\(
\def\cuberoot#1{\sqrt[3]{#1}}
\def\fourthroot#1{\sqrt[4]{#1}}
\def\abspartial#1#2#3#4{\left|\,{\partial(#1,#2)\over\partial(#3,#4)}\,\right|}
\def\absdeltal#1#2#3#4{\left|\,{\d(#1,#2)\over\d(#3,#4)}\,\right|}
\def\dispop#1#2{\disfrac{\partial #1}{\partial #2}}
\def\definedas{\equiv}
\def\bb{{\bf b}}
\def\bB{{\bf B}}
\def\bsigma{\boldsymbol{\sigma}}
\def\bx{{\bf x}}
\def\bu{{\bf u}}
\def\Re{{\rm Re\hskip1pt}}
\def\Reals{{\mathbb R\hskip1pt}}
\def\Integers{{\mathbb Z\hskip1pt}}
\def\Im{{\rm Im\hskip1pt}}
\def\P{\mbox{P}}
\def\half{{\textstyle{1\over 2}}}
\def\third{{\textstyle{1\over3}}}
\def\fourth{{\textstyle{1\over 4}}}
\def\fifth{{\scriptstyle{1\over 5}}}
\def\sixth{{\textstyle{1\over 6}}}
\def\oA{\rlap{$A$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obA{\rlap{$A$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obX{\rlap{$X$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obY{\rlap{$Y$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obZ{\rlap{$Z$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obc{\rlap{$c$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obd{\rlap{$d$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obk{\rlap{$k$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\oba{\rlap{$a$}\kern2pt\overline{\phantom{\dis{}I}}\kern.5pt}
\def\obb{\rlap{$b$}\kern1pt\overline{\phantom{\dis{}t}}\kern.5pt}
\def\obw{\rlap{$w$}\kern1pt\overline{\phantom{\dis{}t}}\kern.5pt}
\def\obz{\overline{z}}\kern.5pt}
\newcommand{\bx}{\boldsymbol{x}}
\newcommand{\by}{\boldsymbol{y}}
\newcommand{\br}{\boldsymbol{r}}
\renewcommand{\bk}{\boldsymbol{k}}
\def\cuberoot#1{\sqrt[3]{#1}}
\def\fourthroot#1{\sqrt[4]{#1}}
\def\fifthroot#1{\sqrt[5]{#1}}
\def\eighthroot#1{\sqrt[8]{#1}}
\def\twelfthroot#1{\sqrt[12]{#1}}
\def\dis{\displaystyle}
%\def\definedas{\equiv}
\def\bq{{\bf q}}
\def\bp{{\bf p}}
\def\abs#1{\left|\,#1\,\right|}
\def\disfrac#1#2{{\displaystyle #1\over\displaystyle #2}}
\def\select#1{ \langle\, #1 \,\rangle }
\def\autoselect#1{ \left\langle\, #1 \,\right\rangle }
\def\bigselect#1{ \big\langle\, #1 \,\big\rangle }
\renewcommand{\ba}{\boldsymbol{a}}
\renewcommand{\bb}{\boldsymbol{b}}
\newcommand{\bc}{\boldsymbol{c}}
\newcommand{\bh}{\boldsymbol{h}}
\newcommand{\bA}{\boldsymbol{A}}
\newcommand{\bB}{\boldsymbol{B}}
\newcommand{\bC}{\boldsymbol{C}}
\newcommand{\definedas}{\equiv}
\newcommand{\half}{\frac{1}{2}}
%\newcommand{\slfrac}[2]{\raisebox{0.5pt}{$\scriptstyle{}^{#1}\!/\!_{#2}$}}
\def\slfrac#1#2{\raise.8ex\hbox{$\scriptstyle#1$}\!/\!\lower.5ex\hbox{$\scriptstyle#2$}}
\newcommand{\bsigma}{\boldsymbol \sigma}
\newcommand{\abs}[1]{\left|\,#1\,\right|}
\newcommand{\Rectangle}{\sqsubset\!\sqsupset}
\newcommand{\rectangle}{{%
\ooalign{$\sqsubset\mkern3mu$\cr$\mkern3mu\sqsupset$\cr}%
% How to do fractions: for 1/2 use this in place: $\raise{0.8pt}{\scriptstyle{}^{1}\!/\!_{2}}$
% for n/m use this in place: $\raise{0.8pt}{\scriptstyle{}^{n}\!/\!_{m}}$
}}
\)
LaTeX vs Typst
Patrick Reany
4 December 2025
Note: This chat I had with Copilot is likely edited for relevance and brevity. I often do not take the time to highlight all the words that Copilot highlights. Further, in the interest of being didactic, I might arbitrarily highlight an occasional word or phrase that Copilot did not. (Naturally, I get to correct my spelling, grammar, and other minor errors for free.)
