Hi Don, You are write about this. But, the people over at DP want more. Which is also fine. The problem is that over at DP they seem to me to focused on output formating. What they do seem to understand that you can use markup as pseudo-code or pseudo-mark for that matter. The want to keep as much information as possible. That is not hard if you use pseudo-code. The first step would be as you said is to match the scanned image. So you have a text containing alot of code marking the original linebreaks, chapter beginnings, page marks, page numbers, bold, italics, indentation, images. This markup will not be easily human-readable, but computers do good work of rendering/display in an appropriate fashion. Then all you need is a simple tool that parser this format into the output format you want. e.g for plain text: throw-out page breaks, images, hyphenation convert footnotes to PG Style convert bold, italics, PG Style start output PG Style output PG Header output text PG Style two linebreaks for paragrahs before Chatpters, etc. wrap accordingly This an other simplification. for HTML (everything in one page) throw-out hyphenation create tags for bold, italic create tags for chapter header, with anchors create tags for paragraphs repecting indentation for verse and such. throw-out linebreaks create footnotes. with anchors create tags for images create TOCs You could also have the system produce a more complex HTML-structure, directories for chatpters, one file per page, etc. The same procedure can be applied to other output formats. That is the cool thing about pseudo-code it does not produce output if you do not want it or need it!! regards Keith. Am 21.02.2010 um 06:14 schrieb don kretz:
It's not trivial that it would make shared proofing a lot easier and less ambiguous.
Just match the image.
_______________________________________________ gutvol-d mailing list gutvol-d@lists.pglaf.org http://lists.pglaf.org/mailman/listinfo/gutvol-d