Amen to the need to help people narrow down, focus their searches; amen to BT/NT in LCSH.  I'm working in a smaller subject domain now than I used to, theology and religion. It makes the idea of projects like mining seminary reserve lists for recommended works, [I really wish ATLA would let us mine book reviews], or mst-cited-author lists, or other selection tools aimed at users, seem possible.  And how to combine browsing the the classification with what LCSH terms are linked there...
Cindy 

-----Original Message-----
From: Code for Libraries [mailto:[log in to unmask]] On Behalf Of Greg Lindahl
Sent: Wednesday, April 06, 2016 11:44 AM
To: [log in to unmask]
Subject: Re: [CODE4LIB] Google can give you answers, but librarians give you the right answers

On Wed, Apr 06, 2016 at 07:42:11AM -0700, Karen Coyle wrote:

> Also, without the links that fuel pagerank, the ranking is very 
> unsatisfactory - cf. Google Book searches, which are often very 
> unsatisfying -- and face it, if Google can't make it work, what are 
> the odds that we can?

Karen,

I wouldn't generalize so far for either web search or book search.
Pagerank is close to useless on the modern web thanks to webspam.
When Google first launched, its focus on anchortext was just as important as pagerank. On the books side, properties like publisher authority, book usage, and used book sales+prices make nice ranking signals. Book content also contains a lot of citations, which can be used to compute impact factors. Google Books has only scratched the surface of what's possible for book search and discovery.

-- greg

http://blog.archive.org/2016/02/09/how-will-we-explore-books-in-the-21st-century/