On Mon, 5 Nov 2018 00:36:13 +0200 Dr Eberhard Lisse <nos...@lisse.na> wrote:
> Steve, > > I fundamentally disagree about the relevance. > > LyX is a front end for LaTeX, not a document format. And it is a > FANTASTIC front end, which can be twsited to do a lot of things :-)-O Sure, but LaTeX isn't the only game in town the way it was 10 years ago (unless you took Docbook seriously a decade ago). Most new books are ePubs or derivative mobis or whatever. > > pandoc can produce an epub from (reasonable) LaTeX (exported from > LyX), which kindlegen can translate into mobi. Yeah. I've had hundreds of people recommend pandoc and XSLT and the like. Have *you* ever successfully used pandoc to create HTML or ePub or mobi formatted to your desires? If you have, you're one in five hundred. Everyone recommends Pandoc, but finding people who have used it is like finding a needle in a haystack, and when you find such people and ask them how to do the conversion, they point you to Internet sites with procedures that make installing Gentoo or Arch a one click process. By the way, same thing goes for Asciidoc, Asciidoctor, Markdown, and Multimarkdown. Evvverybody recommends it, but few have used it to make books in which the author declares and uses styles. > For LaTeX there is lwarp at > > https://ctan.org/pkg/lwarp > > which also looks interesting. I wish I had a dime for every hour I spent, on solutions to this problems, that "look interesting". 99% of them turn out to be converters whose first step is to convert your styles into appearance, guaranteeing garbaged up output. > > XML would be a great step, and not only for epub. But that would be a > fundamental change, and who's going to do it? And that's where the rubber meets the road. Look back to the thread, starting on 7/22/2008, subject "Progress on the MS Word to LyX conversion". In that thread, against my warnings, by the way, several top Lyx developers promised an XML native format for LyX 1.7x. Not pidgeon XML. Not almost XML. Not halfassed XML. They promised XML. With a DTD, no less. Now my position was that XML is much harder to parse with Unix core utilities, so I was against it. But at least I figured that if it went XML, I could find an XML parser to do what I had been doing. With much more difficulty. But doable. But they went only half way, harming the inherent coreutils parsability without enabling the file to be processed by an XML parser. If memory serves me, 1.6 already had some XML-ish changes to the native format, I don't remember a 1.7, and 2.0 introduced the pidgeon XML we know today. Over 7 years have elapsed since 2.0's introduction, over a decade has elapsed since it was decided to have a well formed XML native format validated with a DTD. Retina display and iOS and all this Apple compatibility is nice, but I'll repeat, it's not 2008 and PDF is no longer the only game in town, and I think priority should be placed on finishing what was started in July of 2008. SteveT
