ok, james, let's begin the process... :+) i will attempt to reintroduce the correct linebreaks tomorrow, but you can do some stuff meanwhile... *** first, here's a draft version of the editor:
you can't do any editing in it yet, but that's the point. the first thing you'll do with a page is to _look_at_it_, to see if the formatted text looks like the page-scan... if it _doesn't_, then the first type of editing you will do is change the text so that it _will_ look "like it should". (once you can edit, that is, not now; just look for now.) once the formatted text _looks_like_ the page-scan, you will do the proofing. this is the reverse of what distributed proofreaders does, because (quite frankly) they're doing it wrong... in my most humble opinion. *** so we first need to ensure the interface "works right" on whatever machine(s) you'll be using to do editing. so take a look at it, using such machine(s)... the editor shows the text on the left, scan on the right. and, for now, it's showing both my text and your text, with mine at the top because mine is formatted better, as the formatting is our specific focus at the moment. eventually, my text and yours will be merged into one. you'll adjust the scan by resizing the browser-window. start with a window the full-width of the screen, then narrow it until the entire page-scan comes into view... you want to have a window which is as _big_ as possible, so everything is easier to read, but also you wanna have everything in the window, all the text and the whole scan, and you'll want all of the text lines to remain unwrapped. here's an example of a good fit:
zenmagiclove.com/bhaga/good-fit-interface.jpg
if the lines wrap inappropriately, or are just too narrow, you must make the text smaller or bigger, respectively. you can do this by clicking the "smaller/bigger" buttons you will find at the bottom of the text. that resizes text by an increment of 2%, so it might take you a few clicks. you can also resize the text by zooming it (under "view", in most browsers, although it'll differ on some of 'em)... use whichever seems to work best for you, and gives you a view includes the entire page of text and the full scan. tell me if it works for you. and if not, save the source, and see if you can fiddle with the .css and make it fit... *** then, once you've got a good fit, give it a good test -- run through a few hundred pages... you proceed forward one page by clicking the scan... there are some pages -- like the tables and trees -- which i didn't even attempt to format, but the rest of them _should_ look fairly accurate. tell me about any which do not appear to be similar to the page-scan... because the paragraphs are indented, and headings are centered and bigger, and the font is proportional, it's easy to tell if the text is being formatted correctly. the paragraphs on both sides should even align nicely. this will make it easier to check any proofing questions, which will be the next step, once the linebreaks are ok... *** you can also start to work on the text... here's a list of the "good words" in the text, the ones which are found in the dictionary, and thus the ones which we'll use as this book's dictionary from now on.
what you'll want to do is run these through a spellcheck. they _should_ all pass -- since they passed mine! -- but you might be surprised... but then you will understand... the other thing you _can_ do with the list is to go over it. just because a word passes spellcheck does _not_ mean that it's the _correct_ word. it could be a stealth scanno. so that's what you'd be looking for, words that might be "in the dictionary", but which _don't_belong_ in this book. the example often given nowadays is the word "modem". they didn't have modems when this book was published, so that would almost certainly be a scanno for "modern". again, this check is only if you _want_ to do it. there are a lot of words in this text -- the book has 450 pages -- and it'd be a big job to look closely at every single one... plus the odds of catching a glitch are really very small... (but i found a few "possible" ones -- "para" and "tanks"; go ahead and search your text to see if they're scannos.) so, all in all, this task has a low return of benefit to cost. nonetheless, i often enjoy looking over the "good words", because it gives me a feel for the vocabulary of the book, (it's a bit similar to reviewing the index of a paper-book.) *** next up, we'll look at the _flagged_ words -- ones that were _not_ found in the dictionary -- for more action... -bowerbird