abebooks.co.uk has an advert in the current London Review of Books. It suggests that 'if you can't find it here, it doesn't exist' and reports that it has '80 million new, secondhand, rare and out-of-print books'. And here is a note from (both the UK and US) website:
With over 80 million new, used, rare and out-of-print books for sale from 13,500 booksellers, Abebooks.com has the world's largest selection. So if you can't find a book on our site, it probably doesn't exist. [Abebooks: The Not-Book Campaign]
An interesting claim, and I was curious to test it.
I picked the first 'non-mainstream' volume that was on an adjacent shelf.
Spoken English in Ireland 1600-1740: twenty-seven representative texts assembled and analysed by Alan Bliss.
Yep, impressive - they currently have two, one in Ohio no less! Moving further afield, I assembled a small pile of items:
What about Preparing documents with UNIX?
Hey! A whole bunch of them ....
Pascal: user manual and report , by Jensen and Wirth (I have a 1978 corrected printing of the Second edition)
Another whole bunch! And I appear to be able to track down the particular one I have.
What about some more library oriented stuff?
The architect of the much discussed British Library building at St Pancras, Colin St John Wilson, has published several books including The other tradition of modern architecture: the unfinished project.
Ooops .. the first failure. An ISBN search reveals nothing. An author search reveals several works by St John Wilson in multiple copies, but not this one.
Closer to home? Here is a report prepared by Ann Lennon, my wife, and David Pearson some years ago, and published by the British Library.
Nope, not there. Other works by David Pearson are available.
And finally, what about that seminal record of European union catalogue organizations Library bibliographic networks in Europe : a LIBER directory, so wildly popular that it ran to an updated second edition?
Surely some mistake? No copies available ...
I liked using abebooks. I liked seeing that there were all these books in all these places. I did wonder why they didn't try to do more clustering of results, FRBR-style work clusters or others.
Update: as I was typing in the words from the London Review of Books, I paused over 'secondhand' thinking that if this were a US publication this word would not be used. And indeed, I notice that 'used' is used in the US version on the website (quoted above). Well done abebooks. Their sensitivity does not extend to the UK website, however, where 'used' appears, not 'secondhand'.