Linking to Google Books

      Comments Off on Linking to Google Books

This must be the month of the OPAC or something. Today I worked up a quick way to link from a book in our Voyager system to Google books via the ISBN number. You can see how it works with this sample record in our OPAC:

Building An Intelligent Web

This particular record illustrates a problem I haven’t yet solved: some records have multiple 020 tags (ISBN number) and since the presence of an 020 tag triggers the inclusion of a link to Google Books, I end up with multiple links as this excerpt shows:

In this case, it’s not too bad since both links seem to go to the same information. But in other cases, you’ll have to do quite a bit of detective work to figure out why certain ISBN’s are attached to the same bibliographic record. For example, this title:

Another turn of the crank : essays / by Wendell Berry

has two different ISBN’s in the bibliographic record, so the OPAC displays two links:

Search ISBN 1887178031

Search ISBN 1887178007

The first one (1887178031) yields just what you’d expect, but the second? A title called Last of the Curlews. I’m sure you could eventually figure out the thread linking those two ISBN’s—but it’s not obvious. Ever noticed how librarians seem to enjoy doing this?

No fact is too obscure or so insignificant that it doesn’t merit reflection in one MARC tag or another—which suggests to the casual consumer of our cataloging efforts that everything is important. Of course, that conflicts with their life experience and ends up just annoying them as they try to sift through too much information. Either that or I’ve stumbled onto that one-in-a-million error in an OCLC-derived record. Hard to say…

I can tell you that searching the Last of the Curlews in our catalog yields a record with a single ISBN (and no reciprocating confusion back to Another Turn of the Crank).

But rather than speculate on the motivations of library scientists, let’s look at the code.

It’s really simple. Just add this line to one of your Voyager displayX.cfg files. In our case, it’s included in display1.cfg:

[sourcecode language=’html’]

Google Books:
HTML:020||a:Search ISBN {a}

If there is no 020 tag in the bibliographic record (e.g., a journal), then nothing displays. That’s good. If you have a record that has an 020 tag but it’s not a book (e.g., this video) then Google will give you an error message if you click the “Search ISBN” link. That’s not so great.

My brief testing revealed yet another problem with using ISBN’s to link to Google’s content: Mason’s library has this copy of Geronimo’s Story of His Life (a 1989 imprint) in our collection. There’s no ISBN in the MARC record and thus our catalog doesn’t automatically generate a link to Google Books. That’s OK as far as it goes, but it doesn’t go far enough. Search the book’s title on Google Books and you’ll find the full-text is there and available to download (1915 edition). I took a look at another area OPAC that uses a record’s title to link to Google Books but that method missed the availability of this full-text version as well. Clearly, this is an area that would benefit from further hacking…

“Doctor, It hurts when I search Google books for DVDs…”

I think I’m content to live with the video ISBN problem. I’ll assume that following those links once or twice will convince most users that searching Google Books for DVDs isn’t very productive. That’s not a great solution but it fits the scale of the problem.

I’m running a query now to count the number of bib records for videorecordings that contain an ISBN number but I have a feeling it won’t be more than a tiny fraction of a percentage of our total collection. Will update this post when I have that answer.