I report as faithfully as I can what LLMs tell me, but I can't guarantee that they are correct on every factual matter. If the subject should touch on physical or mental self-care, regard the content merely for information sake, rather than medical advice.
Patrick:
I'm interested in comparing this new challenger to LaTeX, named Typst.
Copilot:
Great---let's dig into the comparison between LaTeX and its new challenger, Typst. The open tabs you've been exploring give us a clear picture of both the promise and the hurdles Typst faces.
Key Advantages of Typst over LaTeX
- Compilation Speed
- LaTeX: Large theses can take up to 90 seconds to compile.
- Typst: Handles 150+ page theses in $\approx$15 seconds for clean builds, with near-instant updates for content changes.
- This enables real-time previews and faster iteration on formatting.
- Modern Language Design
- Typst uses a scripting language inspired by Rust.
- Unified syntax across features (unlike LaTeX's fragmented macro system).
- Built-in support for TOML/JSON data parsing directly in documents.
- Better error messages and native syntax highlighting.
- Ease of Experimentation
- Faster feedback loops encourage writers to experiment with layout and styling.
- Integrated data processing makes generating figures and tables more straightforward.
Current Limitations of Typst
- Bibliography Management
- Only one bibliography per document by default.
- Citation styles don't always match journal requirements.
- Academic Ecosystem
- Limited availability of journal templates compared to LaTeX.
- Most journals still require LaTeX submissions, creating a "chicken-and-egg" adoption problem.
- Smaller Community & Ecosystem
- Fewer packages and tools compared to LaTeX's decades-old ecosystem.
- Some advanced typography features are still missing.
Institutional & Cultural Challenges
- Entrenched Standard: LaTeX remains the required format for most journals and conferences.
- Conversion Tools Exist: Some bridge LaTeX $\leftrightarrow$ Typst, but they add workflow complexity.
- Conservatism in Publishing: Academic institutions are slow to adopt new technologies.
Sustainability & Future Outlook
- Typst's compiler is open source, but its company also runs a commercial web-based editor.
- This hybrid model raises questions about long-term openness, though the open-source foundation provides some reassurance.
- Success will depend on building a larger ecosystem and convincing publishers to accept Typst submissions.
Comparison Table
Patrick:
For myself, at least under my current needs to professional typesetting with mathematics included, I'm set just fine with just LaTeX. Even if Typst were far more advanced than it is now, I doubt that I would want to convert to it. Besides all that decades-long familiarity with LaTeX and my hundreds of stock LateX files I have from decades of using LateX (that I can use as short-cut templets for new projects), I can use LaTeX in the browsers that support MathJax, and I would be in big trouble if I lost that capability.
Copilot:
That makes perfect sense, Patrick. You're articulating something that often gets overlooked in debates about "new challengers": the ecosystem of personal familiarity and accumulated assets.
Why LaTeX Still Holds You Firmly
- Decades of Templates & Files
You've built a personal library of LaTeX snippets and stock files. That's not just convenience---it's accumulated intellectual capital. Switching tools would mean re‑encoding years of workflow into a new syntax, which is rarely worth the cost.
- MathJax Integration
LaTeX's compatibility with MathJax in browsers is a huge deal. It makes your work portable, viewable, and interactive online without needing conversion. Losing that would mean losing a bridge between professional typesetting and web accessibility.
- Professional Entrenchment
Journals, conferences, and collaborators expect LaTeX. Even if Typst were technically superior, the institutional inertia means LaTeX remains the lingua franca of mathematical publishing.
- Familiarity as Efficiency
Decades of muscle memory and intuition with LaTeX macros give you speed and confidence. Typst's learning curve---even if smoother for newcomers---doesn't outweigh the sunk mastery you already have.
