Here's an idea, and I hope you will forgive me for not making a formal proposal, as I don't feel one is warranted.
How about a livejournal community for people who want to learn more about livejournal? I joined /support/ mainly to learn more about how lj works, on the inside. I think it would be neat to have a community where people can go to learn things about how livejournal works that's simpler and more straightforward than trying to piece things together from following dev, support, biz, et al. It would be a place where devheads could post about upcoming projects, and suggest languages for up-and-coming web programmers to learn, if they want to help. That could open up into a debate about web languages, but if it's kept genteel, I don't think that would be a bad thing. Also, we could have a fact-a-day (week?) on blah blah about how lj processes thing x ("So and so permission is granted allowing friends to see blah post and not friends of). Am I just talkin outta my ass here? Stay with me, folks!
Instead of people posting to biz or support with their advanced Q&A ("What is S2?"), we would have a place to deal with questions that aren't exactly tech support (like what I normally end up posting in /support/, but maybe that's just me) but will help people understand the mechanics of the service better. It would also be a great place to train support personnel, as per the idea being bandied about in lj_support.
The thing is, I don't know enough to run the place, at least, not by myself. But if someone wants to start up (tech_ed, lj_101, newbies2? What to call it?), I'm more than willing to be a dilligent assistant. And yes, I reference newbies.lj because it's one possible direction to take this sucker.
BOLD, new community for training support people
A place to learn about webcoding applications from experienced webheads on a practical, hands on, real life project (lj_dev for dummies?)
Regular "fun" facts
Smarter lj users > improved (targeted? more knowledgeable? not just "can you change the thing with blah?" "no because we'd have to restructure the entire database, and we can't even autodetect that anyway") suggestions > newer, whiz-bang, optimized lj services
Someone has to spend time on it; it won't run itself, not to start with, at least.
Dark highlights by bolding and underlining like a college brochure
This "non-proposal" is longer than I'd ever imagined it would be. Sorry!