5 minutes read | 896 words by Ruben BerenguelSome links are affiliate links
I am on holidays (starting yesterday)! Two weeks of probably more programming than usual 🤣
This week I have started rewritting another personal project: awkrdeck, porting it from AWK to Haskell. It wisas a Frankenstein project, takes a valid presentation formatted in Deckset and generates a relatively similar presentation in reveal.js (via Pandoc). You still need Deckset, and you definitely should get Deckset if you don’t. For many reasons.
The new version is turning out pretty nice. The main reason is that the underlying process structure is different. Awkrdeck is a one-shot parser-and-rewrite engine. This means that it (mostly) reads line by line and outputs new lines according to the rewrite rules. Haskset (the new project) has properly separate parsing and rewriting steps, and using parser combinators makes the whole parsing ordeal a breeze. Expect a full write-up soon.
You can see a screenshoft from a slide from both below:
You can also get these posts as a weekly newsletter by subscribing here.
This is impressive, and I’m looking forward the two weeks of holidays I have to delve into three.js for data visualisation. It was already on my radar for generative ideas, now I have an additional project for it.
This started more meh than the others (using GitHub Actions) but it picks up the pace when you get to deploy to PyPI and TestPyPI and several other interesting things to do as part of your deployment process.
It is a classic technique in writing (have ever heard of Benjamin Frankling cutting-and-pasting articles?) and painting, but not so much in the tech spaces. For web design, UI/UX work it seems a no-brainer to improve, but I wonder how it can be applied to other areas (areas I use more). Now I think a bit more about it, some of my generative work is technically copywork, in particular Schotter, Going crazy and Crazy nights (although this one was more inspired by other’s, it has many things only I have done as far as I know).