Weird issue with background covering almost all windows

Duncan 1i5t5.duncan at cox.net
Sun Jun 12 04:24:56 BST 2011


Richard Hartmann posted on Sat, 11 Jun 2011 17:24:53 +0200 as excerpted:

>> But we can probably speed the process even further, as from your
>> description, the problem is very likely kwin or plasma related.  So
>> after getting your backup, try deleting just the kwin files under both
>> of the dirs above, and see if that cures the problem.  If not, try the
>> plasma files, but I'm suspecting kwin so try it first.
> 
> I was plasma's fault. For now, I am more than happy with deleted plasma
> settings, but if anyone is interested in the config files to they can
> poke them, I will gladly send them on.

Well, I got it right that it was one or the other, but missed on which 
one.

I should have guessed plasma's files, as one of them in particular is 
horrendously complicated (it's still section and setting=value, but a 
section line has multiple ID segments, most of which enumerate the 
parents, so it's not as simple as just bisecting by section there, one 
must figure out the tree relationship and delete whole subtrees at once, 
when bisecting).  I've had it go bad on me before and had to trace the 
structure because as I said I'm a heavy customizer and didn't want to try 
to recustomize all my plasmoids and their positions from scratch.  But it 
was an experience I don't want to repeat, and towards that goal, I now 
keep backups.

And of course as any engineer will tell you[1], complexity breeds bugs 
and brittle configurations, so I might have figured plasma was the 
problem, as I already know how complex one of its config files (share/
config/plasma-desktop-appletsrc) in particular is.  But I went with kwin 
instead, guessing it was ether an always-on-top misconfiguration, or kwin 
getting mixed up and still placing windows on the now disconnected 
monitor's area.

---
[1] This is basic first-year Engineering-101 material.  More complexity 
means more things that can and ultimately some of them will, break.  When 
it's a useful or even functionally necessary feature, the cost of the 
additional complexity in higher maintenance burdens and/or risk of 
failure is however arguably justified, tho exactly where one draws the 
line on "useful" remains open to debate.  But if it's a mis-feature...

FWIW, It's for this reason that when I saw MS going down the whole 
product activation mis-feature route with eXPrivacy, that I decided that 
was one platform I wanted off of.  They were previously infamous for 
their bloat, but all those features were useful to some users somewhere 
so one could reasonably argue that they were justified.  But the 
activation mis-features added complexity and therefore risk of failure 
without benefit to ANY users AT ALL, only potentially to MS itself.  The 
implications of a company now adding mis-features of NO benefit to users 
at all, in fact rather the opposite as it's a hassle even when it works 
and the users are entirely legal, KNOWING it's going to break things for 
some of them... that a company is deliberately breaking legal customers 
as they certainly knew it would do, with NO possible benefits to 
customers... that was NOT a platform I was interested on being on any 
longer!!  So I upgraded from MS Windows 98 to, at the time, Mandrake 
Linux 8.1, as the best alternative to eXPrivacy.  Of course in doing so I 
quickly realized the benefits of software freedom and as I did so, I 
found there was even less reason or desire to go proprietary than I had 
realized before -- in fact, quite the reverse, since any such attempt to 
add a similar mis-feature in the land of software freedom will, to borrow 
the saying most often applied to attempts to censorship the Internet, be 
interpreted as the real damage it *IS* and quickly routed around.

So, rather ironically, in a very real way I have MS to thank for giving 
me that last push it took to switch to Linux and freedomware. =:^)

Meanwhile, realizing the implications of where MS was headed and that it 
had "jumped the shark" as the saying goes, I correctly predicted that 
with the precedent it was setting, within a few years other companies 
would be following its lead with their own malware, upto and including 
rootkits, etc.  And what did we have a few years later?  Sony's rootkit 
CDs fiasco.  I realized the implications and correctly predicted the 
result, but wasn't having anything to do with it.  I wanted off and I got 
off.  But in a way I understand how Sony must have felt, since MS got 
away with it and CONTINUES to get away with it, while Sony couldn't, even 
on the same platform that MS was pulling its tricks on.  That certainly 
must have seemed (and probably still seems) unfair to them.  What I 
/don't/ understand is why so much of the rest of the world took it, just 
like the proverbial frogs sitting in the beaker on the Bunsen burner.  I 
suppose it was /just/ that, that the "increase in temperature" was simply 
too slow for most folks to notice...

-- 
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.


More information about the kde mailing list