Carrot2, a clustering search engine

Mary Ellen Bates raves about Carrot2 in her latest InfoTip newsletter. Carrot2 clusters search results, much as Clusty.com does. Carrot2 differs in that it's using a Swiss meta search engine, etools.ch, as the basis for its initial group of search results, while Clusty uses US-based meta search engines. Both the Carrot2 and Clusty home pages look like mirror images of each other, down to the various selection tabs on offer. As a test, I entered "information retrieval" as a search term in both. I didn't do a hard analysis, of course, but I found Clusty's clusters generally more scannable and valuable as a starting point for further searches, as the clusters tended to be more granular. Carrot2's fewer clusters seemed to survey the landscape at a slightly higher level; specifying different sorting algorithms (available under Show Options) was fun though--"Rough k-Means" and "HOAG-FI" shook up the clusters and yielded a more interesting display.

By the way, I'm also subscribing to Mary Ellen's Info-Entrepreneur newsletter. I'm able to visualize myself doing that kind of work soon; up to now, I've not had a real picture of where my IS degree may take me. The Info-preneur/Information Broker idea at least gives me a start at something to form ideas around. I also consider it a good omen that her initials (MEB) are the same as mine. :)

Mary Ellen also runs a blog on the side, Librarian of Fortune, where she "contributes white noise to the blogosphere." Highly recommended, as are her newsletters.

Done, done, and done

For the last month, just as I thought I was nearing the finish line or reaching a milestone where I could catch my breath, another deadline or commitment loomed, both at work and at school. I spent last weekend binge-grading grant projects submitted by other teams in my Digital Preservation and Archiving class, reading an article, drafting a critique of said article, and drafting a research proposal. The grant info was due Monday, the critique due Wednesday, the proposal due Friday. Ho ho, thought I, can I turn in the proposal on Wednesday and avoid a commute to campus on Friday?

Well, no. The grant stuff and critique got done, but the proposal was a disaster. I just finished it tonight, printed it out, and after tomorrow morning, Christmas shopping can finally begin.

But here are lessons learned on the proposal:

  • Start early. Crucial to me, since I had to junk my entire first draft and start over from scratch.
  • Get a fellow student to read your paper and critique it for you. I'd read about this idea in other blogs, but this was the first time I'd done it. She was supportive but put her finger on a key weakness that I couldn't write or think around. She also knew what he liked to see in papers and student work and provided good advice. Hence, my need to scrap it and start over.
  • Go back and read the professor's directions. The weakness she pointed out was clearly delineated in his instructions for the proposal, had I but re-read them. Be a lawyer and read the fine print.
  • Don't research forever--timebox it. The danger here is that I had left myself so little time that I barely skimmed the articles I found. No time for fancy research techniques; scan, skim, ingest. But the earlier you can do this, the more facts you can feed your brain so it can go to work in the background.
  • I started to feel panic a second time as I started over on the writing. Classic fear response. I relaxed and fell back on my ol' NaNoWriMo skills and tips: Write a vomit draft. Don't edit. Lower my standards. Think quantity, not quality. The more you write, the more you can write. Just keep your fingers flying. If you just don't know what to write, the trick here is to write about your inability to write. Describe the frustration. Describe what you want to be able to say. Lo and behold, this always seems to unjam the blockage for me. (It's all going to be deleted anyway, no one's going to see it, so go crazy.)
  • I used InstantBoss (freeware), set for the standard 10 + 2 * 5 routine. By focusing for just that 10 minutes on writing and not diverting myself with editing, I got a good two pages done my first night. Tonight, I worked about 45 minutes total to finish it.
  • The key is not to finish the paper; the key is to keep starting. Eventually, you'll reach the end.
  • I also decided that it's OK to relax and do B-level work on this proposal. My class participation and other work have been more than up to the mark. No need to torque myself into a perfectionist knot.
  • It's OK to feel like the slow kid in class. Three of my fellow students had finished their proposals early and I was disappointed that I couldn't be a member of their club. Oh well--next time.