Nova Googeliana

[Klaus Graf <graf _at__ uni-koblenz.de>]

Aus Peter Subers FOS-Newsletter:

Advanced searching with Google variations

Google is the second-most used search engine for chemistry research by
professional chemists, after ChemWeb (see FOSN for 2/25/02,
4/22/02).  Google's sorting algorithm and index size make it useful for
serious scientific research even in direct competition with searchable
databases dedicated to scientific content.  
http://www.freepint.com/issues/040402.htm#feature
http://www.managinginformation.com/news/content_show_full.php?id=371

One of Google's smarter moves recently was to publish its API so that
programmers could build Google searching into their own programs. 
With a few lines of code programmers can incorporate all the power of
Google into a program, and then with a few more lines tweak and vary
this power to suit their needs or visions.  Some search innovations
wouldn't work on their own but would very well when added to the Google
feature set.  Some innovations would work very well on their own but
would work even better when applied to Google's index of more then
two billion continuously refreshed web pages.

Google's decision to open its API will trigger an explosion of
creativity in search technology.  If you have a special searching need
not met by
existing search engines, it's likely that someone's Google-variant will
soon meet your need.  If not, you can take a whack at doing it yourself. 
Here are some of the Google-variants now online.  

Google email interface, from CapeClear.
http://capescience.capeclear.com/google/
(Send an email to <google [at] capeclear.com> with the search string in
the subject line.  CapeClear software will send you back an email of
the top 10 results.)

[Das klappt!]

Google API Proximity Search (GAPS), from Staggernation
http://www.staggernation.com/cgi-bin/gaps.cgi
(Find keywords within 1, 2 or 3 words of one another.)

Google API Relation Browsing Outliner (GARBO)
http://www.staggernation.com/garbo/
(Enter a URL, get a collapsible outline either of related pages or of
pages linking to the URL.)

Google Web Search by Host (GWASH)
http://www.staggernation.com/gawsh/
(Organizes results in a collapsing outline by host.  Within each host it
seems to sort by Google's page rank.)

Home grown Google variants cannot be commercial, and cannot query the
index more than 1,000 times a day.   Since Google is willing to
terminate service to entire domains when a user from the domain sends
automated queries to the index, this suggests that Google will give
developers using the API a privilege that it doesn't give to other
users.  If you agree that processing FOS as data can provide services
above
and beyond FOS itself (FOSN for 4/8/02), investigate what the Google API
makes possible.
http://www.google.com/apis/api_faq.html#gen13
http://www.google.com/apis/api_faq.html#gen7
http://news.com.com/2100-1023-883558.html?tag=cd_mh

Google's instructions for downloading and using its API
http://www.google.com/apis/

There will be endless Google variations as the word spreads, and I don't
plan to cover them all.  After this list, which should stimulate your
imagination, I'll only cover new variations especially helpful to
serious scholarly research.

I haven't seen a page collecting links to Google variations.  If you
have, let me know and I'll link to it here.  Meantime, try one of these
links to
find new variations.

ResearchBuzz by Tara Calashain
http://www.researchbuzz.com/index.shtml
(Tara tracks Google variations.  I learned about the three Staggernation
variations above from ResearchBuzz.)

SearchDay by Chris Sherman
http://searchenginewatch.com/searchday/
(Chris also tracks Google variations as they appear.)

Google search for "google api"
http://www.google.com/search?sourceid=navclient&q=google+api

* Postscript.  This week AOL dropped Overture and adopted Google as its
default search engine.  Overture invented the pay-for-rank
business model for search engines, which assumes that users are more
interested in shopping than research.  Overture is the leading search
engine with the model, and Google is the leading search engine that has
refused to adopt it.  From this point of view, the AOL decision is a
victory for objective searching over the commercial rigging of search
results.
http://www.nytimes.com/2002/05/02/technology/ebusiness/02GOOG.html