29

Disclaimer: typically soft question.

I was reading about expl3 and LaTeX3 development and try to stumble something about overall future of LaTeX. All I can find is one article.

The article mostly covers the output: fonts, encoding, hyphens, positioning; but I found no information of future of LaTeX internals: interaction with other languages (luatex, different python packages, knitr as a preprocessor), namespaces, more robust syntax, making TeX more functional with lazy evaluation and purity (I found TeX very similar to Haskell ideologically).

Is there are any overview of this?

m0nhawk
  • 9,664
  • 3
    As often with such questions, I have a feeling you are conflating different areas. For example, do you want to talk about changes to the TeX binaries (or rather the TeX-derived binaries: TeX itself is frozen), alternative approaches to TeX (which would be needed to change the syntax), developments in the LaTeX kernel, the directions in package development, etc. – Joseph Wright Aug 07 '14 at 09:41
  • @JosephWright then... It's more like a TeX derived language with a backward compatibility: with new syntax and features, but can compile old documents. – m0nhawk Aug 07 '14 at 09:54
  • 4
    The problem there is that you see either derivatives of TeX (LuaTeX most obviously) that add programming ideas to TeX but retain the core macro expansion stuff, or ideas inspired by TeX (see http://tex.stackexchange.com/questions/120271/alternatives-to-latex where some of the answers don't use TeX at all) which then mean you can forgot back-compatibility. To date, it's questionable if anyone has found an approach as flexible as TeX but with more 'purity' (there are lots of solutions to parts of the typesetting problem, but all that I know make some compromise on flexibility). – Joseph Wright Aug 07 '14 at 09:58
  • 1
    related: I asked a similar question before on comp.text.tex about one year ago and there was a long discussion on it. Google "So, what will replace latex in the year 2020?" – Nasser Aug 07 '14 at 10:04
  • @JosephWright "but its programming language is based on 1978s idea of what computers are limited by and what a macro language should look like" I like that; I'm myself agreed with Knuth to make TeX frozen, so people can build their own software with a backward compatibility with a total subset of a frozen TeX language. Definitely the replacement of TeX should be fully support compiling plain TeX documents (and, maybe, LaTeX support). – m0nhawk Aug 07 '14 at 10:22
  • @m0nhawk Any replacement which can support plain TeX can support LaTeX2e: the current release does not use in the kernel itself any primitives other than those Knuth introduced. e-TeX would I think be sensible, and with with LaTeX3 hat on \pdfstrcmp as well. However, such a system will still look much like TeX! – Joseph Wright Aug 07 '14 at 10:39
  • 15
    a presentation at tug 2014 (just last week, in portland, oregon) shows promise in the area of a backward compatible "successor" to the tex engine; see the slides: "Liberate TEX: Progress on Building a New TEX-Language Interpreter", by doug mckenna. not yet ready for prime time, or even (probably) for third-party testing; i hope some more detail on that aspect will be included in the article for tugboat. – barbara beeton Aug 07 '14 at 11:01
  • 1
    To add to Barbara's comment, there will be online videos of the talks, including Doug's, and including a panel at the end called something like The Future of TeX. FWIW, my $0.02 is that the next challenge is to improve video and audio, and executabilty and interactibility, such as you see with JavaScript in a browser. Doug's system adds a capability for at least some of that, as I understand it. – Jim Hefferon Aug 07 '14 at 12:36
  • 1
    @barbarabeeton: I just looked at the presentation, which seems interesting and ambitious. But to be honest, slide 7 scares me: "About 100,000 lines of code, half of it comment(ary)", and "Currently implemented as one C file, two header files". – yori Oct 02 '14 at 23:57

1 Answers1

4

Here's one answer, by Deyan Ginev: LaTeX is Dead (long live LaTeX): Typesetting in the Digital Age. He focuses quite a bit on output, but touches on content and process as well.

The article is predicated on this two-part claim.

The web-first scientific manuscripts of 2015 are HTML5 documents. LaTeX is one of several viable, yet imperfect, authoring languages for the web.

His "nutshell summary":

Structural document formats, such as HTML, generalize over and may eventually supersede print-oriented formats, such as PDF.

His workflow for producing the linked document:

This blog post is cross-hosted on both Authorea and a static web site. For the static hosting, I currently write content in LaTeX and have a smart bit of Ruby create a PDF and an HTML5 blog post out of my “text program”, using XeLaTeX and LaTeXML. I then upload the bundle to GitHub and deploy to my Ruby on Rails site on Digital Ocean. This sounds like, and in fact is, an exercise in programming and juggling technology stacks.

My own opinion, as a LaTeXML user and also based on looking at some recent discussions on tex.sxe, like this one: How to embed Javascript in LaTeX Beamer? is that there are certainly quite a few points of convergence.

Joe Corneli
  • 4,340
  • 1
    Sounds a bit like mouse wheel will kill emacs type of forecast (for my taste) which would make me happier actually had it been so :) Can you also include the other comments so that we are done with this once and for all? – percusse Mar 30 '15 at 00:56
  • percusse, for me it works the other way, I have to disable mouse wheel effects in Emacs to make it work the way I want. Is your suggestion to include the other boxed bits of text from Deyan's post? I'll add one more... – Joe Corneli Mar 30 '15 at 01:02
  • Ah no sorry. I meant barbara's and others' comments under the question. – percusse Mar 30 '15 at 01:08
  • I hope that they will add them as multiple different answers - I'm not an expert on that stuff. However I did read the JSBox slides, pretty impressive. Cheers. – Joe Corneli Mar 30 '15 at 01:18