Zork: The Great Inner Workings (2020) | Hacker News Reader

Zork: The Great Inner Workings (2020)

medium.com

205 pointsgus_leonel68 comments

If anyone's interested, here's a pretty good overview of Infocom's internal development tooling (how they got from the ZIL source code to a compiled story file). https://github.com/ZoBoRf/ZILCH-How-to

Additionally, this site is a really cool visualization that lets you play Zork while being able to see its internal game state and the corresponding source code to where you currently are. https://eblong.com/infocom/visi-zork1/

Re: the Visual Zorker visualisation, see Andrew Plotkin's associated blog post: https://blog.zarfhome.com/2025/01/the-visible-zorker

[ commentless HN discs: https://news.ycombinator.com/item?id=42769679 ]

Modern interactive fiction is pretty amazing.

I like Zork a lot, but I've never made it even halfway through the game. Knowing that you can permanently lose by doing certain things (does breaking the mirror count? I can't remember) puts a damper on my enthusiasm. The infocom parser is pretty good, but I sometimes run into issues where I know what I need to do, but can't get the parser to accept my commands. This happened in planetfall where I needed to get a key under a grate, but eventually gave up and looked up what the parser wanted.

Some modern additions are automatic maps and fast travel so you don't get lost in an infinite labyrinth and have to write your own maps (it was fun once, but I wouldn't want to do it each game). Fast travel is nice, so I can get back somewhere without having to type N N E NE N and so on.

As for this thread and more relevant comments, I had no idea the parser variables were named after the authors. HN knowledge never ceases to amaze me. I wish there was a convention or something once a year.

I was 15yo in 1977 when our two school paper terminals connected to a Milwaukee Public Schools PDP-11 provided access to the FORTRAN version of DUNGEO, aka mainframe Zork.

Memorizing the map was considered a high calling among the half dozen students who became obsessed with the games (ADVENT aka Adventure or Colossal Cave was also on the PDP).

At that time, death was just a part of the puzzles of the game. Repetition was not a hindrance.

Of course we did learn how to save, but you could only have one save file at a time.

Then we learned how to “detach” our session (RSTS/E operating system) and rename save files as well as holding a current game’s state until we later reattached that session.

Mind you this was all on green bar paper.

I’m still involved in the IF community, board member of the IFTF, and will occasionally try to play Dungeon from memory.

Even made a map a few years ago.

https://plover.net/~dave/DungeonMap.pdf

I think any gamer worth their salt should take on the challenge.

I also think Dungeon would be a great challenge for teenagers in a computer science class.

I had almost the exact same experience/environment in another state. It was on a vax though.

and green bar paper reminds me of part of dungeon where they had the stack 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).

the original source can be found here:

https://simh.trailing-edge.com/software.html

> At that time, death was just a part of the puzzles of the game. Repetition was not a hindrance.

Some of this design philosophy continued into arcade games (where inserting a new token was profitable) and onwards to early home consoles. ("Nintendo Hard.")

That is one of the better maps I've seen. Thanks for the lovely historical insight! I wish these games were still popular in my youth as it would have been a blast to play with friends (although GoldenEye was fun too).

> Modern interactive fiction is pretty amazing.

Got any recommendations?

Like you, I was never able to get into Zork. I really enjoyed it, until I inevitably became stuck. Hearing you say this could be permanent really seals it. It's one of those games that's more fun to recollect than to actually play.

Anything by Emily Short has really good writing and prose. They vary greatly by length too. I think she mainly writes for commercial projects these days, so I haven't seen a release in awhile of just something for fun.

https://ifdb.org/search?searchfor=author%3AEmily+Short

Alabaster, Bronze, and City of Secrets were all excellent. Check out the really short Galatea to see how much parser improvement has occurred since the old Infocom days.

I also recommend Andrew Plotkin

If your looking for a puzzley experience, Emily Short's "Counterfeit Monkey" is consistently top rated and I found it wonderful

My canonical answer is "Violet" by Jeremy Freese:

https://ifdb.org/viewgame?id=4glrrfh7wrp9zz7b

My favorite short IF is Glowgrass:

https://ifdb.org/viewgame?id=b1xy3s75cjlty973

The best regarded long one I know is Anchorhead, basically a Lovecraft story:

https://ifdb.org/viewgame?id=op0uw1gn1tjqmjt7

…these are both from 1998 so clearly I haven't been paying attention to what anyone is up to lately.

Since each person has their own taste, I suggest to browse IFDB's recommendations (https://ifdb.org/search?searchbar=played%3Ano+willplay%3Ano+...) and choosing something that piques your interest.

Personally I found Violet excellent as first-time adventure. The puzzles are hard but fair, and no mapping is involved.

CEJ Pacian makes excellent modern IF. I recommend checking out Gunmute and Red Prince.

