[delayed]
In two places the article states that the original game had the ability to save the updated tree ("it had the ability to save progress and load it during the next run" and "It is an amazing example... of how, even with such a simple language, ... and the ability to save new questions").
The later part says the opposite - that the original implementation had "No ability to save progress" and that this is new in the C++ implementation.
I can't help but wonder (also due to other language features) if the author ran the article through an AI to 'tidy up' before posting... because I've often found ChatGPT etc. to introduce changes in meaning like this, rather than just rewriting. This is not to dismiss either the article or the power of LLM's, just a curious observation :)
File IO was extremely hard on the early home computers, so it was very unlikely to have the ability to save and load the tree. And with such short code, there's no way to balance the tree either.
Bit misleading title as in the 80'd descision trees were concidered part of the discipline of AI, and you today might be surprised that machine learning was considered a diffetent discipline. This was the result of scientific kerfuffls about funding.
In practice many of us in the 'Nouvelle AI" movement that countered the ruling Symbolic AI paradigm (GOFAI was the commonn slur) had at least one foot in the ML, EA. And alife spaces as once you abandon symbolics straight programming becomes awkward fast.
I remember the version of this called Pangolins on the ZX Spectrum. Because pangolins were an unusual case (scaly mammals).
Looking it up, apparently it was a type-in in the programmling book that came with the computer.
https://spectrumcomputing.co.uk/entry/15157/ZX-Spectrum/Pang...
That's the book my (blind) dad got me to read to him which taught me to program!
This looks like "Pervading Animal" (or just "Animal") from the 1970's [1]. Said to be the first computer worm.
let's not forget the 60s: https://en.wikipedia.org/wiki/Matchbox_Educable_Noughts_and_...
I read about this idea and actually built a variant in college for the Washington University ACM (our CS student club). We ran a demonstration at one of the engineering school events and ended up refilling the matchboxes multiple times that day - it was a hit!
If I recall, there is a way to significantly reduce the number of matchboxes needed by taking advantage of mirrored board conditions. Somewhere I have a Perl script that generated HTML for all the matchbox covers and possible next states.
That's immensely more interesting than a program that plays 20 Questions. I'm amazed I've never heard of it!
chapter 8 of this book goes into detail and how to run your own. fascinating stuff! https://ia904503.us.archive.org/13/items/gardner006/gardner0...
This program was one of my first interactions with a pc. This was a wonderful shot of nostalgia.
Opening image: AI-generated picture of a slightly melting computer, with keys that can only type words like "kgonfutz" or "oøoøøo".
I can only assume the rest of the article is also AI-generated, with a similar attention to detail. Tab closed without reading.
Well, dear Peggy, the program in question is, in fact, known to have existed in 1974 on Multics (https://multicians.org/terminals.html), so you're not half wrong.
The other major issue is that it isn't machine learning—it's a classic example of an expert system, even if there is a bit of a gray area around whether a binary decision tree qualifies as ML.
The worst part, of course, is that it takes less time to slap "GUESS THE ANIMAL" on a stock image of a glass terminal than it does to prompt a diffusion model to generate the same thing with passable fidelity... and it still wouldn't be an accurate depiction of what the program actually did.
It can be difficult to find hero images for articles like this, so people generate them. Why is that bad to you?
Guilty. I worked with a version of this program.
I've occasionally thought about trying to organize a filesystem that way.
I am surprised that the program understood English grammar.
It didn’t. I implemented this program in BASIC as a teen. The brilliance is that, when it isn’t able to guess your animal, it asks the user for the question to ask, as well as the correct answer. All “intelligence” and grammmar comes from the user, but for someone playing it for the first time on a well-populated database, it looks impressively smart.