BBgee... somehow i got the impression that -- under the rubric of "crowdsourcing" -- you were _soliciting_ "improved" versions of p.g. books, versions that were "fixed" and "tweaked" such that they gave people "better" experiences, either because they had had their typos and errors "corrected", or because they were "more friendly" with some hardware/software, or a combination of both those factors. i _also_ thought that you proposed to "host" these "new" versions, _and_ that the "best" one -- out of many that were "submitted", thanks to dedicated efforts by volunteers -- might get "folded back into" the p.g. library....
What I heard Greg say was indeed he was going allow these various different "improved" versions to be posted targeting various platforms, and that indeed he said that WW'ers would have the option in the future of folding back *parts* of the new effort into the existing PG source files when they felt doing so would make a contribution. I don't think he ever said anything about totally replacing current versions with totally new versions in totally new source languages, and I certainly didn't hear him suggest that the current source file formats of the current books were going to be totally replaced with new source file formats. And then what I heard next was Marcello saying "H*ck No, I'm Not Going To Allow That!" And then again like always Marcello used the opportunity to push his own agenda (in typical SU fashion.) And I still don't see where people who want to post these "improved" versions targeting specific platforms can post their efforts where PG customers can find them, so, I guess, in practice, Marcello holds the keys to the fortress. TEI is too geeky, and creates a source file which is dead from the minute it is written, because the only person who cares to maintain it is the original author. And the HTML it generates stinks to high heaven. It makes Microsoft Word output look like the immaculate conception. And TEI "defeats" the third party distributors, which one can see by downloading the "TEI" books from one of these 3rd party distributors. The end result, viewed by that indirect customer of PG, is ugly and ill-formatted and loses a lot of what was there in the original. So TEI, in practice *does not* do a superior job of preserving what is there for future generation. It provides an inferior job of preserving what is there, at a cost of a much higher workload to the PP. RST is too simple, relies too much on tweaky escape codes, generates "Python User Manual" output page formats, but does generate reasonable "living" HTML for output which other people can then take as "living" HTML and maintain and build on. There are some people who minds work this RST way, and who don't mind memorizing little tweaky "printer escape codes" and RST is going to work for them, especially for things like Junior Readers. And there are other people whose minds don't work that way and really really don't want to have to memorize tweaky "printer escape codes" and further who don't want have to give up the superior tools that are out there to write and display [X]HTML code, and the quick turnaround development cycle that allows, rather than having to stop and spend a minute running it all through a python dev chain every time they change a comma. Not to mention the pain in the hindee to set up that python chain and try to maintain it. Or the need to keep sending your code to the "SU in the Sky" to see what has happened every time you've changed a comma.