It seems to often make it hard to get back where you were, like you go west and then east and end up in a different place. I don't know enough about z-machines to tell why.
Not ZMachine related, not even with Inform6, where n_to, s_to and w_to e_to are pretty much self-explanatory. It's just that Dungeon/Zork/Zork-I-II-III were made that way.
It has been a while since I've played interactive fiction, so I can't make a specific suggestion, but modern games seem to be better at keeping the directions consistent (or at least providing clues when they are not). As others have noted, older games broke directionality to serve as puzzles -- failing to acknowledge that some people have a sense of direction while following twisty paths!
Newer games also tend to follow some quality of life rules in their design, things like avoiding arbitrary deaths and avoiding situations where the player cannot progress because they missed something earlier in the game.
Anything Non-Zork. Even Adventure from Don Woods was half-consistent in some places. But having an "odd" geometry matched perfectly the environment of a cave.
It's still pretty good. TADS is a more modern alternative, and the one I would go with, but basically these haven't been commercially viable products since the 80s so there's not a lot going into it.
Arguably, the right answer now is to document everything that matters to you about the adventure, and tell an LLM to run it.
It does even better than that.
I have a private interactive fiction game going based on a popular novel. It's very immersive. Still a work in progress. Right now you can do the daily life things the novel skips like making dinner. I make it so you can skip the day but I was enjoying doing the routine things. It's hard having the game keep track of objects and some NPCs knew things they shouldn't. I tightened it but testing is a lot of work. Hmm. Maybe I can have another LLM play test it.
It also allows you to deviate from the novel. This is a romance novel. I had the main character choose someone else and it showed an alternate version of the novel. After the main threat of the story was dispatched things settled into weeks of routine living where I added another person into my routine. I played to see if another event would show up but after 3 in game weeks I decided to stop. This was a test and I needed to improve the game. I can see this becoming addictive to some people because they will be living in the story and having a life there and making friends and finding partners.
This type of game is still uncharted territory. Does one end the game or let the player play indefinitely? Is it more fun to have the player do the daily life things or skip those or give them a choice?
Well, TADS is older than Inform, yes, but the whole idea of VM-based platforms for adventures originated from Infocom's Z-machine starting with Zork I in 1980 -- it's what let them basically release their games for every platform current in the US in the 1980s.
Inform 7 source code, at first glance, looks like plain English. It makes you start to wonder whether Inform 7 is actually an LLM and not a compiler. Then as you look at more code, you quickly start to see the structure of the language, and realize that it is indeed a programming language merely structured to look like English at first glance. But it's very cool that you can do things like this (example from the Inform 7 documentation):
Definition: a room is neighboring if the number of moves from it to the location is 1.
Every turn:
if a random chance of 1 in 2 succeeds:
let current location be the location of the lurking critter;
let next location be a random room which is adjacent to the current location;
if the lurking critter is visible:
say "The critter [one of]slouches[or]slithers[or]shambles[or]lurches[at random] away.";
move the lurking critter to next location;
if the lurking critter is visible:
say "A critter [one of]oozes[or]staggers[or]ambles[or]creeps[at random] into the room.";
[Whether or not the critter has moved, we need to adjust the sword-glow, because the player may have moved.]
if the lurking critter is in the location:
adjust sword-glow to glowing brightly;
otherwise if the lurking critter is in a neighboring room:
adjust sword-glow to glowing faintly;
otherwise:
adjust sword-glow to glowless.
This snippet omits the definition of "the lurking critter" (a standard NPC) and the sword (a standard object with a custom defined method called "adjust sword-glow to"), but it should give you a good idea of what Inform 7 source code looks like. (Inform 6 looked much more like a traditional programming language).
Yes, very easy to grasp and games will be much lighter than Inform7. Tads it's propietary even if the interpreters are free. Inform6 has the whole stack (even free documentation too), from Inform Beginner's Guide to these, where you have libre games for Inform6 and some free as in freedom
related documentation of Inform6 internals.
There's also ZIL by way of ZILF https://github.com/taradinoc/zilf This is the Lisp-like language that Infocom implementers actually used back in the day.
Infocom came out of the MIT AI Lab which was very LISP-centric at the time. Depending upon you you listen to, the gaming part was something of a side effect as opposed to the ultimate goal.
Yes, they actually planned to release cross-platform applications in addition to games and actually shipped one (1985's database Cornerstone). The problem was 1985 was a bit late for cross-platform business software to be successful; the IBM PC and compatibles had taken over the business market, and having apps in a VM just meant they were a bit slower than native ones.
This is very neat! Been wondering what people would do with the Zork code now that it is open source [1].
[1]: https://opensource.microsoft.com/blog/2025/11/20/preserving-...
Nice. Like watching clockwork internals. Came across a similar engine for another text based adventure a while ago: https://github.com/efa/Wilderland
This is neat. Z-machine internals without needing to run a debugger.
Very cool. Did a very little testing for Infocom back in the day and knew a lot of the folks involved.
Neat! Any cool stories? What games did you work on?