Adam Cadre’s games are excellent. https://adamcadre.ac/if/

there is always the save restore so it is never permanent. However you can wonder around for a long time seeming to make progress only to discover it isn't possible to solve the game because of something not obvious you did a while back.

I like IF that is not about puzzles but about adventure stories and social relationships. There opened a new IF universe for me. I recommend the games of "Choice of Games" for that.

If you’re a fan of the genre, Zork I is worth playing through (with hints if necessary) precisely because it’s an example (the example?) of the early explorations of the art — back when it was about, “Hmm, what we should make the game do if someone tries to break the mirror?” and then implementing a response.

It wasn’t so much about finishing the game, it was about exploring the playground.

I never made it through--but then I've mostly always been more of a fan of the idea of gaming than the work of actually playing a game until the end. I did finish a few of the Infocom games generally with the help of a few (literal pre-Web) phone calls to the developer of some of those games who was a friend of mine from school. (There's even a very inside joke related to me in Planetfall. Don't ask because I won't tell.:-))

Now that is a heck of a story :)

Don't stand in the way of history. Tell us!

Basically, a minor character has a name that is, to someone in the know, obviously a derivative of a playful expression that the author would regularly use on me in school. I don't actually remember the character--though, for some reason I'm thinking it was an ambassador.

> Knowing that you can permanently lose by doing certain things (does breaking the mirror count? I can't remember) puts a damper on my enthusiasm.

I ran into that with the Hitchhiker's Guide game. I dropped one item early on and got to the end and couldn't finish without it.

It’s not clear in the beginning that you need to buy the cheese sandwich and give it to the dog. So you get approximately halfway through and then need to restart from practically the beginning.

...the parser variables very much were not named after the authors. The parser variables were PRSA, PRSO, and PRSI (Parser Action, Parser direct Object, and Parser Indirect object). The mention of JW and MARC was in comments they left in lieu of nonexistent commit messages.

Years ago I made some maps of things like which rooms of the game have the “sacred” bit, and what it means.

If you’re interested: https://www.reddit.com/r/raldi/comments/10dtch/i_spent_my_we...

>We can also make an educated guess that MARC refers to Marc Blank, who was one of the initial creators of Zork, but I couldn’t figure out who JW could be.

JW: That was probably Jerry Wolper. He was a coder and designer at Infocom in the later 1980s

https://www.mobygames.com/person/336/jerry-wolper/credits/

I had forgotten Jerry actually worked at Infocom for a bit. And no one else with those initials comes to mind. He was part of the film committee that Marc and others were involved with as well.

I tried making a random text adventure generator a while ago, where the idea was I would fill it with items and item combinations, and ask it to look for the most crazy/interesting chain of item manipulation puzzles e.g. key opens chest, to get the matchbook, to light the dynamite, to blow open the safe, to get the money, to buy the ticket.

I thought the idea was really promising at first but my finding was there's just not that many interesting puzzles based around real-life item interactions, so it wasn't fun to play (without going further by adding a plot, NPCs and graphics).

If the item interaction is too everyday, it's too obvious and boring (like a key to open a door), and if it's too obscure it feels unfair (like a blowtorch to the neck of a wine bottle to open it), so it has to be somewhere in between and there's only maybe a few hundred types of interactions like this. When you look into it, you'll notice the same item interactions reused in lots of adventure games too e.g. bolt cutters + chain, torch + batteries, spade + ground.

Maybe sounds obvious when you think about it, but it wasn't obvious when I was prototyping.

I think this is one of the reasons escape room games devolve into obscure logic puzzles. And also why adventure games got criticised for having super obscure (moon logic) puzzles. And also why a lot of adventure games feature things like time travel and magic and NPCs, as it lets you introduce new rules for item interactions.

Another aspect is filling the game world with lots of items, locations and general red herrings so that the connection between items that is usually obvious isn't immediately obvious because you can't keep them all in your head.

The bugs were funny though. Like you had to tell it you can't put a car in a tree, that you shouldn't bury bread in the dirt, and a vending machine can't dispense vending machines.

> The bugs were funny though. Like you had to tell it you can't put a car in a tree, that you shouldn't bury bread in the dirt, and a vending machine can't dispense vending machines.

My god, leave those in!

It does sound fun, but there would have to be some layer to control the amount of wackiness, or it would feel random to the point of meaninglessness, like "procedural oatmeal"?

Sibling comment mentioned Zelda but really this reminds me of the Zelda Randomizers.

Fixed set of items in a fixed set of locations. Certain locations need certain items to access. So therefore randomize which items are in which rooms while still being able to complete the game.

Then they have a door randomizer which changes where doors warp you to - effectively changing the map and potentially the order of items needed.

> Fixed set of items in a fixed set of locations. Certain locations need certain items to access. So therefore randomize which items are in which rooms while still being able to complete the game.

