One of the first pieces I was asked to edit was a student guide for a life insurance law textbook (and you think the stuff I read now is tedious). The person who had the technology was the department secretary/AA. You'd give her the manuscript (handwritten?) and she would key it in using an OCR (optical character recognition) ball on an IBM Selectric. OCR = the kinds of numbers you see at the bottoms of checks. That output would get scanned by a data processing (later information services) department, and thus would the file be saved.

And OCR wasn't too bad back then. If there were errors going in, there would be errors coming out. Often not too many errors resulting from the scanning itself.
I've received recently a little (66-page) book from one of my publishers, thinking it was going to be an easy job.
Wrong.
The job appears to have been scanned in from what is probably an existing out-of-print book. The easiest ways to tell are missing punctuation, certain letter pairs always being wrong; probably the most common is seeing modem for modern. You can see how a scanner would get that wrong.
But from a proofreader's perspective, so much can go bad with scanned-in copy. While it seems like an easy way to produce a book (I really don't know how fast page proofs scan), to me the task is fraught with peril. Do publishers increase the size of the originals so that the scanner has better input with which to work? I don't know whether they do or not. Is my job as a proofreader to catch mistakes? Of course, but . . .
. . . but what I end up with, especially with documentation (footnotes and bibliography), is ultimately a copyediting task, because the print was usually so small on the pages to begin with. And as I've mentioned before, being paid for proofreading but essentially delivering copyediting gets my goat.
No comments:
Post a Comment