One of the content feeds I've subscribed to is the Resource Shelf, which describes itself as "a daily newsletter with resources of interest to information professionals, educators and journalists." A recent entry (I'm a little behind -- it's from July 14) looked at what editor Shirl Kennedy called "niche statistics" and contains a diverse array of really specialized statistical sources on the web.
The link that really caught my eye was to wedding industry research, from Library of Congress Business Reference Services. Not the first part, the second -- the LOC's Business Reference Services, which I hadn't been aware of. Clicking the link and looking around a bit got me to the main page for their Business Reference Services, and from there to their list of internet resources. Thinking it looked pretty cool (and noticing it has a section for marketing research, something I'm always on the lookout for), I started wondering how I would find it again.
I not a big fan of bookmarking. I also don't like accumulating links via Del.icio.us particularly, either, although I make use of the hive mind there to find stuff sometimes. I prefer figuring out the search that's going to get me to the site again. (Yeah, this backfires sometimes.) I generally assume that I'm going to forget the actual name of the page but remember something about its location -- in this case, I might remember it's at LOC. So in Google, the search to find it again will be ["marketing research" site:loc.gov]. Or just ["marketing research" site:.gov] because this search (and I always test the search) puts the site I want on the first page of results, and also adds some other interesting stuff.
Fun, huh? I just wish I could calculate the odds of ever needing to find the site again -- or of succeeding when I try.