KDE's rough edges... what are your experiences?

[Duncan]

Michael posted on Mon, 28 Oct 2013 15:14:40 +0100 as excerpted:

> Am Mon, 28 Oct 2013 16:10:16 +0530 schrieb dE <de.techno at gmail.com>:
> 
>> I think KDE is not suitable for production environment. Just for casual
>> enthusiasts.
> 
> I guess that view is a bit too extreme, but interesting nevertheless. As
> annoying as the "typical" KDE-issues might be or get, there will be a
> point where users know the issues and can adapt to the "buggy"
> situation, so that KDE is not a general "show stopper" to their
> workflow.

I think it's extreme at one level, but not at another.

I already stated that I think kde 4.5 (and will further specify here 
later 4.5, so 4.5.4 and 4.5.5) finally reached "release quality".  Since 
then, the quality has gone up and down a bit, but as a general rule, by 
4.x.4 or so it tends to be reasonably good, "production environment" 
level, for at least some definition thereof.

What's more disturbing to me, however, and why I agree that at a 
different level, dE's correct and kde is not production environment 
quality, is the longer term record of claims made vs. reality.  We have a 
situation where a very public statement was made that kde3 support would 
continue as long as there were users, which WOULD have been production 
environment quality, but then exactly the opposite occurred, support was 
dropped for what real users were saying was the only reasonably working 
version.  At the same time, kde was publicly insisting that the new 
version was ready, while the facts were very clearly otherwise, both 
because even the devs were saying various bits weren't ported yet, and 
because the real users were simply finding the new version unworkable in 
the state it was in at the time.

That's not production environment quality by pretty much any measure, so 
even if the code does arguably literally reach production environment 
quality at some point (as I assert it did with late 4.5), taking the 
project as a whole including the claims made and evident behavior seen, 
no, kde in its 4.x state is NOT production environment quality -- it 
cannot be depended on over time to maintain a product that can be relied 
upon at claimed level of support, because the quality of the in-support 
code drops *WELL* below production environment quality for **YEARS** at a 
time, with users being left in the lurch.

It remains to be seen if that has changed.  With the 5/frameworks effort 
in full swing now, we'll see over the coming couple years just how much 
kde learned with the early kde4 disaster.  If they continue to support 
kde4 code until *USERS* say 5/frameworks is actually ready, or at a very 
minimum, refrain from claiming that they'll do so and then dropping 
support just like that, then when USERS say 5/frameworks code is 
"production environment" ready, a reasonable argument can then be made 
that kde has learned from its earlier mistakes and really /is/ production 
environment ready, even if literal code quality does drop below that 
level from time to time.

I'm actually quite optimistic, as the plan I've seen stated is to allow 
and support both 4.x and 5.x applications running side by side for a 
time, so users can upgrade individual apps to their frameworks-5 versions 
as /users/ consider them ready, while continuing to run other apps at the 
4.x version level until they (the users) consider the frameworks-5 
versions suitably stable to upgrade to them individually.

This actually fits the whole more modular emphasis and theme of 
frameworks-5 as well, so as I said I'm optimistic.  OTOH, the more 
cautious side of me says we've seen promises of continued support before, 
and we know how THAT turned out!

So we'll see, but I really AM optimistic, hopefully not to my own 
detriment.  Regardless, once bitten, twice shy, and I'm better prepared 
for a less-than-smooth transition this time, in part because over time 
I've been forced off of kde based technologies for one thing after 
another, so there's less kde on my system now TO be affected and the 
effect will thus be much more limited however it turns out, and if worse 
does come to worse, it'll be far easier to switch entirely off of kde, 
since there's simply less to switch out, now.

We /will/ see!

>> As of your problems -- if you continue to use KDE, you'll get used to
>> it. For e.g. now removable disks will now show up in device manager.
>> I've to restart KDE to fix it.
> 
> Used to it? That is most unlikely. I could tolerate such issues for some
> time but I guess I could never "adapt" to a point where I would not even
> realize the issues anymore.

Arguably, this is actually what happened to the kde devs themselves -- 
they became used to their workarounds to the point they ceased to even be 
aware of them as workarounds any longer, thus explaining their claim that 
kde 4.2 was ready for ordinary use.  Only the new users still trying to 
upgrade from the by then unsupported 3.5.x could see how horribly broken 
4.2 and 4.3 still were, because they still had to come up with their own 
workarounds, and weren't yet so used to them that they ceased to see them 
as workarounds but considered them normal practice, as the kde devs 
evidently did.

And in perfect honesty, I have to wonder if that's what happened to me by 
4.5 as well, thus my considering /it/ to be finally release ready -- what 
should have been 4.0.  I have only the dual supporting facts of all the 
protest more or less dying down by then, and my own personal knowledge of 
specific "deal breaker" bugs being fixed by then, to say otherwise, and 
in all honesty, I do still wonder sometimes if they're enough, or if I 
too was simply deluded due to familiarity with workarounds that had 
become so routine they ceased to appear to me as workarounds.

-- 
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

___________________________________________________
This message is from the kde mailing list.
Account management:  https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.