Academic blogging workflow

Minimising the friction of advertising my thoughts in order to maximise the chance each thought gets advertised. Not covered: maximising quality of thoughts.

This blog, and virtually all my notes, are in plain text files on my computer, published online as plain html files. It’s an informal open notebook.

I had to jump through some hoops to make this work, because I need mathematical markup support and basic citation management. Vanilla, non-academic plain text blogging is simpler. But even this academic stuff is not complicated.

It helps keeps me rigorous and coherent, knowing that the public can see what I am doing, and which half-cocked opinions I am holding. It encourages people to contact me about my ideas.

Further, having a bunch of plain text files is the most simple, convenient and reliable way of taking notes. I’d do it this way even if it weren’t going to be online.

This is being mainstreamed, to the point that blogging is part of writers’ advice books such as Nick Higham’s, Handbook of writing for the mathematical science.

Great. enough principles, let's do stuff. Practicalities follow.

There are lots of tools to render academic blogs. There are hundreds of static site generators for plain text blogging. Particular ones of note for academics are…


See Blogdown et al.


See Blogdown et al.

distill + radix

By the makers of Blogdown et al and thus cover along with it.


Wowchemy (formerly Hugo academic) is a collection of hugo extensions targeted at academics. source is oft recommended for scholarly types.

USP: can handle both rmarkdown like blogdown and jupyter. Judging by the bug tracker is brittler than vanilla hugo. However it solves many niggly themeing problems at once and has a strong community. Leslie Myint’s introduction gives us the gist of it.


The classic static site generator endorsed by github can pump out academic blogs too. FWIW I found it needed too much tweaking out of the box and got bored because I am not a ruby hacker who enjoys that kind of thing. I’m even less of a go hacker but I still found hugo less trouble because it is designed cleanly not to require any code to make a site go apart from some HTML templates and a config file, which did not seem to hold for jekyll for me. 🏗


Pelican is the python-based one that I used for this site for a long time; works well for academia especially with my custom pandoc reader. I gave it up because it was boring getting rmarkdown integration and blogdown gave me that for free.

If that is not so important to you, I would recommend it. Nafiul Islam gives some clever shortcuts on getting live preview using livereload. Michael Toth shows how to use RMarkdown+knitr to approximate blogdown. Open source.

Michael Toth’s blog documents a clever rmarkdown solution for pelican that would also work AFAICT with blogdown.


Hakyll is a Haskell-backed static site generator which boasts most-favoured-nation status with pandoc. (The less prominent yst might fit that bill also.) A highly evolved example of this in full flight is gwern’s blog.


pandoc-scholar extends pandoc with useful academic features. 🏗


Matthew Butterick’s Pollen is a dark horse.

I created Pollen so I could make my web-based books Practical Typography, Typography for Lawyers, and Beautiful Racket. Sure, go take a look. Are they better than the last digital books you encountered? Yes they are. Would you like your next digital book to work like that? If so, keep reading.

At the core of Pollen is an argument:

  1. Digital books should be the best books we’ve ever had. So far, they’re not even close.
  2. Because digital books are software, an author shouldn’t think of a book as merely data. The book is a program.
  3. The way we make digital books better than their predecessors is by exploiting this programmability.

That’s what Pollen is for.

Yes, it supports mathematical markup amongst all the other elaborate typographical.


Daan Leijen’s Madoko is an even darker horse than Pollen. It includes some nifty publish-to-the-web features, native citation support, and fancy maths rendering. However it seems quirky, solo, and undermaintained.


jupyter can render and export markdown as a flagship part of its policy of doing everything easily but nothing well. One would probably use it as a piece of a full site generator.


Hexo (javascript) (will render equations on the server)


b-ber (javascript) is an ebook system which happens to generate suave websites as a side effect. It has a “prosumer” feel, and a designerly motivation where they want to target e-readers as well.

b-ber is both a method and an application for producing publications in a variety of formats—EPUB 3, Mobi/KF8, static website, PDF, and XML file, which can be imported into InDesign for print layouts—from a single source that consists of plain-text files and other assets. b-ber also functions as a browser-based EPUB reader.


franklin.jl is a static site generator in julia, for julia purists I suppose. I probably will not use it as I am already committed to multiple languages and purism is dead to me.