Wow. Knowing that my writing requirements are spread out over a number of days and that I'm making steady progress on the different things I need to write certainly takes a load off my mind. I feel that my reading paper is moving at a good pace, and I think that I've scheduled in enough time to do well. What works for me:

1. Read source material and write all my notes down in a notebook, together with page numbers and other reference information.
2. Think up an outline. Add the outline to my schedule.
3. Go through my notes and write more detail about each point, either going chronologically or skipping around to points that fit into my flow. This step might have been easier if I had written my notes on index cards instead of a ring-bound notebook, but oh well. Anyway, repeat until all of my notes have been included or dropped.
4. Edit.

Yes, yes, perfectly obvious to everyone. =)

I also bonked myself over the head for missing another obvious thing. For my reading paper, I should start by looking up all those classic papers on keywords for classification (there should be a few!). If I could find a paper or two that talked about user interface differences that make it easier to assign keywords, that would be great. The key thing that I need to explain in my paper is why the social aspect is so very interesting, and what that enables: browsing, searching, mapping... =) What's old is new again, after all. I think the Internet's radically transformed keywords, though, making the social dimension non-negligible...