It is useful to determine commitment [...] to the site, but past a certain point [...] it ceases to mean that you are any more committed to the site, and just that you know things.
This is somewhat true. In a first instance, reputation doesn't measure how much you do for the broad community (in a sense of cleanup, or whatever you deem necessary). Essentially, it measures how much the community likes your answers (or thinks your questions are interesting or well written, for that matter).
And hey, we're a Q/A site, so I don't think there's a problem with that.
While that is important, that is not related to being able to be a good "semi-mod"
You do have a point there. However, we somehow need to find a way to quantify how much we can trust a user, and unless there's a better alternative, we have to do it through reputation.
By reading through meta.SO, you'll find a lot of similar questions about this.
Here's a bit more discussion:
I assume you see a problem in high-rep users getting privileges – otherwise you wouldn't have written this. Now what happens if a user reaches "semi-mod" powers? As you can imagine, both of these extremes could exist:
- users who don't make use of their given privileges at all (sometimes even counter-acting the community by refusing to make use of them)
- those users who get to the rep barriers just to unlock the features they need for doing what they actually wanted to do (i.e. only closing stuff, only editing, only flagging)
The main reason why I think it's not hurting to judge this by reputation per se is: It's also not mandatory to use all of that.
- If you know you're not feeling comfortable editing other people's stuff, just don't do it.
- If you don't feel confident enough to vote to close other people's questions, you don't have to do it (but please don't flag the moderators to do it for you).
- If you don't feel like checking
/tools regularly, because it's wasting too much of your time, just don't do it.
You're not hurting anybody. If you're a high-rep user who likes to just answer and cast the occasional close vote, that's completely fine.
Why do we differentiate between high rep users (top 5% of users above 200 or so) and the top .2% (20000 rep)?
These numbers are a bit arbitrary. They're the same for different leagues of sites, such as Super User and Stack Overflow.
Example? On Stack Overflow, to find a user who didn't (theoretically) earn
/tools (10k) privileges within the last year I have to scroll to the tenth page. On Super User, well, we stop at the third row of the first page.*
10k users are somewhat needed to help moderators. I think we have a good balance on the site – so I don't see a need to change these limits.
* this is obviously not very representative in reality, since high-rep users also count in these rankings