Thursday, April 12, 2018

Things They Should Invent: use fruitless library catalogue searches to feed the library's acquisitions list

I heard about an upcoming book that sounds interesting, so I searched for it in the library catalogue. Unfortunately, it wasn't there yet.  But, instead of going to the trouble of putting in a request for them to acquire the book, I just figured I'd search again closer to the publication date, then wandered off to do something else.

I'm sure I'm not the only one who does this. Asking the library to acquire a specific book is A Whole Big Thing, and it may well already be ordered but not in the catalogue, or get ordered through whatever their normal channels are by the time the release date arrives.

But what if simply searching for a book and finding it isn't present in the catalogue could automatically inform the library that someone wants the book?

The library already knows what people are searching for.

The technology already exists to determine when a user arrives at a webpage and doesn't click on any of the outlinks (which, in this case, would be the books listed in the search results) and to generate a list of pages where this occurs - the free stat counter I use on my blog even has this functionality! This list could then be sorted in by frequency, to identify what multiple people are searching for but not finding.

Then the frequently-fruitless search terms would need to be compared with a list of current and forthcoming books. Does such a thing exist? I know Books In Print is a thing, I don't know if there's also a "Books Soon To Be In Print".  (Although even if there isn't, comparing frequently-fruitless searches with Books In Print could be useful in and of itself.)  I also don't know if it has some method to allow you to write your own program to search its database.  Google Books has an API, which might be a starting point (although I certainly can't rule out the possibility of there being better starting points that I haven't thought of.)

But comparing terms on List A with terms on List B is totally something a computer can do.  And once it's done, you've got a list of frequent fruitless searches that are also titles of books.  Which is most likely a list of book titles that people are searching the catalogue for but not finding.

Which seems like useful information to have when deciding which books to buy.

2 comments:

laura k said...

This is a really good idea!

Why is it a Whole Big Thing to suggest a title? I'm curious about this, because I answer emails from customers requesting titles. They email a title and author, and we try to order it or get it by inter-library loan. The email address may be hard to find, but that's because our library's website sucks. Presumbably on a good website (supposedly coming soon) it would be easier to find.

impudent strumpet said...

Apparently there's some very specific form that needs to be filled out that isn't available online? I don't remember exactly - I looked it up once, went "Ugh, that's A Whole Big Thing" and have since forgotten the details. I have suggested acquisitions through the regular feedback form, and they acquired some of them but not others so I don't know if it was actually my doing.