[Web4lib] Google Allows Downloads of out-of-copyright Books

[Jim Campbell]

I'd add on the plus side the simple fact that the books are available at
all.  With a very little hunting you can find digital versions of
Middlemarch elsewhere, but Google's got the complete works of George Eliot
up there.  Now why the 1895 edition is available for download and full view
and the 1900 edition is snippet view only, ......

On the down side, the error rate, the quality is sufficient only as long as
you are talking about Roman alphabet books. They seem to do OCR on
everything, but for most older German books, for Russian, Greek, Yiddish,
Arabic etc. it's useless.  Worse than useless, because they don't mention
anywhere that it's useless, and people may think no result means the text
they want to read isn't there at all.
 
- Jim Campbell
 
Digital Access Coordinator and Librarian for German
E-Mail: Campbell at Virginia.Edu | Voice: 434-924-4985
 
Digital Access Services, University of Virginia Library
http://www.lib.virginia.edu/digital/das/
 

> -----Original Message-----
> From: web4lib-bounces at webjunction.org 
> [mailto:web4lib-bounces at webjunction.org] On Behalf Of Karen Coyle
> Sent: Wednesday, August 30, 2006 1:04 PM
> To: Richard Wiggins
> Cc: Web4Lib
> Subject: Re: [Web4lib] Google Allows Downloads of 
> out-of-copyright Books
> 
> Richard Wiggins wrote:
> >
> > -- The downloadable PDF appears to be images wrapped in 
> PDF, not OCRed 
> > text.  That means I can't search it locally, and I can't copy text 
> > from it to use in a research paper.  Is this correct?
> Yep. So essentially what you have is an electronic book with 
> none of the features of an electronic book except that it is 
> stored as ones and zeroes. We know what people want in ebooks 
> (see "E-Book Functionalities: 
> What Libraries and Their Patrons Want and Expect from 
> Electronic Book Technologies" Lita Guide 10, Susan Gibbons, 
> et al.) At a minimum they need to be able to bookmark their 
> place in the book, and navigate to chapters and pages. In 
> addition people want to be able to make notes in the margins, 
> "dog-ear" pages to return to, copy passages to the clipboard, 
> search within the text, and see visually how far into the 
> book they are (scroll bar or whatever). These books have none of that.
> 
> On the plus side, you can have your computer read it aloud, 
> which is not possible with most commercial e-books since this 
> is considered to be covered under a separate "performance" 
> contract with the author.
> >
> > -- What's the error rate?  Google lists the particular document I 
> > found as being published in 1850.  The document is from 1856.
> High. Very high. As a quick and dirty keyword index to books, 
> which is what Google originally announced as their goal, the 
> quality is probably sufficient. As a book to be read, it's 
> definitely inadequate. Not this particular book perhaps, but 
> the Google corpus generally.
> 
> kc
> 
> -- 
> -----------------------------------
> Karen Coyle / Digital Library Consultant
> kcoyle at kcoyle.net http://www.kcoyle.net
> ph.: 510-540-7596
> fx.: 510-848-3913
> mo.: 510-435-8234
> ------------------------------------
> 
> 
> _______________________________________________
> Web4lib mailing list
> Web4lib at webjunction.org
> http://lists.webjunction.org/web4lib/
>