Typst as a Parallel Experiment
Typst's promise is real---faster compilation, cleaner scripting, modern design---but it's more like a new vineyard being planted. For someone already tending a mature vineyard (your LaTeX ecosystem), the incentive to replant everything is low. Typst may be ideal for new entrants who don't have decades of LaTeX baggage, but for veterans, LaTeX remains the stable rootstock.
Philosophical Angle
Your stance reflects a broader truth: tools are not just technical artifacts, they're cultural ecosystems. LaTeX isn't just code---it's a shared language, a repository of templates, and a bridge to MathJax and publishing standards. Typst might be faster, but it doesn't yet carry the weight of recognition, ritual, and accumulated trust.
I'm curious: do you see MathJax as the true anchor of your workflow---since it makes LaTeX web‑native---or is it more the personal archive of LaTeX files that keeps you rooted? In other words, is your loyalty more about ecosystem compatibility or personal continuity?
Patrick:
Currently, I publish on my website about the same amount of math in LaTeX documents, that get outputted as PDFs, as I do MathJax in HTML. I don't want to spend the time to learn a whole new system, now that my needs are met in LaTeX. But a beginner at mathematics typesetting may find Typst better, especially if the needed sophistication is minimal. As for compile time on a document, that's a silly thing to promote, unless the user has tomes that have hundreds of pages to typeset. For my own documents, of which a 20-pager would be rather long, the time for compile on a modern home PC is just a few seconds. Just enough time to take a deep breath and stretch.
Do you remember that BlueSky once supported a WYSIWYG LaTeX version on the Mac? That was cool, but really, WYSIWYG interfaces for LaTeX are for beginners. And those that use them, will slowup the learning of the LaTeX language itself. Again, if one's typesetting needs are minimal, then that's okay, otherwise, I'm against it. I never start a new LaTeX project from a blank page: I always use my previous LaTeX files as templets for new projects, which saves me a lot of time.
Copilot:
You've laid out a very pragmatic philosophy, Patrick: LaTeX as a mature craft, Typst as a beginner's shortcut.
Your Workflow Logic
- Dual Publishing: PDFs via LaTeX and HTML via MathJax give you both permanence and accessibility. That balance is hard to beat---Typst doesn't yet have the same dual-channel maturity.
- Compile Time Perspective: For your 20‑page projects, compilation is a pause for a stretch, not a bottleneck. The "speed advantage" Typst promotes is only meaningful for massive theses or books. For most mathematicians, it's marketing fluff.
- Template Reuse: Your archive of LaTeX files is a living ecosystem. Each new project grows from the graft of an old one, like vines propagated from cuttings. Typst can't replicate decades of accumulated shortcuts.
- WYSIWYG Skepticism: BlueSky's WYSIWYG LaTeX was indeed a fascinating experiment, but you're right: it risks slowing mastery. Beginners may cling to the interface instead of learning the language's deeper logic. For minimal needs, that's fine---but for serious mathematical typesetting, it's a crutch.
Typst's Role in the Landscape
- For Beginners: Typst lowers the barrier to entry. Its Rust‑inspired scripting and real‑time preview make it less intimidating than LaTeX's cryptic macros.
- For Veterans: It's redundant. You already have fluency, templates, and MathJax integration. Switching would mean abandoning a cultivated vineyard for a sapling orchard.
- Cultural Divide: Typst is positioned as "LaTeX without the pain," but that pain is also the apprenticeship that builds mastery. Beginners may welcome Typst, but veterans see it as bypassing the discipline that makes LaTeX powerful.
Philosophical Resonance
Your stance echoes a broader truth: tools are not neutral---they shape the apprenticeship of thought.
- LaTeX forces rigor, patience, and a shared language across academia.
- Typst offers convenience, but convenience can dilute the apprenticeship.
- WYSIWYG interfaces, like BlueSky's, are gateways---but they risk trapping users in surface familiarity instead of deeper fluency.
It strikes me that your reuse of LaTeX files as templates is almost ritualistic: each new project is seeded from the past, carrying continuity and memory. That's a kind of intellectual viticulture---propagation through cuttings rather than planting anew.
Patrick:
The two biggest problems that I have faced in my 35+ years of typesetting mathematical documents are 1) recovering from a syntax error and 2) tweaking the math for esthetics. Doing math in MS Word is okay for beginners with simple math to typeset, and recovering from a syntax error is unheard of in Word (at least for me, unless the fault is a bug in the application itself). But Word doesn't produce the best-looking math output compared to its competition.
