[Web4lib] google & library catalogs

[Lars Aronsson]

Drew, Bill wrote:

> Its going to suck if any of us do it!  What needs to be done is to
> improve the placement of OpenWorldCat in Google searches.  OpenWorldCat
> is the closest thing we have to a free world wide union catalog.  If
> myst udents find records for books in just your library, it is of no
> benefit to any of us.  I remember when all of the book records showed up
> in gopher searches as some one mentioned earlier.  That was pretty bad.

It will be pretty bad if only one library does this, and that is 
what happened in the Gopher days.  If I search for Shakespeare's 
Hamlet and all I find is a copy at some public library in Kansas, 
it's going to be pretty weird.  Today if I search Google for the 
exact phrase "here are some pictures of my kitten" I get 18 hits 
and "one hour before the library closes" gives 215 hits.  This is 
a much richer (thicker, fatter) web than ten years ago, and search 
engines need to find the most relevant among all potential hits.  

Geographic distance might be a useful filtering criterion, if the 
search engine can compute the geographic distance between me and 
the object described in the web pages.  Then if every public 
library had their catalog indexed, the search engine could present 
the hits that are closest to me, and I wouldn't have to bother 
about the other ones.  If I don't see Kansas, it's not so bad.

I'm not really suggesting that we all do this with today's Google, 
but the principle of finding library books in the same user 
interface as the rest of the world-wide web shouldn't be alien.  
But then it would also have to be something more useful than 
today's OpenWorldCat.

OpenWorldCat, RedLightGreen or Wikipedia, if they were closer to 
an IMDb.com for books, might serve as a stepping stone in a web 
search process, where I can learn that there are many translations 
and editions of Shakespeare's Hamlet, before I refine my search.

The other day I added a paragraph to the Swedish Wikipedia's 
article on Jack London, enumerating the various Swedish 
translations of "The Call of the Wild".  This information could be 
found in the Swedish union catalog, but it was hard to dig out.  
The information is much more accessible now, in the Wikipedia 
article.  No less than 12 translations into Swedish by different 
translators have appeared between 1907 and 1999 of this novel 
alone, using two different titles, 
http://sv.wikipedia.org/wiki/Jack_London

Suppose we had this kind of information, easily navigable, for all 
authors, titles, translations, translators, editions, 
illustrators, with links to reviews and criticisms.  How were the 
12 Swedish translations of the Call of the Wild received by 
Swedish newspapers of the time?  How many copies were sold of each 
edition?  The solution for the future cannot be that every library 
catalogs (= keeps inventory of) the books in their collection, but 
a unified global project where we accumulate the sum of human 
knowledge about books.  Maybe it's OpenWorldCat, maybe not.  I can 
make contributions to Wikipedia, but WorldCat is Closed to me.


-- 
  Lars Aronsson (lars at aronsson.se)
  Aronsson Datateknik - http://aronsson.se