Dr Cross, this is a very good point. The problem is that the Bookshare automated header stripper is far too radical for its own good. It removes both chapter headers and page headers completely, without even leaving a virtual indicator of real page 1 in the book. A kinder /gentler stripper can be implemented easily from a purely technical point of view, as demonstrated by the 'carefully' option in Kurzweil 1000 9.0. Even this stripper though is not perfect, because it does not have an option to leave uncorrupted page numbers intact while removing redundant portions of page headers. I have already suggested to the Bookshare staff that while page numbers should probably be removed from the raw body text, they should be inserted into page header tags, so that a Daisy reader can voice them or ignore them at will. Furthermore, there should be a mechanism for volunteers to tag manually page 1 in the book for best sequencing by the Bookshare system. If page one were taggable, the Bookshare system can then easily create page header tags without any need to preserve original page headers. Needless to say, the stripper should not remove chapter and section headers alltogether, but smarts should be implemented to create the appropriate tags around them. In the meantime, one way to get around the problem is to split chapter headers on two consecutive lines to defeat the stripper. For those books where the physical access to page numbers is important, omitting a blank line between the page number and the body text will also defeat the stripper. Guido Dante Corona IBM Accessibility Center, Austin Tx. Research Division, Phone: 512. 838. 9735. Email: guidoc@xxxxxxxxxxx Web: http://www.ibm.com/able "Kenneth A. Cross" <crossk@xxxxxxxxxxxx> Sent by: bksvol-discuss-bounce@xxxxxxxxxxxxx 01/17/2005 03:31 AM Please respond to bksvol-discuss To <bksvol-discuss@xxxxxxxxxxxxx> cc Subject [bksvol-discuss] The Broker--strengths and weaknesses Congratulations to the BookShare staff for the timely availability of THE BROKER, John Hrisham's newest book. For the person who reads for pleasure, it is great and most enjoyable. But the determination to provide the book without pagination is a real problem to some users. For example, since I was once an English teacher, I spend a fair amount of time either in book discussion groups or running such groups for young adults. Since the page numbers have been carefully removed, I can't ask the members of my groups to consult specific pages. And since there are no chapter headings, I can't use those either--I don't even know without great effort whether they occur. For the same reasons, I can't write or check out the writing of others if they have provided footnotes about the book. I can't even check the print book quickly to check possible errors in the copy. For example, I am pasting here a sentence which I think is in error. Everyone sw r allowed hard and waited for the words to escape through the heating vents. Now if I had any idea what page of the print book that was from, I could check it almost instantly, but all I know is that was on page 17 of the BookShare copy. What makes this disturbing to me is that, in the scanning process, the page numbers were there. They had to be carefully eliminated. And that careful elimination means that the book has some real limitations outside general reading. Paradoxically, the initial submissions of books are probably much more useful to someone doing work in teaching and research than are the books which have gone through an editing process. Is there not some way we could preserve information and, simultaneously, not plague the general reader?