Onto the whole reason I decided to post! First off, I've been working on a UI system. As many of you probably know (because I've stated it on numerous occasions), I love UI development. At some point in my career, I'd love to develop and release a library that could compare (NOT compete) with ScaleForm.
Anyway, I've been working on one and getting all of the little things working together is a pain. At one point or another, a lot of us go through the hassle of designing and implementing that nice UI system with panels that the user can drag, a huge list of controls, and a skinning system. In the end however, it's mostly for naught. 90% of the controls aren't used, the project either is never finished or never gets big enough to need skinning, and the game doesn't really NEED panels that the user can drag around.
My main issue with a static system is the fact that it is static. I can't make things larger/smaller when I need to unless I draw different size skins for everything. That and everything has to have it's own skin, so it increases content requirements. The other option is just stretching, and that looks horrible IMHO.
What's your opinion though? Are you mad when your RPG won't let you drag your inventory window around? Or your MMO won't let you resize your chat window? Or you can't use a custom skin?
As a proposed solution, I've thought about just implementing a layout system. The user can select from multiple built-in layouts (or make their own.)
Or even just allowing different states for things. For example, in Baldur's Gate you can make the game's main output window larger or smaller by clicking a button.
It's really not much of a question, seeing as almost every AAA game out there uses a static UI, but I'm still interested in your input. The only real reason I can see for implementing a full on UI system is for in-game editors, but even that could use a static UI.