Some nerds are out there claiming old-timey typesetting tools like man, mdoc, {g,n,t}roff and such are great, actually, and we should use them for things like CVs and blogs and research papers.

I used to work in scientific publishing as a copy editor, toiling away in the LaTeX mines of academia, and have had strong opinions about typesetting and its beauty beaten into me. Like many people I’m also deeply unhappy and am ever looking for something, anything, to fill the void inside, such as a new old typesetting mechanism.

So I got some groff and dug around on the internet for a couple of hours to figure out how to get a basic Hello World document going, and then went slightly deeper into the rabbit hole to figure out how to load a macro package that let me write sane things like a document title and author and to write equations. I then re-typeset the first page of an article of mine that I wrote in LaTeX.

The results are bad. For the price of learning a weird macro language designed for the constraints of late 70s computers (other than TeX, that is, which is a weird macro language designed for the constraints of early 80s computers) and spending much of my time digging through scattered websites, each of which answers maybe 5% of my basic question of how to typeset this trivial thing, I get to enjoy a postscript document that looks — and I mean this — exactly like what I’d get from Microsoft Word and its Equation Editor. I also get to enjoy keeping manual track of section and environment numbers, and would have the pleasure of formatting my citations by hand if I had not given up way before getting to that point.

The nerds are wrong this time. This is not good or desirable, and despite being the weapon of choice for Brian Kernighan these are clumsy tools for a less civilized age.