I agree... mostly. (That may surprise some people, but read on.)
I think it's entirely reasonable that some old requests can be deleted. Requests that are (a) trivial and (b) have resulted in tag synonym(s) being created aren't really relevant any more.
But the requests should meet both of these criteria to be candidates for deletion! Otherwise, we end up deleting valuable information that we may need later on, or removing a legitimate request which is not actually completed.
As I said below in a comment, the score of these posts has absolutely nothing to do with the value of the information the question contains. In fact, retag requests with the most information are lower-scoring. Why? People tend to pass over the complicated ones without reading. Since I've given a lot of answers in these questions, I definitely noticed that the more complicated my solution, the less votes I got on my answers, presumably because not many people want to read that stuff. (All this is totally fine, btw; I'm just saying what happens.)
<soapbox>
As far as I can tell, essentially zero thought went into this decision, and it certainly would never have gotten the community's blessing either. Given how Jeff tends to do whatever he wants and then ignores the backlash until people forget about it, maybe that was the whole point; I don't know.
</soapbox>
I'm all for site cleanup, but most of these deletions make absolutely no sense.