On Sat, Apr 15, 2006 at 01:15:57PM -0700, Bryce Harrington wrote:
Another thing I'm wondering about is how much bug fixing work will be needed. For 0.43 I think we actually spent _too_ much time in bug fix mode, to the point that it almost burnt out a few people (which may be the cause for the long period between releases). Yet at the same time, much development has gone on, and we may have a lot of defects to eliminate. We'll have to find a fair balance there...
Hmm, actually it looks like the bug tracker is in quite good shape. In November last year, around the time of the 0.43 release, we had around 250-300 bugs (out of 1736 total), and today we're at 339 (out of 2068). Looks like Bulia and others have been doing well keeping on top of the tracker. That puts us in a good spot going into the release. :-)
How do people feel about the overall robustness of inkscape presently? I assume there are still some issues in recently implemented features, but beyond that do you feel there is a lot of stabilization work needed, or just a little?
Bryce
Thanks for the status Bulia. I have been chatting with individual developers to collect current status, so have the details to fill in some of the holes. I will take the action to check with some of the additional developers you've listed, and will post a roll up summary later. I'll also check and see who would be available and willing to be release coordinators this time.
From what I've gathered so far, we need at least a couple weeks' more
development before feature freeze. Perhaps we could tentatively identify May 1st as the earliest possible freeze date. As we get more status we can decide if that's doable or push it back to something later.
Bryce
On Mon, Apr 17, 2006 at 02:49:42PM -0300, bulia byak wrote:
In preparation for 0.44, I'd like to share what I know about the status of various developments, as well as ask people to submit their updates. Ideally I'd like to see EVERY developer to respond to this email with their status updates.
But first, some food for thought. Judging by SF statistics, 0.43 is the first Inkscape version where we failed to double the number of downloads. Up to 0.42, the approximate doubling rule held (if we count 0.42 and 0.42.2 as one version), but 0.43 got only a bit more downloads than 0.42: 347000 vs . 300000, even though this release was longer. May be a stats glitch, but may be an indication that 0.44 needs to be marketed more aggressively.
Here are the items I'm aware of, in no particular order:
- Snapping. I sent a call for Carl and Mathieu some time ago but got
no response. Even if we leave this in its current status, we must do at least two things: fix the freezing when dragging multiple selected objects (http://sourceforge.net/tracker/index.php?func=detail&aid=1429049&gro...) and write the release notes for the new stuff. In the absense of the main authors, help is welcome from anyone who has an idea of the changes since the last release (Ralf?).
- Andrius has a very interesting rendering speedup patch. It can't be
committed yet since it has a problem with scrolling, but the author says he can fix this. In that case it definitely should be in 0.44, even though it's a bit risky. But we have time to test it still.
- Scislac has taken on collecting/making screenshots and updating the
tutorials. I don't know of the current status of this.
- Layer dialog: Mental has promised to commit it before the release.
Looks like it's the biggest remaining feature item, and its completion will determine the date of the release. Any status update will therefore be especially appreciated.
- Jon Cruz seems to be in the middle of a hacking bout, but again I
don't know exactly what's done and what is planned. Filling out the release notes would help a lot. Jon, please do it NOW.
- Aaron has finalized quite a number of useful features, the
remaining one being the "crop canvas" thing which just needs to be hooked up. After that (I think) he was going to work on the path effects branch - which means he will, at least, have something to hack on while we all will be painfully frozen :)
- I need to do one interface tweak in Calligraphic (requested by
Mental) and maybe one other small thing. Both pretty small.
- Richard (cyreve), can you please fill in the Release Notes on the
bugfixes you did to the Windows font handling?
- Last I tested, extensions still don't work on Windows for lack of
PyXML. It would be a pity to not enable them because of this. Ishmal, can you look into it?
- Bob (ishmal), what is the status of the SIOX tracing and ODG
export, as well as DOM? Will any of them be usable for 0.44?
- One small but useful thing that seems to be almost there is help
text for extensions, to be shown on extension params dialogs. Ted, what is the status of this?
