yes, i did and i actually tried out the tool and did some looking around for Tolstoy's works. which i assumed to be outside of copyright proptection, but haven't looked into it.
anyway, i was more thinking of searching the page I'm actually looking at, in the same way we CTRL-F web pages.
agree with the picture books but that's a specific example and I dare say that's only very small minority of all published books.
again I agree that must do some OCR to privide the search function, but then why not provide this in raw html for viewing? I'm thinking about readers of poor eye-site, babelfishing etc
scanning a whole load of books is no doubt time consuming but I fail to see how this is high tech!? Surely the mass storeage requirements would be lower for text than images, not to metion browsing speed...
CHeers
Whilst I agree with what you are saying, what pains me more is the low-tech way in which Google implemented there library: i.e. digital scans.
With all the benefits that OCR would bring, Google would get much wider support for their actions beacuse as it stands they have simply ripped these books off. No variable sized fonts; multiple parse images, no text selection, no onpage searching, no multiple page printing / viewing. A community effort (Wiki) could corrcet the % of OCR errors. A missed opportunity.
IMHO, this tool will not do for books what google earth will do / has done for maps. Nothing to worry about yet as the tree-edition is still more useful.
On the flip side, this is yet another example of an **PA falling asleep at the wheel, waking up to see their demise staring them in the face, and sueing as a last resort. Where is the better alternative they are offering?
CHeers
Slashdot Mirror
yes, i did and i actually tried out the tool and did some looking around for Tolstoy's works. which i assumed to be outside of copyright proptection, but haven't looked into it. anyway, i was more thinking of searching the page I'm actually looking at, in the same way we CTRL-F web pages. agree with the picture books but that's a specific example and I dare say that's only very small minority of all published books. again I agree that must do some OCR to privide the search function, but then why not provide this in raw html for viewing? I'm thinking about readers of poor eye-site, babelfishing etc scanning a whole load of books is no doubt time consuming but I fail to see how this is high tech!? Surely the mass storeage requirements would be lower for text than images, not to metion browsing speed... CHeers
Whilst I agree with what you are saying, what pains me more is the low-tech way in which Google implemented there library: i.e. digital scans. With all the benefits that OCR would bring, Google would get much wider support for their actions beacuse as it stands they have simply ripped these books off. No variable sized fonts; multiple parse images, no text selection, no onpage searching, no multiple page printing / viewing. A community effort (Wiki) could corrcet the % of OCR errors. A missed opportunity. IMHO, this tool will not do for books what google earth will do / has done for maps. Nothing to worry about yet as the tree-edition is still more useful. On the flip side, this is yet another example of an **PA falling asleep at the wheel, waking up to see their demise staring them in the face, and sueing as a last resort. Where is the better alternative they are offering? CHeers