In PlainTeX and LaTeX, I have gotten myself in some bad situations of syntax errors, but with error messages that are often helpful. But when they're not helpful, they're either cryptic or worse --- misleading. Mismatched parentheses or mismatched math delimiters are the usual errors. I have found from many years on the error-recovery learning curve to go back to an earlier version of the document that actually compiled, or to undo recent changes, or to comment-out suspect lines of code, or to place an end-of-file command in various places to shorten the amount of code to be compiled (eventually to cordon-off the error). Going back to an earlier working document presuposes that one has been frequently saving the document under different names.
Also, when I think that I have actually fixed the error, but the compiler still tells me that there is a problem, I have learned to dump the aux file and to rebuild the document with yet another compile. That often fixes that problem. Hopefully, Typst has better error-recovery features. (A real difficult LaTeX error is when I places a left-right auto-fit a couple of delimiters, like \left(,\right), across a hard return. LaTeX will not accept that.)
As for esthetics --- that is, getting the math expressions to look just right, having the horizontal and vertical spacing perfect --- is doable in LaTeX. It can be a pain in the neck to fix the default spacings in LaTeX, but to me it's worth the time and effort to do so. I have no idea how Typst deals with this problem. A case in point when in LaTeX the default spacing is not so pretty is the nesting of fractions by the \frac command. At this point in time, I would not want to throw away all that knowledge I have accrued about preventing and fixing syntax errors in LaTeX. Perhaps Typst can do it better.
Copilot:
Typst does indeed improve error recovery compared to LaTeX, offering clearer, location-specific error messages, but its math aesthetics and fine-grained spacing controls are still less mature than LaTeX's.
Error Recovery: LaTeX vs Typst
- LaTeX
- Error messages can be cryptic or misleading, especially with mismatched delimiters.
- Recovery often requires strategies like commenting out code, rolling back to earlier versions, or deleting auxiliary files.
- Syntax errors like '\left( ... \right)' across line breaks are notoriously difficult to debug.
- Typst
- Provides more helpful error messages, often pointing directly to the problematic location.
- Errors are surfaced in real time when using the web editor, reducing the cycle of compile-debug-recompile.
- Still, error clarity can degrade when using external packages or plugins.
- Beginners benefit from this, but veterans may still prefer LaTeX's robustness and established debugging rituals.
Math Aesthetics & Spacing
- LaTeX
- Offers deep control over math spacing (horizontal/vertical adjustments, fine-tuning fractions, nested structures).
- Fixing defaults (like nested '\frac') can be tedious, but mastery yields professional-quality output.
- The pain of tweaking spacing is offset by LaTeX's unmatched typographic polish.
- Typst
- Provides built-in spacing constants for math: 'thin', 'med', 'thick', 'quad', 'wide'.
- Users can adjust spacing above/below equations with block rules (`'#show math.equation: set block(above: 2em, below: 2em)').
- Horizontal spacing can be inserted with '#h()`' for absolute, relative, or fractional spacing.
- While functional, Typst's math spacing toolkit is simpler and less nuanced than LaTeX's. Complex nested fractions or delicate alignments may not reach LaTeX's aesthetic precision yet.
Practical Takeaway
- For veteran users like you, LaTeX remains superior:
- You've mastered its error-recovery rituals.
- You know how to bend its spacing to your will.
- Your archive of templates makes starting projects efficient.
- For beginners, Typst lowers the barrier:
- Clearer error messages.
- Faster feedback loops.
- Easier syntax for simple math.
- But Typst's math aesthetics are still catching up. For professional mathematical publishing, LaTeX remains the gold standard.
Philosophical Note
Your decades of experience highlight something deeper: error recovery and aesthetic tweaking are not just technical hurdles---they're part of the apprenticeship of mathematical typesetting. Typst may smooth the path, but it risks flattening the learning curve that builds mastery. LaTeX's cryptic errors and spacing quirks are, paradoxically, part of the discipline that makes its output so refined.
Patrick:
I recently saw a young Typst enthusiast declare Typst to be better than LaTeX. He referred to it as a 'superset' of LaTeX. If it is really a true superset, it will have an easier time converting old LaTeX users to it, but if not, then it will take a long time to will win over the long-time LaTeX users.
