[kde-linux] Another KDE 4.x print problem?

[Duncan]

James Tyrer posted on Sat, 31 Oct 2009 16:13:02 -0700 as excerpted:

>> tho seriously improving with each release.
> 
> But, will it.  How many bug fix releases will there be in the 4.3
> branch?  Or, will it just go to 4.4.0 which means more features not bug
> fixes -- actually new things that are not quite fully baked yet.  So,
> there will just be more things to go wrong.

The kde 4.x bumps are on a six-month schedule.  The 4.x.y bumps are on a 
monthly schedule, so they get to 4.x.4 before the x bumps. They've been 
sticking to that reasonably well, tho not absolutely set in stone.  

But they *DO* still need the 4.x bumps, because the 4.x.y bumps are 
relatively limited -- I believe they can't change UI wording much, for 
instance, as they're string frozen for translation purposes.  And there's 
still enough /big/ things wrong that they *NEED* the 4.x bumps regularly 
in ordered to be able to fix a reasonable number of them each time, which 
they *HAVE* been doing.

And actually, they do seem to have calmed down on the featuritis somewhat 
too, and be focusing to a reasonable extent on fixing the still gaping 
holes.  4.4 is badly needed, as fixes are already in for a number of 
them, but it's not scheduled until February.  4.3 was the same way, as 
was 4.2 and 4.1definitely because 4.0 was basically hacked-up demo apps 
around finally frozen core-libraries.

There are yet three major things I disagree with, and one of them I'm 
skeptical of but they /might/ pull it off, so two I can honestly say I 
fully disagree with at the moment.  #1 is the statement that 4.2 was 
ready for the normal user -- IOW, release quality.  I flat-out disagree 
with that entirely.  4.3 isn't even there yet tho it's certainly closer 
that 4.2, but 4.2 missed it by about as much as that NASA Mars mission 
missed when they screwed up the metric/imperial conversions!

#2 follows from that, and has to do with the fact that they aren't 
following thru on the very public promise to keep kde3 supported as long 
as it had users, or more in general (that is, disregarding the promise), 
keeping the last decently working version, 3.5.9+, fully supported until 
the new version can properly replace it.  Of course, they say that it 
does with 4.2, but as I said, that's a huge difference of opinion there, 
as there's no way I would or could ever even attempt to make that claim, 
and pretend I still had any integrity at all left.  Yes, I DO feel that 
strongly about it!  But anyway, the fact is that 4.x is seriously broken 
in enough areas that it is NOT sufficiently ready for many 3.x users to 
make the switch, and while I believe 4.4 will at least be arguable to 
change that (IOW, I expect that I could make that claim personally 
without losing all respect for my own integrity), I think it'll really 
happen, in general, with 4.5.  And as long as that's the case, 3.5 should 
remain supported, but it's not, and this is my second major point of 
disagreement.

The third one, still potential, is that given the kde4 track record so 
far, I'm NOT particularly looking forward to the transfer kmail transfer 
to akonadi, scheduled for 4.5.  In fact, that one thing /alone/ IMO has 
the potential to derail the entire 4.5-is-finally-stable-enough-to-switch 
train.

The ray of hope, however, is that they /did/ have the good sense not to 
try it for 4.4, where only kab, kaddressbook, is supposed to make the 
change.  Given the kde4 track record so far, and the fact that it seemed 
/almost/ ready, I could have seen them /trying/ it.  But they aren't.
/Maybe/ they're actually learning that people depend on stuff and they 
can't simply ship broken pieces and claim it's ready for normal use.  The 
kde4 track record so far doesn't give one a lot of hope, but OTOH, the 
track record so far would have had them trying to ship it with 4.4, and 
they're not, they're actually taking their time a bit, hopefully to get 
it right and test it well, and hopefully to ship something that's 
actually reasonably mature with 4.5.  We'll see on that one.  They may 
yet surprise me.  OTOH, that's what I've been saying since the first 4.0 
betas came out, too, and the track record so far is again, one I could 
not say I'd be proud of, and keep my own integrity, because IMO it'd be 
about as bald-face a lie as that promise of kde3 support as long as there 
were users turned out to be.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman