Oxygen's new window-move feature fatal for KJumpingCube game

[Hugo Pereira Da Costa]

On Friday, June 18, 2010 04:57:41 pm Thomas Lübking wrote:
> Am Saturday 19 June 2010 schrieb Ian Wadham:
> > No.  Although the question is off-topic, it is seriously put and I would
> > like to see answers.
> 
> Speaking only for myself, I introduced it to Bespin because of the seamless
> client/decoration transition (which Oxygen shares) made it somehow
> necessary to catch clicks on the menu/toolbar and dialogs.
> 
> I then extended it to statusbars (mostly to have easy dragging on the other
> end - some sort of "coutertitle") and when Hugo contacted me this April
> (w/o even bothering to reason his move :-) we refined, "fixed" and
> extended it a bit more :-)
> 
> Given that the WM/client separation is hard to understand for Joe User
> anyway (i recall my early X11 days here...) and is now even visually
> absent, this feature -aside usability- makes sense and despite being
> opt-in (in Bespin) I got quite some positive feedback on it.
> 

Hi Ian,

Same here. Most feedback has been positive. 

As for answering your question: 
> The new window-move feature is a done deal, but
> how much consideration will core developers give in future to the possible
> costs and inconvenience caused to application developers and
> end-users (not just "advanced" users) when new KDE features
> are introduced?

Obviously, a new feature that breaks otherwise well working applications, is 
not acceptable. I believe everybody's aware of that and carefully tests his 
features against anything he can think of possibly being broken. At least I 
did, for this dragging stuff. Now, in open source world, you understand it is 
hard to test _everything_, and I think this is why we have beta cycles: 
notably to allow "daring" users to test their favorite application against new 
experimental features. The counterpart of having the brand new kde "in 
advance" with respect to basic user, being that yes they might have to 
dedicate some of their time to write bug reports. 

In fact, as a dev, I rely quite some on these daring users, on top of the 
people I talked to on IRC and by email, to catch the corner cases that none of 
us have caught before. Beta cycles is usually a time during which a lot of bug 
reports are filled, which is good. The number of bug reports and my ability to 
fix them, during the beta, is what decides whether this or that feature will be 
enabled or not for the true release. Does that make sense ? 

Hugo

> Cheers,
> Thomas