I have my own yacy search engines running internally (non-peered) for similar reasons. One crawls some key code documentation sites that I need for work, and another crawls a whole bunch of music blogs.

While I typically still use RSS for reading music blogs, I find having the search engine is a great way to go back and find something or discover something new! Every time I find a new blog, I just add it as an index to yacy to crawl.

I think it'd be great to see people spinning up larger instances that are highly specialized. For example, maybe a search engine that is dedicated solely to sci-fi and only crawls high quality boards, personal sites and blogs, and skips all the spammy, seo-optimized sites.