Lisa, as Donna already said, it takes no more than 90 minutes to scan even a relatively sizable book of 400 pages from front to back. Fixing a book in bad shape is an open-ended proposition which can take from just a couple of hours to. . . the sky is the limit. having said that, When performing quality checks on a book I go through the following steps. Please note that not every step is applicable to every book. 0. Do not remove blank pages beyond the front matter. Blank pages count as pages and their removal can screw up the page integrity check below. 1. Page Integrity Check: This essentially means attempting to match the page number at the top of bottom of recognized pages with the pages reported by your text processing program. This is especially easy in Kurzweil 8 or 9. a. Find in the book what should be page 1. Bear in mind that sometimes you need to guess a little because you may not find a page number for page 1. b. Assign this page to be page 1 in the document. In Kurzweil 9.0 you can do this in navigation/set user page number. Now page numbers at the top or bottom of each page should match what is reported by Kurzweil c. Go to page 20. In Kurzweil 9 you can go to a specific page from the navigation menu. d. Does the page number reported by Kurzweil match the page at the bottom/top of the page? If pages match, repeat the check on page 40, 80, 100, etc... until the end of the book. If there is a mismatch, you have either missing pages or duplicate pages. e. If Kurzweil reported you are on page 20 but the page number at the top/bottom of the pages is smaller than that, e.g. 18, you have some duplicate pages between pages 1 and 20. Go back 10 pages and check again. If pages match now the duplicates must be between pages 10 and 20. If pages still do not match the problem is between pages 1 and 10. split the difference again in the direction of the problem pages until you find the duplicates. This is called a binary search technique. Its strength is in the fact that you could find what you are looking for in very few steps. Eventually you will discover where you a double copy of two pages. Delete the first copy of the duplicate pages. f. If instead when looking at page 20, you find that the page number at the top/bottom of the page is 22, this means that you are missing two pages. Use the binary search technique outlined above to find what pages are missing. If you have a copy of the book, use scan/insert scan in Kurzweil to scan and insert the missing pages in the correct spot. If you do not have a copy of the book take a note of the missing pages. If at the end of your check you find lots of missing pages reject the book without feeling guilty: as Donna said, submitters are responsible for ensuring their books are complete, not reviewers like you. g. Do not worry: the above sounds complicated, but in reality page integrity checks run very quickly even for very long books. 2. Chapter header check This does not apply to every book, because not every book has clear chapter headers. Typically you search for the word chapter in the whole book and ensure that there are no terrible spelling mistakes in the chapter title. It is best to put the word chapter and the title of each chapter ion separate lines, e.g. lines 2 and 3 of the page. This is to avoid the automatic Bookshare header stripper from removing chapter headers. 3. Page header: In some cases you can strip all page headers from a book using the automatic header stripper in Kurzweil. You will find this utility in file/utilities/remove page headers. For best results use the option called 'carefully'. Automatic header removal tools do not ketch every header. After automatic removal I usually page through the whole book to ensure that all page headers have been removed correctly. Remove manually those headers that are still there. note that in most cases it is perfectly OK to remove page numbers after you have ensured that the book is complete without pages missing or duplicated. But for academic type books it is best to leave page numbers in the book for easier student reference. In this case you should not use automated header strippers, but remove headers manually and leave page numbers in th book. 4. Remove tab chars: In kurzweil you can search for tab characters by typing \t in the search field in edit/search. Remove tab characters manually. You will very often find there are junk characters immeditely to the left or the right of the tab. remove those manually as well, or replace them with the appropriate character if required in a misspelled word. 5. Remove / correct junk chars: Junk chars are those that have usually no business being in a book. I usually go through most special symbols generated by the keyboard. I search methodically for every occurrence of each in the book and take appropriate action: remove or replace with the correct character. 6. Check for odd combinations of digit 1, like 111 instead of "I'll": Also check for the single character word 1. In many cases this should be replaced manually with the word "I". 7. Remove/fix single alphabetic words: In most cases single chars b through z are not valid words. Search for each and take appropriate action on individual bases. Note that for 'i' you should search for lower case only. 8. Spell check: Not much to say about this step, except it may take several hours per book. Guido Guido D. Corona IBM Accessibility Center, Austin Tx. IBM Research, Phone: (512) 838-9735 Email: guidoc@xxxxxxxxxxx Visit my weekly Accessibility WebLog at: http://www-3.ibm.com/able/weblog/corona_weblog.html "Lisa Leonardi" <lml5280@xxxxxxxxxxx> Sent by: bksvol-discuss-bounce@xxxxxxxxxxxxx 09/29/2004 09:15 PM Please respond to bksvol-discuss To <bksvol-discuss@xxxxxxxxxxxxx> cc Subject [bksvol-discuss] Editing books Hi, my name is Lisa and I am a new volunteer. I've noticed that many of the books that I download for validation seem to be of fair quality. i would like to edit them to improve this. Does anyone have any tips on doing this? It seems a bit time-consuming to read the entire book to check for errors. Any suggestions?