Ever wonder why some ebooks don’t have the same pictures that the paper books have? It’s because that for older books the publisher might not have the digital rights to use the images and then a decision has to be made about whether it’s worth the time, and money, to try and get them. That’s why some ebooks have no images at all.
How about converting the backlist to ebooks? It can be expensive if the publisher no longer has a digital file for the book. It then has to be scanned, OCRed and manually proofread. Expensive.
The above is from an article in the New Zealand Herald which you can find here. Lots more interesting information.
There is no valid reason why publishers who have to scan and OCR back list titles (because no digital files exist) can’t do what Project Gutenberg and Mobilereads have done- use volunteers to proof read the books. If they do that with several pages at a time, they won’t really have to worry about the drafts bing pirated on the internet (or some other sensible prtected idea). This is an imminently logical solutiopns- AND many ereader owners will volunteer on this.
I do agree that many scanned books have terrible (read NO) proof reading. I have posted reviews of two James Clavell novels that are replete with these problems. They were NEVER proofed, and not even spell-checked which would have solved more than 95% of the problems.
So, I’m not buying the publishers laments and projected costs on this. A minimal amount of spellchecking, etc will fix most of the problems,. The rest is also handleable- IF they really want to do it. But they don’t- not unless, like PENGUIN, they are charging outrageous/ridiculous prices for backlist titles.
But,we already have the solutions- let the authors do it themselves and keep 70% of the royalties. Bye Bye publishers.
Publishers could even “pay” OCR readers with X cents of store credit for every Y pages proofed. It might even be free advertising. If I found some random page of a novel interesting, I might make a note to get the work when it’s released (presuming I know what the work is; if pages are unlabeled, I might well not).
Easiest way to proof scanned/OCR’ed pages:
upload the scan and the pc’ed page to MTurk.
A good cheap way to handle slushpiles, too.
http://en.wikipedia.org/wiki/Amazon_Mechanical_Turk
But the publishers would rather cry woe is me than to pay ten cents a page for proofing.
Uh, that is: “upload the scan and the OCR’ed page”.
Typos. Typos. Typos. 🙁