Quote:
Originally Posted by JSWolf
There is a really really good chance there are errors there. And unless you do a compare of the result to the PDF to find them, they'll remain. You can get split sentences, split paragraphs, & run on paragraphs. Oh and then there are the fi & fl errors that cause you to lose fi and fl depending on what software is used with the resulting file. It's a lot more work to make sure the PDF conversion is correct then just running it through Mobipocket Creator.
|
Those errors you described sound to me more like the kind of thing you'd get in the original OCR of scanned page to PDF. If the PDF is text-based, I wouldn't think you'd get so much of that.
I've only made a cursory scan through the Seas of Ernathe file, but as far as I can tell without comparing to book pages, it looks like a clean conversion. (And a good job on someone's part of scanning and doing the OCR.)
A more detailed look will have to wait while I do more pressing things.