by the way, i just thought i would reiterate an old screenshot... perhaps some of the people planting bugs in rfrank's ears will consider pointing him in the directions suggested here:
that's a big screenshot, because i've got a big screen, but the idea is that the proofer does the word-by-word scan _not_ against a web-page's textfield version of the page, but rather against an .html-realized version of the page... (if proofers wanted, you could even use the d.p. font on it.) the main benefit is that you free 'em from having to look at the markup, because that's an unnecessary distraction. they see actual rendered italics, not the markup for italics. it's also possible this way to red-flag any possible scannos, as well as capitalization and punctuation improbabilities... you can also colorize quotations, which helps locate any missing or incorrect quotemarks. like i said, i have a big screen, so i can put up 3 pages -- the textfield, the original scan, and the .html version, but for a smaller screen, you'd put up the scan and the .html. then, only if there are changes to be made will the proofer summon the textfield for editing. i also show lots of buttons on the screen. some are there to add words to the book's custom dictionary, so that you don't have to have them flagged the next time they appear. the others are just marked with numbers, to indicate things that they could be used for, such as italicizing selected text. again, proofing the .html version is easier than the textfield. -bowerbird