Awesome, thanks so much!

Bouquin looks like a very comprehensive tool, I'll give it a deeper look!

And I will definitely work on those improvements, they seem like must-haves w/o compromizing the minimalist approach I took.

Thanks again and all the best!

Thank you! Great idea about revisions, will definitely add this!

Regarding AI, I was thinking about this as well. Perhaps for feedback/reflection on your notes, or to summarize longer periods and look for behaviorial patterns over weeks/months/years. What would you implement an agent mode for?

So Zork was written in Lisp? It had to be!

---

<ROUTINE V-ADVENT ()

  <TELL "A hollow voice says \"Fool.\"" CR>>

MDL, actually, which was derived from LISP.

https://en.wikipedia.org/wiki/MDL_(programming_language)

I’m curious why they chose MDL rather than Lisp for it. Sure, it would have been ancient MACLISP or whatever, but why not leverage what was already in wide use at MIT at the time?

MDL is what was in wide use at MIT at the time, the PDP-10 era. The M in MDL is sometimes "MIT" in the various backronyms of what it stood for. (Mostly it was apparently just short for "muddle", a self-deprecating description.)

(Also, to be technically correct, these source files aren't even MDL, they are a further descendant called ZIL [Zork Implementation Language].)

Yea, I get that MDL came from MIT, but I have to think that MACLISP was far more used at MIT at that time. But maybe not.

It sounds like from what I've read MACLISP and MDL were side-by-side for a while at MIT and something of a department choice. MACLISP sounds like the serious effort and I read MDL as the "hip" or maybe rebellious upstart with a weirder sense of humor (it was called Muddle and spelled MDL to make it seem like an appropriately serious acronym), which would also make some sense that Zork originated in that allowed to be sillier language.

(Also, in reading other comments around here, I've learned there's a deeper connection in MDL to Scheme than I knew before, so I hadn't realized the Lisp/Scheme split even has ties to this "competition" of Lisp languages at MIT.)

Because Zork was written on the MIT Dynamic Modeling PDP-10. MDL was an important part of the software ecosystem on that computer, but Lisp wasn't. On the other MIT PDP-10 computers, Maclisp reined.

Was there any particular reason they did that, or was it just a random coincidence (that was the team that wrote it and the hardware they had access to was that particular machine and that particular machine ran MDL, otherwise, it would have been MACLISP)? Was there anything about MDL that helped with writing an adventure game?

MDL is also from MIT and supposedly stood for More Datatypes than Lisp. According to wikipedia "MDL provides several enhancements to classic Lisp. It supports several built-in data types, including lists, strings and arrays, and user-defined data types. It offers multithreaded expression evaluation and coroutines."

Seems that most of it's novelties were eventually added into LISP proper.

maybe they just made a mini-lisp and called it MDL?

It’s very Lispy, but it’s not strictly Lisp. Why, for instance, use “<“ and “>” to surround various forms but not others? If they were to make a mini-Lisp, I’d expect something more like Gnu Emacs Lisp, something that’s obviously a Lisp, but heavily influenced by the Lisps of the day. I’ve found a few old MDL manuals linked from Wikipedia, but none of them have any sort of “Here’s why we created MDL” section that I could find.

MDL is Grue Emacs Lisp ;)

Heh!

From one perspective ADVENT is just SHRDLU turned inside out, after all. (Though of course from another perspective it's a fancier WUMPUS.)

('ADVENT' is https://en.wikipedia.org/wiki/Colossal_Cave_Adventure , for anyone who isn't familar.)

(This is not an original observation, to be clear: see eg. https://nickm.com/if/riddle_machines.html )

A nice overview of the source code for that is:

http://literateprogramming.com/adventure.pdf

I remember playing it and finding a bunch of listings

  There is an enormous stack of line-printer paper here.  It is barely
  readable and totally unintelligible.

and:

  <DEFINE FEEL-FREE (LOSER)
    <TELL "FEEL FREE, CHOMPER!">
    <MEMQ ......
  The rest is, alas, unintelligible (as were the implementers).

I read a while back it’s a language called zil based on MDL.

https://the-rosebush.com/2025/07/studies-of-zil-part-2-how-d...

You mean an e-quine?

GPT-5 can't handle 2 things: an esoteric quine or an aquatic equine

You get the "more clever than GPT5" award today!

Pebble was amazing! Whimsical, useful and a loooong battery life. Strangely, there’s no alternative on the mass market. Or there has been no alternative.

P.S. your homepage is terrible.

I love the homepage design! Never had a pebble, but if it has ECG like Apple watch, i'm interested.

BangleJS

Sounds better than many books on Audible.

I built a similar thing a while back: https://github.com/katspaugh/interlinear.io

Useful when you're at a very low level of a language. Past B1 or so, it's mostly looking up individual words.

This is nice! Have been building something similar for Sanskrit (links in my other comment).

And yes, this is not helpful beyond a point. But most people never reach that point. So, any help is useful.

We’re also seeing super slow dev build times with Next.js. It takes around 12 seconds for the initial page and then 2-3 seconds for other pages.

I always thought it was an inherent issue in this app due to the many dependencies (9000+ modules for each page) but I’m going to try what the article suggests.

Does Rsbuild or any other build tool provide an alternative to Static Site Generation with automatic code chunking? Because this part of Next.js is great. The production app is super fast to load.

Edit: looks like Rspress does exactly that.

Love the book! It's fun and inspiring to write programs recreationally!

Some of his best movies are Le Circle Rouge and Le Samuraï.

His sister is in Le Samuraï as well.