It seems to often make it hard to get back where you were, like you go west and then east and end up in a different place. I don't know enough about z-machines to tell why.
Perhaps you could try making a map.
Or in this case; just click on "map"
Nothing to do with the VM. The game was designed that way, navigation is part of the puzzle.
Not ZMachine related, not even with Inform6, where n_to, s_to and w_to e_to are pretty much self-explanatory. It's just that Dungeon/Zork/Zork-I-II-III were made that way.
It has been a while since I've played interactive fiction, so I can't make a specific suggestion, but modern games seem to be better at keeping the directions consistent (or at least providing clues when they are not). As others have noted, older games broke directionality to serve as puzzles -- failing to acknowledge that some people have a sense of direction while following twisty paths!
Newer games also tend to follow some quality of life rules in their design, things like avoiding arbitrary deaths and avoiding situations where the player cannot progress because they missed something earlier in the game.
>Modern
Anything Non-Zork. Even Adventure from Don Woods was half-consistent in some places. But having an "odd" geometry matched perfectly the environment of a cave.
Would Inform 6 nowadays be still the language of choice for developing text based games like Zork? Or even a reasonable choice for that matter?
It's still pretty good. TADS is a more modern alternative, and the one I would go with, but basically these haven't been commercially viable products since the 80s so there's not a lot going into it.
Arguably, the right answer now is to document everything that matters to you about the adventure, and tell an LLM to run it.
I wonder if you could make a "choose-your-own-adventure" book uning LLMs.
It would write the book ahead of you, kind of like how railroads could send trains with supplies out to the track as it was built.
Like with any LLM produced project, can you? Sure. Will it be engaging, well written or consistent? Almost certainly not.
You could, but no one wants to read or play GenAI content.
It does even better than that. I have a private interactive fiction game going based on a popular novel. It's very immersive. Still a work in progress. Right now you can do the daily life things the novel skips like making dinner. I make it so you can skip the day but I was enjoying doing the routine things. It's hard having the game keep track of objects and some NPCs knew things they shouldn't. I tightened it but testing is a lot of work. Hmm. Maybe I can have another LLM play test it.
It also allows you to deviate from the novel. This is a romance novel. I had the main character choose someone else and it showed an alternate version of the novel. After the main threat of the story was dispatched things settled into weeks of routine living where I added another person into my routine. I played to see if another event would show up but after 3 in game weeks I decided to stop. This was a test and I needed to improve the game. I can see this becoming addictive to some people because they will be living in the story and having a life there and making friends and finding partners.
This type of game is still uncharted territory. Does one end the game or let the player play indefinitely? Is it more fun to have the player do the daily life things or skip those or give them a choice?
TADS is hardly modern. It’s actually the oldest VM based platform from the early 90’s.
Well, TADS is older than Inform, yes, but the whole idea of VM-based platforms for adventures originated from Infocom's Z-machine starting with Zork I in 1980 -- it's what let them basically release their games for every platform current in the US in the 1980s.
But not publicly available as a platform.
I think Inform 7 is pretty cool. I suggest checking in to that as well.
Inform 7 source code, at first glance, looks like plain English. It makes you start to wonder whether Inform 7 is actually an LLM and not a compiler. Then as you look at more code, you quickly start to see the structure of the language, and realize that it is indeed a programming language merely structured to look like English at first glance. But it's very cool that you can do things like this (example from the Inform 7 documentation):
This snippet omits the definition of "the lurking critter" (a standard NPC) and the sword (a standard object with a custom defined method called "adjust sword-glow to"), but it should give you a good idea of what Inform 7 source code looks like. (Inform 6 looked much more like a traditional programming language).
> [Whether or not the critter has moved, we need to adjust the sword-glow, because the player may have moved.]
That must be a comment, right?
In that case: "The critter [one of]slouches[or]slithers[or]shambles[or]lurches[at random] away.";
[one of] [or] [at random] are also comments?
In code, square brackets indicate comments. But inside a string literal they indicate variable interpolation or other commands.
Yes, very easy to grasp and games will be much lighter than Inform7. Tads it's propietary even if the interpreters are free. Inform6 has the whole stack (even free documentation too), from Inform Beginner's Guide to these, where you have libre games for Inform6 and some free as in freedom related documentation of Inform6 internals.
https://jxself.org/git/
There's also ZIL by way of ZILF https://github.com/taradinoc/zilf This is the Lisp-like language that Infocom implementers actually used back in the day.
Infocom came out of the MIT AI Lab which was very LISP-centric at the time. Depending upon you you listen to, the gaming part was something of a side effect as opposed to the ultimate goal.
Yes, they actually planned to release cross-platform applications in addition to games and actually shipped one (1985's database Cornerstone). The problem was 1985 was a bit late for cross-platform business software to be successful; the IBM PC and compatibles had taken over the business market, and having apps in a VM just meant they were a bit slower than native ones.
I built https://github.com/chicagodave/sharpee/ as a modern IF platform (Typescript) that runs natively in a browser.
Also https://sharpee.net/