Google and Google Scholar (was RE: Central site for IR)

["Mark Johnson" <mjohnson@highwire.stanford.edu>]

Everything Joe wrote about Google is correct so far as I know, 
but Marty specifically mentioned Google Scholar and there is a 
difference between the two tools owned by THE VERB down in 
Mountain View.

Google Scholar actually is getting very good at searching 
professional and scientific information, and usage of it is 
definitely growing in the academic community.  The results are 
not returned using Google's link-counting PageRank method. Google 
Scholar attempts to sort articles like a researcher would: 
presumably weighing the key words in the full text, the authors, 
the source publication, and the quantity of citations in other 
publications.

Also, Google Scholar crawler data is being supplemented by 
article specific metadata.  For example, HighWire offers a 
service that supplies Google Scholar with metadata on behalf of 
our clients, greatly improving the accuracy of search results for 
those participating publications.

And the eggs are not all in Google's (or Google Scholar's) 
basket. Microsoft has recently entered the scholarly searching 
field with its Windows Live Search Academic product.  Like Google 
Scholar, Microsoft has shown great interest in building 
relationships with publishers and hosting services (including 
HighWire Press).

Cheers,
Mark

Mark Johnson
Journal Manager
HighWire Press, Stanford University
mjohnson@highwire.stanford.edu

-----Original Message-----
[mailto:owner-liblicense-l@lists.yale.edu] On Behalf Of Joseph Esposito
Sent: Tuesday, August 01, 2006 3:56 PM
To: liblicense-l@lists.yale.edu
Subject: Re: Central site for IR

No quarrel with the distributed repository idea, but let's not 
put all the search eggs in Google's basket. Google is wonderful 
for what it is wonderful at, but it is not nearly as good for 
professional and scientific information.  Google is a keyword 
search engine with an outstanding method for ranking results 
(PageRank).  PageRank analyzes links between sites. This is great 
when there ARE links, but academic papers can be both relevant 
and important without any links whatsoever.  Google can't help 
much here (though it will find the key words).

It's easy to forget that Google is all of 8 years old.  With the
amount of research going into search now, Google may have
disappeared from our memories before it turns 16.

Joe Esposito

On 7/31/06, Martin Frank <MFrank@the-aps.org> wrote:
>
> Why do we need it?  That is the argument against a central PMC
> repository.  With the search technology that exists today, a
> central repository is unnecessary.  Let Google Scholar crawl
> journal sites as it already does to provide access to the
> literature, or invite NIH to extend PubMed/Medline backward with
> links in parallel with the journal legacy projects that are being
> undertaken, so the public and scientific community can readily
> find what they are looking.  Central is unnecessary, duplicative
> of distributed journal and institutional repository sites, and a
> diversion of research dollars unnecessarily.
>
> Martin Frank, Ph.D.
> Executive Director, American Physiological Society
> email: mfrank@the-aps.org
>
> ________________________________
>
> From: owner-liblicense-l@lists.yale.edu on behalf of Richard Feinman
> Sent: Fri 7/28/2006 7:27 PM
> To: liblicense-l@lists.yale.edu
> Subject: Central site for IR
>
> Wouldn't it be good to have a central site for IR supported by
> grants or all the institutions that wanted to use it as a
> repository?
>
> Richard D. Feinman, Co-editor-in-chief
> Nutrition & Metabolism ( http://www.nutritionandmetabolism.com  /home )