Greetings,
I hate to disagree with the other poster, but I do. I'm inclined to
assume that the elderly book you're describing, as well as most
elderly books that are republished after their copyrights have
expired, is being republished using scans of the book, NOT text
extracted by OCR. Rendering OCR'ed text suitable for publishing is
(I'd assume) a major undertaking unless the publisher has no
professional standards whatsoever. For them to undertake such a
project on a book with tricky symbols, graphics, etc. would be
remarkable. If they did undertake such a project, one would certainly
assume that they'd include the graphics/etc.
I also don't know why you are directing your enquiries at your
booksellers, rather than the publisher. I can't imagine how you'd
expect your bookseller to know how a book, originally published in
1906, is now being published. (I'm guessing that the book isn't
terribly popular, at least not anymore.)
BTW, have you considered searching for a scan of the book on
http://books.google.com and elsewhere? As long as you're prepared to
accept a copy of the book, why not print your own or have it printed
for you?
Cordially,
Richard Kanarek
On Thu, 9 Oct 2008 16:28:14 +0100, "Ciarán Ó Duibhín"
wrote:
>An old book which I have been looking for has recently appeared as "publish
>on demand". But I am hesitant to buy because this is a technical book, with
>lots of obscure symbols.
>
>I have asked two online bookstores whether the book I get will be reproduced
>from images of the old edition, or from a stream of letters made by OCR from
>the images. In the second case, I would have little hope of the book being
>accurate or readable.
>
>Neither bookstore answered my query. Would people here have an idea how
>this is likely to be done?
>
>Thanks,
>Ciarán Ó Duibhín
> >> Stay informed about: Publish on demand - images or letter-stream?