This should really link to https://sile-typesetter.org/what-is-sile/
(unless it just released a new version of course, in which case the title should probably be about that)
Sile: A Modern Rewrite of TeX
71–80 of 181 posts
Re: Sile: A Modern Rewrite of TeX
#72TeX is one of those bits of software that's so complex to replace, it'll take another few decades (and likely a few false starts) before we can get there. The fundamentals of TeX's typesetting are amazing, but everything else is just bolted on. You'd probably want to rebuild it with its own, custom language, perhaps inspired more by modern XML/markdown rather than the old TeX language. The biggest problem that we're…
Ideally there would be a replacement of TeX that shares little of its non-math typesetting capabilities but retains (or only slightly modifies) the math syntax. The rest of the system is where the the problems lie, or at least where I waste a lot of time on little things.
Re: Sile: A Modern Rewrite of TeX
#73Reminds me of reportlab, which I loathe only because of their lack of documentation. Typesetting is hard. But at this point why not use semantic html and a specially crafted css?
For computer-generated documents like invoices, that is totally the way to go.
Re: Sile: A Modern Rewrite of TeX
#74Earlier quoted context omitted.
I work with TeX quite a lot and I can't even really understand intuitively the principles of the language design let alone the grammar. Like 15 years later and I'm still copy and pasting spells. I'd love to see an CST of my document and figure out why I'm writing the characters I'm writing!
My experience with TeX is a complete opposite. I feel like I can do pretty much anything in plain TeX, with a very good feel of what it is doing in its 'guts'. I wrote parsers in compilers in pure TeX (the latter was admittedly a silly academic exercise). It took a very thorough reading of 'The TeXbook' but after that I am constantly amazed at how deliberate and tasteful Knuth's choices were. TeX is simply ... elegan…
I suspect that the vast majority of people who say they've used TeX have actually used LaTeX - and don't realize just how massively powerful TeX itself is. Much of the LaTeX macros are quite simple; TeX does the heavy lifting for so much of it.
Re: Sile: A Modern Rewrite of TeX
#75Something I've wondered, and the Tex users here can probably answer - why not just use Acrobat or Word for math? Does it not support something?
Microsoft recently improved the math notation support in Word. You can press [alt]+[=] on Windows or [control]+[=] on OSX to enter "Math Mode" and begin typing mathematical characters. Word tries to guess what you mean as you type, and it's still much more cumbersome that typing plain-text code in (La)TeX; you still have to reach for a mouse for some notation.
Here's what it looks like if you're interested:
Re: Sile: A Modern Rewrite of TeX
#76Earlier quoted context omitted.
For enterprise environments, where demands usually are low to non-existent, Word does fine. In academic environments, when journals are involved, when a document is being worked on by several people simultaneously (in varying meanings of that word) and when data may come from several different outside sources, it is massively inadequate.
I think there is demand of these features for Word users in the enterprise. Version tracking or reusing templates is huge productivity boost and also helps reducing mistakes. It is only that enterprise users have never thought of that because they don't know it is even possible.
Word with integrated git-like history would be a godsend if someone could develop it. As it is, it's easier to either cripple along with track changes and friends or teach everyone how to use Markdown or LaTeX.
Re: Sile: A Modern Rewrite of TeX
#77Something I've wondered, and the Tex users here can probably answer - why not just use Acrobat or Word for math? Does it not support something?
LaTeX is an art form. The output is perfect, based only on the code used to make it. Typesetting mathematical expressions or even creating complex diagrams using a library such as `tkz-euclide` is just code. There is no WYSIWYG. No invisible bold spaces[1]. No forgetting which changes you made to try to get it to look right. Everything is right there in the code. (I am aware of the difference between LaTeX and TeX; I…
One of the most important things I've found for making LaTeX documents is do not worry about what it looks like until you're almost done. Otherwise you fall into the rabbit hole of adjusting things for hours to find your work is wasted anyway because you changed that page later.
But once the text is done, it's really nice to work with, and you never "accidentally break the whole document by adding a table" - at least not irrecoverably.
Re: Sile: A Modern Rewrite of TeX
#78Even if LaTeX3 brought some nice things to the TeX world, TeX programming is every year lagging more and more behind modern programming languages and practices. SILE has a long way to go before it can compete with math-heavy publications made with TeX, but I think it is already a good solution for typesetting technical documentation or prose fiction.
For prose fiction, I don't see much here that is going to drive people away from Indesign. Designers and typesetters aren't really interested in doing that kind of work in a system that would require generating new output for every correction. Even relatively simple tasks like balancing columns in a spread would become a pain.
Re: Sile: A Modern Rewrite of TeX
#79I'm excited for the project, but disappointed that they kept the arcane syntax of TeX/LaTeX. Math expressions are hideous to write and read in LaTeX syntax.
Re: Sile: A Modern Rewrite of TeX
#80Does it come with a context-free syntax? If not, that would be a grave mistake. Being unparseable is one of the biggest problems of TeX, IMO. Not the least because you cannot automatically work with TeX "documents".
It's damning that there no truly functional grammar or spellchecker for latex, even ignoring crazy things that can be done with macros. If some billionaire wanted to 'move the needle' and change society for the better I think building a truly better latex would be worth it. Each year some of the most brilliant people in the world sacrifice on the altar of latex, that could instead be spent on more productive research…
I'd like to know what a "better LaTeX" means, and how that would move the needle, given that the only things you can't easily do in modern (Xe)LaTeX are things that you shouldn't be using TeX itself for in the first place, but have other tools do so you can simply embed their output. Especially given the availability of WYSIWYG-esque editors with "one-click" compile buttons that run all the nonsense that we used to have to do by hand back in the dark days of "you want a GUI? Get out of my office, you simp" paired with the modern "just install missing packages without asking me, that's the whole point: I do the writing, you make that frictionless."