SV: [Web4lib] One consequence of the digitizationprograms

[Anders Ericson]

Yes, why not as well be concerned about the NEW bad - or controversial -
books that are published. Why not then link from the Opac to (good)
Wikipedia articles and even good debates at web forums or in articles about
controversial books?  

Re the old scanned ones: Heaps of e.g. incorrect history and genealogy from
the 19th century are on their way to the web. Some of these books are
subject to Wikipedia articles. Others should be. Why not link to them?

My point is: Mainly the old books appear in full text on the open web,
searchable by Google. The new (bad or good) ones are in print only or within
closed e-book portals. 

Anders e


> -----Opprinnelig melding-----
> Fra: Jonathan Gorman [mailto:jtgorman at uiuc.edu]
> Sendt: 6. november 2007 14:09
> Til: Anders Ericson; web4lib at webjunction.org
> Emne: Re: [Web4lib] One consequence of the digitizationprograms
> 
> 
> 
> ---- Original message ----
> >Date: Tue, 6 Nov 2007 13:26:26 +0100
> >From: "Anders Ericson" <anders.ericson at norskbibliotekforening.no>
> >Subject: [Web4lib] One consequence of the digitization programs
> >To: <web4lib at webjunction.org>
> >
> >Libraries and others do a lot of digitization these days. But one of the
> >(unintended) consequences is an increasing amount of very easily
> available
> >texts in Google - however old and often unreliable and false information.
> >(Not unlike the new, but you get my point?)
> >
> >I'm looking for digitization efforts that include some "consumer's
> >information" on the quality of digitized documents. Like links to
> Wikipedia
> >articles or librarians' input.
> >
> 
> 
> I guess I'm a little confused here by what you mean by
> quality.  It seems that you're asking if anyone has reviewed
> or looked over the material.  That's an interesting question,
> after all, bad books have always been published.  Maybe the
> author was considered a quack.  I'm not sure of any projects
> attempts to do anything like that.  The interesting part I've
> found as I've played with these works though is that I've
> found references to the books I'm looking for in other books.
> At some point we might be able to datamine those connections,
> but it can be a tricky issue.
> 
> Of course, I might be misreading you.  There are usually
> quality metrics associated with scanning books, such as the number of
> errors per page or the average run of correctly
> converted words.  If you're asking if anyone has done sampling
> or displays estimated error rates, I don't think so.
> 
> Sorry, no answers, but good questions ;). Good luck on finding
> some information.
> 
> 
> Jon Gorman
> 
> 
> >
> >Anders Ericson,
> >Web editor, Norwegian Libr. Assoc.
> >
> >
> >
> >
> >_______________________________________________
> >Web4lib mailing list
> >Web4lib at webjunction.org
> >http://lists.webjunction.org/web4lib/