[Web4lib] Re: Relevancy-ranking LCSH?

[Walter Lewis]

Walter Lewis writes:
>  > This begins to feel like the engines purportedly used by various
>  > anti-plagiarism "services".
>  > 
>  > Think of the wonderful collection of unintended consequences.  :)
Some one responded to me directly and wondered if I would be less subtle. :)

If we had a relevance ranking mechanism that was aware of the full-text 
of books, I was thinking that it might turn up a variety of instances 
where there was serious plagiarism happening in the publishing world. We 
have seen in the last year or so fiction like the "Da Vinci Code" in 
court defending itself against the plagiarism of purported "non-fiction" 
like the "Holy Blood and the Holy Grail". 

My notion of unintended consequences was that, to the extent that the 
Latent Semantic Indexing that Art referred to could discern patterns in 
sections of books, it might be used as evidence in the "real" world, as 
well as in the classroom.  Could the originality of research come down 
to an index value that measured similarity/dissimilarity to other 
published works? (not an outcome I'm advocating, BTW)

All speculation on my part.

Walter