I can understand sloppy formatting creeping into older books - that is, any title that has not been published in printed format for more than, say, 20 years.
But, surely, OCR and proofreading can't be an issue for more recent titles. For such books, the publisher must surely have a way of extracting the text for the ebook from their traditional composition systems. In other words, the text is already held electronically (complete with formatting tags), so converting it to an ebook should be a mechanical process.
|