I think LaTeX set out to be a decent typesetter (in the sense if the profession) for books.
With human typesetting already becoming a rare profession LaTeX turned out to be the better typesetter for almost everyone in the 90s. Also InDesign came along and fulfilled that promise well for the other half of the market that had money but no inclination to work the WYSIWYM way. This lead to LaTeX' big success in the academic world.
I think typst can't hold a candle to any of the two when it comes to the previous flagship
discipline of setting narrow columns of fully justified and hyphenated[1] text utilizing microtypography to equalize the grey value.
I do not know what the plans for typst are, but I think it will
have a niche even if it will
never come to par with LaTeX and InDesign.
Their capabilities are a thing for old style physical books and not even for what we call books now.
Full justification is as dead as narrow columns and hyphenation. 30 years of web changed our reading habits. What we think of books now is mostly meant to be readable on a screen.
I also think scientific papers should adapt to that fact. Of course without throwing the baby out with the bathwater. Being able to share papers as self-contained files is a big plus and high quality math typesetting is a must.
Columnar and fully justified serif text on the other hand is just baggage.
If typst can be the accessible tool for scientific publication that'd be fantastic. If it gains enough legacy features to
replace LaTeX completely even better.
[1] Especially when it comes to languages with long words and complicated hyphenation rules like German.
P.S. Unironically always enjoyed TeX and LaTeX. Enjoy typst too, just not as a full (La)TeX replacement (yet).
Do you think it is sufficiently respectful of TeX/LaTeX?
As far as proponents go, I will echo the sentiments of many people who have actually used both TeX and Typst: I have been able to accomplish many things in Typst within an hour or two by writing my own Typst code, that in LaTeX I could only accomplish after several days by cargo-culting indecipherable gibberish from years-old forum posts. I freely admit Typst can't (yet) match LaTeX's long-tail package ecosystem, but it is much more pleasant to use and easier to reason about.
I posted that link here earlier last month[0] and even I think the comment was off putting because it's off topic and just a way to put something down. "The link you posted is becoming increasingly irrelevant" doesn't seem to add much to the conversation. To the extent that it does add something (ie. comparison of typst and Tex/LaTex) it could be phrased very differently. The way it is written now also invites similarly phrased criticism the other way as seen in other replies. I agree typst is much more pleasant to write. Also yes I doubt the typst developers would call LaTex irrelevant. In fact the author specifically points out ways the Tex currently outperforms typst. (Not to imply you stated otherwise.)
>in LaTeX I could only accomplish after several days by cargo-culting indecipherable gibberish from years-old forum posts
To learn basic use of LaTeX, takes an afternoon. To understand the language fully takes "effortful learning" like any other programming language.
I believe the difference is that Typst is effectively a scripting language, not much different than many popular ones like Javascript and Python. If you already known the basics of some programming languages this allows you transfer them very easily and start writing your own scripts very quickly. You also don't need to fully understand the language to do this, the basics will mostly be enough.
In comparison Latex has a very particular way of doing computations that you will have to learn from scratch, and even then it won't be as easy or intuitive. The fact that Latex relies so much on packages for many things also means that you will have to learn their details and intricacies when trying to do interoperate with them, which makes this even more complex.
I'm ignorant of Typst. But you're missing an important problem with Latex. Packages are really fragile. The most important property of a programming language is compositionality, and Latex has so little of that that I'm generally afraid of picking up packages because I've wasted so many hours trying to get them to play nice.
I still use Latex because of the output quality and the sunk cost..but we can clearly do better
You're right, people submitting for academic publications will still need to use LaTeX until those institutions change their practices.
If that group comprises the vast majority of people who might have a use for a programmatic typesetting environment, and if the use of LaTeX by academic institutions represents current, expert insight about LaTeX's continued superiority and not simply organizational inertia, then Typst is irrelevant and pointless.
Long term user of LaTeX. I did try Typst and it has is advantages main one it compiles faster. I am sticking with LaTeX and I don't find it difficult to use, write packages and classes, as I did invest the time to understand it and learn the language. Academic institutions, understand its superiority and also want to protect their archives. Maths has a long shelf-life. LaTeX also has a very good civilized community. LaTeX as it stands now has no comparative competitor.
What about my of decades worth of snippets, custom commands, templates for all the journals, my bib files, and of course my published works that I borrow pieces from? I should replace that with something nonfree that I have to learn from scratch? Howabout you do you?
Does typst support the standard Tex math notation? I understand that a lot of effort went into doing math different - and probably better - than Tex, but I'm just very used to the Tex notation.
It doesn't, which IMO was a stupid decision. TeX math notation isn't even that bad, and making Typst compatible with it would've gone a long way towards adoption. It's currently unusable for me because in order to actually use it for something I need to first find time to re-learn everything.
Strict to the extent that they actually expect latex, not just something that looks like latex. So unless typst is willing to output latex, which they are not, it will never work.
no the depth of graphics API's in latex is really something, and an area that is underdeveloped in typst. it'll take a considerable time for typst to be on par.
The compiler is FOSS under a permissive license Apache 2.0). Only the online editor, similar to Overleaf, is not Open Source. Please check the facts before hitting reply.
yes, it does and its realtime.
personally, I use typst locally, but
the online editor utilises WASM, so if I'm not mistaken, the rendering is real-time and handled by the front-end.
no, I pointed towards the joy of writing in a proper IDE.
but to your point: overleaf is a key enabler of latex and its cool to see typst offers a similar route
no going back once you experience realtime rendering of your document, and support in VS Code is stellar IMO.
[1] http://typst.app