryce made a bug report, and sciboy today mentioned similar problems -- see 17:52 and previous at http://inkscape.gristle.org/2005-10-11.txt. sciboy says that it works if connecting to gristle.org, but that none of jabber.org, jabber.com, or jabber.org.au work. Later that day, Michael & sciboy did get a connection. Issues: - Lots of output on the terminal, should be turned off for general use before release. (If these messages are still useful to developers, then wrap the status messages in a macro or function that does nothing in the main version but that interested developers can "switch on".) I wouldn't want to release with these status messages: e.g. because they mask real errors that we'd want to be reported. - Michael & sciboy's session got out of sync somehow: at some point sciboy cleared everything but a star was left on michael's session. Not a show-stopper: this was the only thing we were aware of being out of sync. - Undo is global, i.e. it often undoes what the other person did rather than what the undoer did. This is a usability problem, but changing this is non-trivial, so I don't consider this a show-stopper. Btw, Michael was using MacOS X, so it seems that libloudmouth works acceptably there (as well as on Linux/Gnu). I gather Windows still doesn't work acceptably well, due to threading differences.
sciboy did experience a segfault later in his session with Michael (18:47 in the gristle log), though he didn't have a backtrace at the time. My feeling is that if the crash is specific to inkboard then we not consider it a show-stopper, but just make a note of it in the release notes.
Connectors ---------- Progress has been made. The bug tracker no longer has any connector bugs marked with priority >= 8. We still have some strange behaviour.
Michael comments: "The only known connector crash now is the one bryce reported, which I believe is fixed but he hasn't yet confirmed. I used to be able to reproduce it, but after fixing some other problems I now longer can. Once this sp_use_update fix is in that Bulia and I were just discussing (give me 20 minutes), then I think it's releasable. The other bugs are related to overlapping objects and are things I need to rework in libavoid after the release. The one remaining thing I'd like to look at if it's easy is the slowdown moving endpoints with the xml editor open."
menus.xml localization ---------------------- In principle fixed (there's one huge menus-skeleton.h string to be translated), but in practice none of the translations has been updated. It would be nice if we went back to smaller strings to facilitate keeping the translations up-to-date (and so that we could re-use the existing translations). Personally I'm still inclined to treat this as a show-stopper, particularly as it looks easy enough to fix.
Other ----- Sorting bug tracker items (http://sourceforge.net/tracker/?group_id=93438&atid=604306) by decreasing priority shows the following marked at priority 9:
- a text-anchor regression (cyreve, can you comment?); - the corrupted about.svg problem I posted on just now; - a bunch of windows-specific bugs that I haven't looked at: I'll leave these for our Windows devs to decide on; - "unexpected printout size", where we're wrongly converting to postscript at 90px/inch or 72px/inch rather than 120px/inch. Personally I wouldn't let this one hold up the release.
pjrm.