paul flo said:

Please tell me you didn't infer all that from the discussion that ensued from Jim's thoughts on fixing paragraph separation on the Kindle?

heavens no. i didn't infer _any_ of that, from _anything_ jim has said in the past month or so, when he went into my kill-file... i have an 8-year history on this listserve. so i have _plenty_ of backlog to "infer" things from, paul. in fact, this recent engagement made me realize something. i originally started battling the xhtml people when they were feeling heady, back in 2003-2005. the world was their oyster. back then, almost everyone said xml/xhtml was "the future"... but in the ensuing years, everything ground to a slow crawl, until finally, people realized xml/xhtml had very little future, and only then as a format for machines to swap information... meanwhile, i stayed tenacious in my stance for light-markup. i started doling out more and more examples, until all of my antagonists came to the conclusion they were being defeated. so the arguments became less heated, and then less and less, until finally we were going months on end without having one. but lately, there's been this "rear-guard" action. i know that it's just the dying gasp of some hardcore xhtml enthusiasts. the whole thing makes me feel nostalgic in a way, though... we've gone through the entire cycle. i started as underdog, climbing the mountain, as xhtml slid down the other side... it's like the gandhi quote that ends with "and then you win".

Along with a straw man, apparently.

actually, my request for elucidation of your workflow is an honest and sincere one. i really wanna know how you do it. because what i keep hearing is "i have this xhtml file," and i think people would like to know where that _came_ from.

For me the tools are Vim, Perl, xsltproc, Calibre, kindlegen, git, xmllint and elbow grease.

i don't see how any of those tools help you code your xhtml. did you use abbyy? did you have it save its output as xhtml? do you type in your angle-brackets and xhtml tags manually?

Thank goodness I'm not claiming to have produced a magic button that can produce nineteen formats before breakfast from an input language used by no one.

paul, really. you're making the exact type of mistake that my early adversaries on this listserve made, using terms like "claiming" and so on. have you actually went and _run_ the python script that i'm writing, which does those conversions? have you compared the input file and the output files to see just exactly how the script works. were you able to grok it? was there anything difficult about the reverse-engineering? because if you have some questions, i _can_ answer them... last year i did this demo in perl. wanna see those scripts? (no, of course you can't have them. but you can _run_ em.) and for goodness sake, paul, i understand fully well that i am not showing you all of my cards. i am doing that on purpose. but if you wanna convince yourself that light markup _works_, go look at markdown. look at restructured text. or asciidoc. there are lots of light-markup formats that are all worked out, that have been proving their worth for year after year already, in heavy-duty environments. restructured text has been the format of choice for the python community's documentation since, i dunno, something like 2004... when you pooh-pooh something that already exists, it just makes you look stupid... do just a little bit of homework to spare yourself from that, ok? -bowerbird