Comments on: Quick response to Siva Vaidhyanathan

By: Matt Cutts: Gadgets, Google, and SEO

Matt Cutts: Gadgets, Google, and SEO — Sun, 02 Dec 2007 08:11:17 +0000

[...] Techdirt covered a back and forth and back regarding Google’s partnership with the University of Michigan for book scanning. If I had [...]

By: bowerbird

bowerbird — Tue, 13 Nov 2007 07:39:24 +0000

besides, it’s not as if umichigan can’t do its own o.c.r.

-bowerbird

By: bowerbird

bowerbird — Tue, 13 Nov 2007 07:32:09 +0000

it’s simple to write a program to get the coordinates.
so it’s just a waste of disk-space to store that info…

and since, in the long run, you will serve the _text_
rather than the _scans_ anyway, the coordinates are
unnecessary. worry about important stuff instead…

-bowerbird

By: Chris

Chris — Fri, 09 Nov 2007 15:12:56 +0000

I’m not Paul, but I can say with authority that we do not receive the word coordinates in the OCR from Google. So, we can’t implement the functionality.

By: Ryan Shaw

Ryan Shaw — Wed, 07 Nov 2007 04:32:10 +0000

Hi Paul,

Thanks for addressing my question about the OCR. It’s really great that you’re clearing up some of the misconceptions around Michigan’s collaboration with Google. I had one additional question about the OCR: I notice that MBooks doesn’t highlight keyword search results on the page images like Google Books and Open Library do. Is this because you don’t have the word bounding box info from the OCR, or just because you haven’t implemented it in the interface?

Cheers,
Ryan