29 Jan
2012
29 Jan
'12
6:53 p.m.
2012/1/29 Calvin Morrison <mutantturkey(a)gmail.com>om>:
On Jan 28, 2012 5:21 PM, "Darrell Anderson" <humanreadable(a)yahoo.com>
wrote:
Build bugs should be marked as a blocker. It blocks compilation.
Fact. I even have a proposal for new layout and undelying
engine/structure. But I'm busy too much with exams/PKGBUILDs for arch
to continue work. I'll work on the website more after I finished
PKGBUILDs.
Oh, and layout can bee seen here: http://l0ner.github.com/html/ Please
note that it is unfinished, I need to polish it a little.
Git migration
Qt3/TQt3?
CMake - what's left
Observations/Concerns
Bugs:
what is high priority?
should we mark certain bugs blocker bugs?
Qt3 - do we have any problems with Qt3 right now
(performance or otherwise), that we should change? now is
the chance with an ABI change.
Artwork - seriously, we need to figure this out. Dead
serious. no more joking around
Website - is this even an issue?
How do we audit the bugzilla? What is a blocker to one person might be a
non issue to another. For example, a bug related to compiling a package
might affect only certain distros. Nonetheless, I think all
compiling/building bugs should be bumped to blocker.
The real challenge is which usability bugs are blockers, critical, normal
or just a nuisance? How do we collectively decide that?
I think we mark blockers ones that are fundamentally broken, or something
that effects a wide userbase.
There are dozens of paper cut bugs that must be resolved for R14. Almost
all of them went untouched with 3.5.13. Those types of bugs affect
perception of the project. We still have four months to go before the
tentative release date, which is a lot of time to address those kinds of
bugs. I don't mind if the release date is postponed to quash a majority of
those bugs. There are several people involved here who could resolve those
kinds of bugs while Tim addresses the more difficult bugs. I wish that would
happen. :)
We are making good progress already. A good amount of bugs have been closed
since 3.5.13 and many more will be. Many are marked Patch available.
Perhaps we should have a check box in the
bugzilla that identifies a bug
as a paper cut candidate.
With respect to the web site, I thought others
were working on that. Is
that not going to happen?
L0ner has noted serious interest in the website.
tdebindings.
Whether tdebindings is overhauled is not as important in the
short term as much as everybody is able to build the package. Can anybody
build tdebindings? I was able to under 3.5.13 but not since GIT.
I would say the priority is somewhere around that of koffice and khtml. Who
cares about kdebindings anyway?
cmake. Just about everything needs to be
converted. tdesdk and tdewebdev
were started and they should be candidates for the next wave of conversions.
We once discussed that we should focus on two to three conversions per
release. Unless somebody has devised a fast way to convert without
sacrificing quality we probably should remain with that agreement.
If we had a cmake conversion how-to at the wiki some of us could help
convert packages to cmake. Those of us who haven't yet done that could start
with smaller packages.
We need a quality assurance check list to ensure cmake conversions do not
leave out any component or build option.
Regarding window managers, I prefer that discussion take place post R14.
We have enough on our plate now. I think I read somewhere that to be fully
XDG compliant that any window manager is allowed to be used, but we should
discuss whether that is a project goal.
Xdg is a specification. A guideline. Not an API or a hard target. This is
low priority. An enhancement bug has been filed.
Calvin