Monday, August 08, 2005

Shooting blind--WorldCat in many guises

Info Today says that Amazon has harvested (is harvesting?) OCLC's bib data and incorporating it in their online collection. My first reaction was '@#$@What in the world are we paying OCLC for? but then I did a bit of searching and was amazed at the variety of responses I got.

I tested it by searching for William Welmers (a linguist who worked in popular languages like Kpelle and Fanti). In Google's Open WorldCat I got 9 hits. I thought that was reasonable. (I have OCLC via Google as a search engine in my firefox search box.)

Next I did the search in (their book search looks in Amazon) and got "about 170"--like the computer really can't tell exactly how many?

Then I went straight to the horse's mouth and tried the same two-word search in Amazon and got 158.

1. For Precision choose Google. All the responses were good, solid ciations and they gave information about libraries where you could find it. But they only gave a few items.

2. Comprehensive award goes to and Amazon because they came up with lots, lots more. Where possible they 'Search within the book.' So you get more recent works that cite Welmers. But of course, there were many, many were false hits like Bobby Knight's biography and a book about Danielle Steele.

3. Ranking matters though. and Amazon apparently use different ranking algorhythms and I liked the one better. More of the titles I wanted up front.

4. Finally there is robustness of the engine. When I added the word Africa to the search in I got 119. Still some false hits but it got rid of some of the junk. I tried adding Africa to the Amazon search and it just pooped out and said, nope, nothing like that here. I cut and pasted the search into the box at the top of the Amazon page and got 'about 122.'

BOTTOM LINE-- we got no idea what's going on when we search these guys. OK if you're just looking for a nice read but useless if you're a grad student trying to build a bibliography for a comprehensive exam.

Oh yeah, and the last paragraph of the article is very interesting:

"OCLC member libraries and individual members of the Users Council were contacted for this story. Some were aware of experiments and projects with Amazon; others were not. Attempts to acquire details from OCLC were limited due to a non-disclosure agreement. Attempts to contact Amazon to discuss OCLC records and their inclusion and use were equally unsuccessful."

No comments: