By ideas for the site, do you mean changes for the site code? And if you mean changes to the code, do you mean the change ideas, or the change implementations?

I mean features.

Issues can be reported (i.e. requests made) at the Google Code page. But resources to implement those changes are minimal, as I understand it.

My impression is that because of trivial inconveniences, these are actually huge problems. I'm not sure though. I'm an inexperience developer, and in looking at it I feel like I don't know where to start. Plus, it has the feeling of a dead project (because it is), and this is very demotivating. "Will anyone respond to the Issue I create? The pull request I issue? Is this even the right place for me to do that stuff?"

It seems to me that the primary piece missing is the effort.

Right. I should have talked about this in my post, but my thoughts are still very unrefined. Anyway, here's what I'm thinking:

• My overarching point in this post is that improving LW is a high level action that could be very very impactful. I don't know how to argue this well yet. Some outside view evidence might be that Eliezer judged it worth it to invest his time in the Sequences and in HPMOR to develop the rationality community. I think that there's still a ton of room for improvement, and that it wouldn't take that much effort.
• I think that others could be convinced of this (although I definitely don't expect this post to do it).
• I think that if others were convinced of this, they'd want to contribute. Well, enough people anyway. There seems to be a lot of people who are capable of contributing, and we'd only need a handful. (I wish I could give a good estimate on how many man-months it'd take.)
• I think that trivial inconveniences are a big reason why people don't contribute (sort of). I sense that if a reasonably organized plan was laid out, it seemed like it'd be impactful, and contributing was low friction, we'd get more than enough volunteers.