Typst: Compose Papers Faster(typst.app)
typst.app
Typst: Compose Papers Faster
https://typst.app/
30 comments
Eh, that only makes sense if you believe LaTeX (markup) to have accidental complexity that typst somehow abstracts away. Having used LaTeX, I appreciate that there is a learning curve, but I think all the complexity is essential domain complexity that's not easy to get rid of. (Note that I'm only talking about the markup language, and not about the toolchain -- which I believe is ripe for massive simplification.)
Most likely something simplified will start out convenient and then accrue much of the complexity of LaTeX over years (because that is solving actual needs!) -- the only difference (pro) will be that new users will be slowly onboarded to this complexity over years and (con) have to wait a long time for this to catch up with necessary functionality. An interesting example to observe is Markdown variants thrashing around and diverging as they try to go beyond the simplest case and try to model the essential complexity of writing documents.
Most likely something simplified will start out convenient and then accrue much of the complexity of LaTeX over years (because that is solving actual needs!) -- the only difference (pro) will be that new users will be slowly onboarded to this complexity over years and (con) have to wait a long time for this to catch up with necessary functionality. An interesting example to observe is Markdown variants thrashing around and diverging as they try to go beyond the simplest case and try to model the essential complexity of writing documents.
Hey, I'm working on Typst. We try to open source Typst's libraries as soon as they are ready. The same goes for the Typst compiler: it will be open source when we go into beta.
We do not want to lock anybody into our service, so you'll always be able to download Typst for the command line for free. Only the web app will remain proprietary.
We do not want to lock anybody into our service, so you'll always be able to download Typst for the command line for free. Only the web app will remain proprietary.
That's great. I wouldn't want to rely on something proprietary (especially SaaS) for something so important.
From their about page [1]:
> We will publish Typst's compiler source code as soon as our beta phase starts. From then on, it will form the Open Core of Typst, and we will develop and maintain it in cooperation with the community
[1]: https://typst.app/about/
> We will publish Typst's compiler source code as soon as our beta phase starts. From then on, it will form the Open Core of Typst, and we will develop and maintain it in cooperation with the community
[1]: https://typst.app/about/
Seems interesting. That they’re tackling the whole pipeline feels like a good sign to me. Latex has, IMO, been ripe for replacement for a while to be left as a niche tool for whatever niche aspect of document formatting it does better than it’s replacement.
My only personal disappointment/nitpick is I didn’t see any mention of a web-native rendering. I had always hoped that what came after latex would be basically web pages that could become PDFs when necessary. Desires aside, I imagine having some story around HTML, interactive graphics and web browsers would probably be a good idea.
Otherwise, stoked to see this and best of luck to typst.
My only personal disappointment/nitpick is I didn’t see any mention of a web-native rendering. I had always hoped that what came after latex would be basically web pages that could become PDFs when necessary. Desires aside, I imagine having some story around HTML, interactive graphics and web browsers would probably be a good idea.
Otherwise, stoked to see this and best of luck to typst.
This looks super promising. LaTeX is an ancient technology with arcane error messages and syntax, and subpar processing speed. For example, groff can process documents much quicker as was features yesterday on HN (https://jstutter.netlify.app/), so I wouldn't be surprised if Typst can also outperform LaTeX. Also, collaborative document editing with WebAssembly looks promising.
This looks nice, subscribed.
I’ve recently been looking for a quick way to write short “mathy” notes in markdown (with full MathJax for math) and share will collaborators for commenting and found HackMD to be nice for this (codiMd is the open source version).
I also found Authorea which looks nice but it does not let you comment on markdown text. And there is too much clicking around to write math.
Ideally there should be KB shortcuts for everything so one doesn’t have to keep clicking on the menu just to insert math.
I’ve recently been looking for a quick way to write short “mathy” notes in markdown (with full MathJax for math) and share will collaborators for commenting and found HackMD to be nice for this (codiMd is the open source version).
I also found Authorea which looks nice but it does not let you comment on markdown text. And there is too much clicking around to write math.
Ideally there should be KB shortcuts for everything so one doesn’t have to keep clicking on the menu just to insert math.
It absolutely has to be almost entirely free to work for any academic due to the need for collaboration. Also, in academia things have a lot of inertia. My supervisor chastised me for trying to use MD + pandoc rather than LaTeX for my notes. After all, LaTeX is very good and serves everyone's needs, so it's hard to convince anyone to move away from it
> serves everyone's needs
I’ve seen enough people failing to achieve what they want with LaTeX, and people making significant compromises in presentation because they either don’t know how to achieve something or can’t be bothered figuring it out or doing it. There are a great many things that it is capable of, but doesn’t make practical. LaTeX does not serve everyone’s needs. Not by a long shot.
