Changing backgrounds

Posted Apr 28, 2008 12:54 UTC (Mon) by ekj (guest, #1524)
In reply to: Changing backgrounds by russell
Parent article: The Grumpy Editor encounters the Hardy Heron

Yes. Plain and simple yes.

If doing that USED to work, but doesn't work TODAY, then it is a regression.

It matters not how the developers, or anyone else, INTENDED it to work, what matters is actual
behaviour.

Actual behaviour is that right-clicking the desktop and setting the wallpaper USED to work,
even for people without Nautilus running. Today it no longer works. That is a regression.

It's not a major issue or anything, not as if this makes the computer unusable, but a
regression nevertheless.

Changing backgrounds

Posted Apr 28, 2008 19:40 UTC (Mon) by jospoortvliet (guest, #33164) [Link] (1 responses)

Hehe, talking of regressions, you'd like KDE 4.0 ;-)

More serious, this is a weird thing. I would personally say it's good to 
reuse code (use Nautilus code to paint the background) but it was done 
wrong (needs Nautilus running somehow). Should be factored out in a common 
library, I would presume.

Imho despite all it's shortcomings, KDE does things better in this 
regard - the infrastructure is in order. Let's see if our Grumpy friend 
will get acquainted with KDE again when 4.1 is out - see if he likes it.

Changing backgrounds

Posted Apr 29, 2008 6:09 UTC (Tue) by ekj (guest, #1524) [Link]

I personally don't mind minor regressions much. I deal, and like our grumpy editor I like
living on the edge, and I'm fully aware that that means suffering bruises and cuts every now
and then.

I was just responding to the (imho!) very misguided idea that something that stops working
somehow isn't a regression if the developer never "intended" it to work in the first place.

I do indeed like KDE4, bugs and all.