You won't get any argument from me, or anyone else in the core team. This *is* an important issue. The problem is that the issue *isn't* entirely cosmetic. It's very easy to say that we "just" need to attach a configurable name to applications -- but in order to do this, you need to address the more fundamental issue of what an application *is*. In having that discussion, you hit a whole raft of *other* problems that are related to Django's definition of apps. That's why this patch has taken so long to come to fruition.
There's plenty of consensus about the broad strokes. The disagreement is about the little details. There's no constant activity because it's a big problem; that means we've gone through multiple maintainers over time, and the activity level rises and falls as attention is drawn onto other priorities (such as bug fixing for the 1.5 release).
I last looked at Preston's Github branch during the DjangoCon US sprints, and at that time, it was extremely close to being ready for merging -- it mostly just needed eyeballs, testing, and documentation. If you want to help out, I'd suggest grabbing that code, and trying to (a) get it up to date, and (b) testing it with your own projects, and © helping to stub out documentation.
I'd very much like to see this patch land as part of the 1.6 cycle -- App name translations aren't a big issue for me personally, but all the other related features -- such as having a reliable startup sequence, a place for application-level configuration, and a place for one-time initialisation -- *are* an issue for me, and fixing these problems are all side effects of adding an application configuration object.
Yours,
Russ Magee %-)