[Web4lib] Google Print

[Sebastian Hammer]

K.G. Schneider wrote:
>>Hm... We've got a pretty good handle on producing metadata for books in
>>our
>>profession. Why should Google try to reproduce a service that does what we
>>as libraries and librarians do? Subject searching, for instance, has never
>>been a strong suit of ISI's citation indexes, or Lexis-Nexis. And
>>sometimes
>>we find that frustrating. But still, if we want metadata-based searching,
>>shouldn't we be going to the experts on metadata rather than the experts
>>on
>>link-based relevance searching?
> 
> 
> I think Jennifer and I are on the same page... (so to speak!). If a
> bibliographic item already has the metadata, why not use it in Google Print?
> Though then the argument would be that all items need good metadata... and
> Google's approach is that everything can be solved through search... which
> would mean either automated metadata (which has got to be within a leap of
> what they're doing) or something beyond metadata that enriches retrieval.

I think there is another important reason not to throw away the existing 
bibliographic metadata. As the shared collection of digitized works 
grows (and hopefully lots and lots of it *will* be shared), it is 
eventually going to outpace all but the largest physical research 
libraries. Curating this collection in a responsible fashion is not 
trivial. How do you ensure that the right works are digitized (and by 
what definition of 'right'?). How do you ensure not just that each 
edition isn't put through the scanning process more often than 
necessary, but that you digitize all of the editions, imprints, etc., of 
an interesting work that has some scholarly or archival value? These are 
all tough questions that people are just beginning to figure out how to 
answer, and to my mind, the existing bibliographic metadata repositories 
are some of the best tools we have available to help with that. That's 
why I think it's so exciting (seriously!) that RLG and others are 
providing bib records and expertise to the Open Content Alliance and 
that bibliographic records are being associated with the digitized items 
from the beginning.

--Sebastian

> 
> Karen G. Schneider
> kgs at bluehighways.com
> 
> 
> _______________________________________________
> Web4lib mailing list
> Web4lib at webjunction.org
> http://lists.webjunction.org/web4lib/
> 

-- 
Sebastian Hammer, Index Data
quinn at indexdata.com   www.indexdata.com
Ph: (603) 209-6853