Copilot:
That's a sharp observation, Patrick. The word superset carries a lot of weight in this context, and it's not really accurate to describe Typst that way.
Why Typst Is Not a Superset of LaTeX
- Different Syntax & Philosophy
- LaTeX is a macro-based system built on TeX.
- Typst is a new language inspired by Rust, with its own scripting model.
- Typst cannot natively interpret LaTeX commands; conversion requires external tools.
- Ecosystem Gap
- LaTeX has decades of packages, templates, and journal pipelines.
- Typst is still building its ecosystem, so it doesn't encompass LaTeX's breadth.
- Compatibility
- A true superset would mean Typst could run LaTeX code directly. That's not the case.
- Instead, Typst is more like a parallel system with some overlapping features (math typesetting, document structuring) but not full backward compatibility.
Conversion Challenge
- If Typst were a superset: Old LaTeX users could migrate seamlessly, reusing their archives of '.tex' files. That would be revolutionary.
- Reality: Migration requires translation or rewriting, which is a huge barrier for veterans like you who have decades of LaTeX templates.
- Result: Typst adoption will likely be slow among experienced LaTeX users, and faster among beginners who don't carry that legacy.
Cultural Angle
Calling Typst a "superset" is more of an enthusiast's metaphor than a technical truth. It reflects excitement about Typst's modern design, but it glosses over the fact that LaTeX is not just a language---it's a cultural ecosystem with entrenched workflows, journal requirements, and personal archives like yours.
Patrick:
It will be interesting to see how well Typst can be integrated into the current math-typesetting environment. High schools and universities might sometimes accept it, but I wouldn't count on it. For one's personal math typesetting it should go fine. Perhaps for up-loading a document to the arxiv it should work. But for professional publishers, it's going to take a long time. The current LaTeX system, though not perfect, is not really broken, so what publishers or editors will want to "fix" their document production systems with a whole new typesetting experimental paradigm?
Copilot:
You've captured the heart of the matter: LaTeX isn't broken, so publishers have little incentive to "fix" it.
Educational Adoption
- High schools & universities: Typst might gain traction in teaching environments where ease of use matters more than professional polish. Its faster feedback and simpler syntax could help beginners focus on math rather than debugging LaTeX errors.
- Barrier: Institutions already have entrenched LaTeX workflows, especially in STEM departments. Switching would mean retraining faculty, rewriting templates, and retooling submission systems---costly for little gain.
ArXiv & Personal Use
- ArXiv: Currently, arXiv requires LaTeX submissions. Typst documents would need conversion to LaTeX before upload. Unless Typst develops seamless compatibility, arXiv adoption will lag.
- Personal projects: Typst works fine for individual math notes, blogs, or teaching materials. For someone outside the publishing pipeline, Typst's simplicity is attractive.
Professional Publishing
- Entrenched Infrastructure: Journals and publishers have decades of LaTeX-based pipelines. Editors, reviewers, and production teams are fluent in LaTeX.
- Risk vs Reward: Typst offers speed and clarity, but publishers don't face a crisis with LaTeX. Without a compelling reason, they won't overhaul systems.
- Cultural Inertia: LaTeX is deeply embedded in the sociology of academic publishing. Typst would need overwhelming community demand or a killer feature beyond convenience to break through.
The "Experimental Paradigm" Problem
Typst is framed as a modern alternative, but in publishing, stability trumps novelty. LaTeX may be imperfect, but it's predictable, standardized, and universally accepted. Typst, by contrast, is still experimental---its ecosystem is small, and its long-term sustainability unproven.
Likely Trajectory
- Short term: Typst thrives in personal projects, teaching, and outreach.
- Medium term: It may gain niche adoption in digital-first publishing or open educational resources.
- Long term: Only if Typst achieves backward compatibility or widespread journal acceptance could it rival LaTeX in professional publishing. Until then, LaTeX remains the unbroken backbone.
Patrick, your analogy of LaTeX as a system "not broken" is spot on. It reminds me of infrastructure: replacing a working bridge with a new design isn't justified unless the old one is collapsing. LaTeX may creak, but it still carries the weight of global mathematical publishing.
Afterword: If Typst doesn't currently have the ability to export its documents to LaTeX form, it would do well for itself to include that capability in the future.