- On Jabber, someone called "deadchip" started work on text dialog
and toolbar and even showed some code. If you are reading us, can you please say hi and give us an update/estimate? This is a very important missing piece in Inkscape UI.
- Does anyone know the status of the Windows "dialogs on top" patch
for GTK? Scislac, as you contacted the author, can you please write him again? Not exactly our problem, but would be nice to have it finally fixed.
- Bugs to be fixed: apart from the drag freezing in snapper, I know
of some weird tablet event bugs that Mental is debugging. Anyone willing to nominate bugs for mustfix status please follow up.
Anyone having something to add to this list is very welcome.
Also of interest is this old mail of mine: http://sourceforge.net/mailarchive/message.php?msg_id=13934323 Some things listed there are done, some are as far ahead as they were then. But overall, we made good progress.
Lastly and MOST IMPORTANTLY, we need to designate one or two release maintainers willing to keep and update this list and push people to fulfill their obligations. The release time is the time when anarchy is suspended and a dictatorship is temporarily installed. The maintainers need to be aggressive and tireless: send daily reminders, make noise on jabber, urge and ask and plead and beg to get things done. Every item in this list needs to be followed up until it's resolved somehow, and every new issue that emerges needs to be added to this list. That is the only way to get the release out of the door fast and to avoid the endless and meaningless freeze that so frustrated us last time.
(as you might guess, I can't be a maintainer this time - too much work lately...)
-- bulia byak Inkscape. Draw Freely. http://www.inkscape.org
This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid%110944&bid$1720&dat%... _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
Bryce Harrington wrote:
On Mon, Apr 17, 2006 at 02:49:42PM -0300, bulia byak wrote:
But first, some food for thought. Judging by SF statistics, 0.43 is the first Inkscape version where we failed to double the number of downloads. Up to 0.42, the approximate doubling rule held (if we count 0.42 and 0.42.2 as one version), but 0.43 got only a bit more downloads than 0.42: 347000 vs . 300000, even though this release was longer. May be a stats glitch, but may be an indication that 0.44 needs to be marketed more aggressively.
just a thought: I'm sure there's others like me that now simply go along with the Ubuntu, etc releases which means waiting ~6 mos for the newest app versions rather than getting into the whole dependency game, which means we are still running 0.42 on our production machines. 0.44 will probably be too late for Dapper but may not have many non-Dapper dependencies so it might be an easy backport that we could put on the download site. Maybe this becomes part of distro testing.
John Taber wrote:
Bryce Harrington wrote:
On Mon, Apr 17, 2006 at 02:49:42PM -0300, bulia byak wrote:
But first, some food for thought. Judging by SF statistics, 0.43 is the first Inkscape version where we failed to double the number of downloads. Up to 0.42, the approximate doubling rule held (if we count 0.42 and 0.42.2 as one version), but 0.43 got only a bit more downloads than 0.42: 347000 vs . 300000, even though this release was longer. May be a stats glitch, but may be an indication that 0.44 needs to be marketed more aggressively.
just a thought: I'm sure there's others like me that now simply go along with the Ubuntu, etc releases which means waiting ~6 mos for the newest app versions rather than getting into the whole dependency game, which means we are still running 0.42 on our production machines. 0.44 will probably be too late for Dapper but may not have many non-Dapper dependencies so it might be an easy backport that we could put on the download site. Maybe this becomes part of distro testing.
I do think it's worth considering getting on the Gnome release schedule. For starters there wouldn't be any question on when the next release is. I can see a major benefit being that we know we won't go over 6 months without a release. Not exceeding 6 months seems like it would be good for the marketing aspect... we're always reminding people that we're out there and the project is still being very actively developed.
If we want to have another release in-between the 6 month cycle, tis all good. For us to be on the Gnome release schedule we would have the benefit of a current version going into new releases of Ubuntu and Fedora (as well as any other distros that follow Gnome's schedule). A schedule also provides solid deadlines for people to know when their work needs to be finished. To me, it seems like a little more structure would only benefit both us and our users. Eh, just my .02.
-Josh
participants (3)
-
Bryce Harrington -
John Taber -
Joshua A. Andler