I’ve seen enough people failing to achieve what they want with LaTeX, and people making significant compromises in presentation because they either don’t know how to achieve something or can’t be bothered figuring it out or doing it. There are a great many things that it is capable of, but doesn’t make practical. LaTeX does not serve everyone’s needs. Not by a long shot.
Cant be bothered seems more likely than can't figure something out. LaTeX has a huge stack exchange community and templates out the wazoo. No reason to think they can't accomplish everything they need, in a realistic context. LaTeX isn't for one time use, it's an investment that returns massively.
I haven’t noticed anyone have any overwhelming issues with it for papers, articles, and theses. I’ve seen people struggle with posters and talks but that’s a good thing because talks made with LaTeX have way too much maths or text which is almost always bad
Nobody ever got fired for ...
Giving stuff to students/Universities for a low price is a tried principle, I can see them trying that out.
Giving stuff to students/Universities for a low price is a tried principle, I can see them trying that out.
It looks interesting, so I'll keep an eye on it, but currently there is not that much information available.
What kind of output formats to you target? I assume PDF, but aside from that? HTML? epub?
Will you maintain a history of a document? I assume this will be a hosted platform, and as a user, I cannot just add a document into source control?
What kind of output formats to you target? I assume PDF, but aside from that? HTML? epub?
Will you maintain a history of a document? I assume this will be a hosted platform, and as a user, I cannot just add a document into source control?
We are targeting PDF and raster graphic output at the moment. Typst is designed to allow HTML export and we do have our eyes on it. At the moment, however, we want to focus on getting PDF export right, including options for standards like PDF/A (archival) and PDF/X (professional print).
ePub is a subset of HTML so the latter would come first.
We will maintain a version history and offer the typical change tracking features, but this will likely be a paid feature for users who join after the beta.
ePub is a subset of HTML so the latter would come first.
We will maintain a version history and offer the typical change tracking features, but this will likely be a paid feature for users who join after the beta.
For getting into the academic publishing space this makes perfect sense. Good to hear about HTML being in the roadmap!
How do you envision the process of submitting to journals looking?
MSWord and Latex are normal in this space and I’m not aware of any that take PDFs for the obvious reason that journals want control of the rendering. I can only presume that this is somewhat of a fraught tech lock-in that resists new formats/languages/backends.
How do you envision the process of submitting to journals looking?
MSWord and Latex are normal in this space and I’m not aware of any that take PDFs for the obvious reason that journals want control of the rendering. I can only presume that this is somewhat of a fraught tech lock-in that resists new formats/languages/backends.
Yeah, that might be a problem that must be gradually solved through partnerships, outreach, and an eventual community. We already talked to a publishing industry rep about this.
In the meantime, many conferences just want camera-ready PDFs for their proceedings. Additionally, Typst will be useful in a university/lab/research group context and for tech reports right away!
In the meantime, many conferences just want camera-ready PDFs for their proceedings. Additionally, Typst will be useful in a university/lab/research group context and for tech reports right away!
Once issue will be that many conferences have a specific latex style file you must use, so you may not be able to escape latex
Are the PDFs accessible to screen readers? Even the Maths, and tables?
Looks interesting. I love the quality of latex output (mostly). Text and math are fine, tables are a pain and figures and diagrams are often painful.
Latex just has too many options which lead to a lot of wasted effort to get the result you want, some of which don't play nicely with each other.
Latex just has too many options which lead to a lot of wasted effort to get the result you want, some of which don't play nicely with each other.
Signed up! Really excited to try this out- I feel like I’m forever trying to not use LaTeX and eventually returning for some reason or other. I’m curious to know if you’ll support commutative diagrams (tikzcd) and/or the more popular LaTeX environments.
One of the advantages of Overleaf is that I can share a link with a collaborator and they only need a browser to participate. I assume with this, everyone will need to download the editor?
The editor is a web app! Because we build Typst in Rust, we are able to deliver it as WebAssembly and embed it into the web site.
Is there documentation to be found somewhere? (And is it written in Typst?)
How will you handle placement of figures in a desired location?
Curious to see how you will support aligned equations.
And while they haven't announced the platform itself to be open source, I'm glad they're at least already contributing some of the stuff they're making [1].
[1]: https://github.com/typst/