RE: FAST and GRIS

[Karl Fast]
> Apologies for venting, but the Amazon search-in-this-book feature
> has got me a bit riled.
>
> Don't get me wrong--I love the new feature. It's ingenious. It has
> rough edges but it's damned impressive for release 1.0.
>
> What bothers me is that in less than a decade, Amazon has
> built--from scratch--a vastly superior information retrieval system
> than anything libraries have achieved. This is something we've been
> working on for decades.

Thanks for bringing that to our attention.

I would not be nearly so disheartened, Karl.

My initial reactions:

The Amazon.com "Search Inside the Book" feature provides author, title, and
price plus a keyword-in-context (KWIC) snippet. (The string searched for is
shown in the context of a line or so of the surrounding text.) You can click
to see a scanned image of the page.

Publishers submit a physical copy of the book, and only part of the book is
used. (This seems at odds with Jeff Bezo's letter to Amazon customers, in
which he cites the advantages of "find[ing] books at Amazon.com based on
_every word_ inside them ..." [Emphasis is in original.])

This is certainly an improvement over searches on titles, authors, and
keywords. But from a technology perspective, this is old news. But it's an
impressive investment -- one that only a company like Amazon could make.

Did they check for scanning errors?

Keyword-in-context display of search results is also old news. But it *is*
helpful in establishing relevance of the hit.

The search string is not highlighted on the image of the page -- an
annoyance that is characteristic of this kind of retrieval approach. (Some
implementations of this approach provide the text *plus* an image of the
scanned page. But that might be a problem for Amazon.com, especially from
the perspective of protecting intellectual property rights and inhibiting
researchers from using Amazon.com itself as an alternative to buying the
book or trekking to the library.)


[Karl Fast]
> We are getting beat at our own game on our home turf, and beaten
> badly. It's embarrassing.

LIS experts should definitely take Karl's caution to heart. Preemptive
strikes on a profession's core competences (usually unwitting) happen with
increasing frequency.

However, as a non-LIS person, I am impressed with some of the good thinking
(and solid foundations) in the LIS community. And it seems to me that the
boundary between business in general and LIS is not a high wall.

    Phil

-------------------------------------------
"I have made this letter longer than usual, only because I have
not had the time to make it shorter."
-- Blaise Pascal, Lettres Provinciales, 1657

Phil Murray -- Chief Knowledge Architect
The Knowledge Management Connection | http://www.KMconnection.com
401-247-7899





----------
Thanks for playing. To unsubscribe from this group, send an email to:
[email protected]

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/