I was really looking forward to seeing a lot of your excellent T3 UI work being piped into Blender immediately after the 2.43 release - giving Blender the skin that you designed for the various components really would help both make it far more professional looking ( the visual style was excellent ), and easier to use ( more whitespace due to rounded corners, faded out controls when disabled etc ).

The other major contribution you put in ( just as important IMO ) was your patch-in-progress for a) adding objects and not entering edit mode, and b) adding objects in perspective mode, in non-rotated mode ( perhaps with the exception of text ). I have taught a few Blender / GE classes, and this is vitally important.

Keep up the great work, and be sure that many people are waiting to see all of your hard work integrated into Blender 2.5!

MAl

Check out Elephants Dream website, it looks like it has been hacked or the server move didn’t went as expected …

– Rui –

Yes, those things did seem to work ok, but they were more like proofs of concept, to be further developed into something more stable and permanent. Some of it was a bit hackish, and some of it was using temporary short term fixes that would not be so flexible or sustainable in the long term (perhaps causing similar types of problems as we are facing now, due to Blender’s design which was focused on the short term needs of NeoGeo in the mid 90s).

These structural issues need to be dealt with properly, the problem is that Ton’s the only one who really knows the insides of Blender well enough to be able to do it well, and he is a finite resource.

In any case, although those features I made do work, and I’m happily using those tools to this day, I’ve been told they’re not in a state to be committed at the present time. Just as I’d hope that coders would listen to me for design advice, if an experienced programmer advises me it’s not good to commit, I’m more than happy to defer to his or her knowledge, particularly in a collaborative team project such as this. Of course this has down sides, such as the one we’re discussing :)

Regarding Tons plans, this is truly unfortunate. Seeing the UI constantly being under-prioritized when it’s in glaring need for improvement is sad.

However, you keep saying you need the events refactor in order to commit all these lovely UI changes. They seemed to work very nicely in the form you already coded them, without any events refactor. Are your initial implementations flawed without the refactor? If it’s just a matter of clean code, how about committing most of the changes soon-ish in their current form (which is still a gigantic improvement), and then implement it in a cleaner way whenever Ton decides to do the refactor?