ACAT Improving catalogues

Posting to Autocat

On 9/9/2014 11:21 PM, Cindy Wolff wrote:

I think Scott’s examples are good ones. I don’t have a lot of confidence in IMDB, allmusic, or Google to always provide the information for free. They have shown themselves at times to be corruptible, some more often than others. I think we are the only ones who know what goes into a good catalog. Otherwise, all we might have left is what Google, press agents, and recording executives want you to see.

I sympathize with this, but it seems to be the same argument that people have used against Wikipedia–that the information in there cannot be trusted–and the current results of that argument so far is that it is not much, if any worse, than regular encyclopedias, even for medical information. See the recent article in the Guardian:
“It’s not uncommon for doctors to type unusual symptoms into Google and see what comes up – studies show between 47% and 70% of doctors and medical students use Wikipedia – and so far standards of medical care don’t seem to have plummeted.” http://www.theguardian.com/lifeandstyle/2014/jun/01/is-wikipedia-reliable-for-medical-advice

Whether I like doctors doing this or not, the article is correct: standards for medical don’t seem to have plummeted. If it appears to work for medicine, it would be logical to extend it to directors and musicians, where an error has less drastic consequences.

In any case, it seems that catalogers just copy what is on the item as to who are the director, actors, musicians, editors, etc. That is what I have always done. I have never done any extra verification, except in a case of a strange name that I have thought was perhaps a pseudonym. Maybe others do more, but I have never had that kind of time. From my experience, I have seen lots of video cataloging that just takes everything from the containers. I can’t blame anybody for it; to do as you are “supposed” to demands a lot of time. In any case, it seems to me the final product produces what “… press agents, and recording executives want you to see”. Libraries don’t really change that.

My argument is that the library catalog has never allowed searchers to find names by their roles. For hundreds of years, if searchers wanted to find materials where people worked specifically as directors vs. actors, the searchers found out pretty quickly that they couldn’t do it in a library catalog. That’s because the catalog was never designed for it. No surprises for anybody at all.

But to re-make the catalog into a tool that allows the public to search its contents to find names limited by director, actor, editor, or the huge list at http://www.loc.gov/marc/relators/relaterm.html demands a lot more work than just adding the codes to the records we make today. Otherwise, what we make gives wildly false results because searches only find what has been records made in the last couple of years.

If we don’t care about people being able to find the earlier records (and of course the materials they describe), then OK. However, that is thinking like an IT person, not thinking like a librarian.

So the librarian says that we want people to limit name searches by role, and sees that it would be changing millions upon millions of headings (i.e. since one record has multiple headings, it is a much higher number than the number of records). That is quite a project–probably the biggest one in the history of libraries. We should ask: how long would it take to get an even half-way decent result for the public (besides, what does “a half-way decent result” mean? 25% correct, 40%, 90%?); how much would it cost; where would the staff come from to do it, etc. etc. etc.

At some point before actually doing any work, I hoped someone would ask: what is the ROI (return on investment)? Is this the best place to focus our resources? Would any of this make that much difference to the public? How much? What are the trade-offs?

Of course, those questions should have been asked and answered quite some time back, but they never were. They will be asked eventually, when somebody tries to make a search to find out what are the films where John Huston was a director and finds results they know to be wildly wrong. So, the public will find out, and then there will have to be answers explaining the situation.

And this when our authority records don’t work and haven’t in decades!

Mamma mia!

-215

Share