So this part of the puzzle generation isn't that bad, you also need to make sure you can't make it unsolvable by destroying/wasting items or making them inaccessible (or allow this and call it "Sierra mode"). Making the item usage interesting is the hard part, but that's not as much as an issue for a Zelda game.

I figure its mostly a graph problem. I do like door randomizers, esp ones that don't have bi-directionality.

Yeah, it's essentially https://grumpygamer.com/rtmi_pdc/ where the items are nodes and the edges are item interactions, and you're trying to carve out a subgraph for whatever your definition is of a good adventure game.

I think you've reinvented the game Scribblenauts.

Does it do interesting chains of puzzles anywhere (more like Zork)? From what I've seen, it looks more like isolated puzzles where you summon any object you can describe to solve the problem rather than having to make do with what you find.

You may want to look into the nintendo zelda chemistry engine

Any good links? I briefly looked up that it models fire, electricity, cutting etc. interactions, but if you put this in the frame of a text adventure rather than a 3D action adventure, would it be fun? e.g. typing "shoot arrow at explosive barrel" isn't as engaging as aiming the arrow yourself.

I also experimented with randomized text adventure game development back in the day but went a different route. I created an n-gram markov model scraped off hundreds of old I.F. games. The maze would be generated, and the markov model would be used to generate the description of each of the rooms. I then used a keyword/POS extractors/etc. to extract a structured well-modeled data representation of the room in terms of items, monsters, etc.

It worked kind of well, but it was more of an experience in exploration (almost like a MUD) rather than a cohesive linear game.

Of course, much later down the line, games like AI Dungeon really dialed up this concept.

So I've tried playing with LLMs for this and the problem is you can talk your way around puzzles because you can't guarantee it'll stick to constraints. If you want something more like a traditional adventure game, it would probably work better to algorithmically generate a cohesive linear game, then use an LLM to fill in the gaps in the descriptions, story, NPC conversations and images.

I don't see how you could avoid the LLM saying stuff that implies workarounds to puzzles though e.g. you add a knife to the game that's meant to be the only object that can cut a rope, but the LLM describes a rock as sharp.

This is my experience as well. It's why LLM dungeon masters can tend to fall flat since they have a hard time distinguishing between a genuinely good alternative solution vs somebody clearly cheesing it.

If you want to force a solution, you have to add an additional prompt check - e.g. "Is the solution provided by the player properly making use of items only within their inventory (A, B, C -> from their actual inventory) and does the solution adhere to one of the following acceptable solutions (X, Y, Z)?"

This helps mitigate abuse but it can also unnecessarily constrain the player, so YMMV.

I just don't see how you can stop it from going off the rails though. For linear adventures, you have to carefully pick which items and actions are available to avoid breaking puzzles e.g. if the LLM lets you break a window which creates a shard of glass, you could use the glass to cut a rope, when you were suppose to use a knife. It would probably be able to come up with decent reasons to explain why you can't do stuff though haha, which is often a part of good puzzle design e.g. "I don't want to draw attention by breaking the window".

I imagine there's also a potential exploit if you're ever allowed to name your pet, and the LLM isn't shielded from ever knowing the name.

Puzzles are to adventure games like frosting to a cake.

It’s expected to be there as a convention, but you want to keep it to a minimum because too much spoils the whole experience.

The story is the point and the puzzles are just an accepted contrivance to make you stay longer within the world. They’re not objectively fun, like frosting isn’t objectively good.

While the story is definitely very important, a point-and-click adventure game without puzzles is basically just a book where you turn pages by clicking the mouse.

Of course there are good puzzles that integrate well with the world (like reading from the world's most boring book to help the oyster fall asleep in King's Quest VI), and bad puzzles (cough arbitrary sliding tile puzzle cough).

I partly agree, but there are puzzle heavy games with throwaway plots and people do mention they like games because of the puzzles, so I thought my observation was non-obvious.

People still talk about the death of adventure games being caused by moon logic puzzles, but I don't hear people mention that part of the problem is the design space of item based puzzles is more limited than you'd think (I mean non-magic, non-NPC stuff) and has been mostly explored.

Weird to learn that the three Zork games were one game split, given that they're so tonally different. (the first having basically no narrative at all, the second being profoundly silly, and the third being fairly serious and dark)

Zork 2 and 3 actually contain a lot of new material, which is the origin of the tonal shifts. The original mainframe Zork was basically a larger version of Zork 1. Zork 2 took most of the rooms that had to be cut for space, weaved them together into a new map, and added a lot of new things as well, including all the sillier plot elements, like the wizard. Zork 3 was almost entirely new, except for the brief endgame which was the original endgame of mainframe Zork.

Huh! I've only played the original mainframe game. I thought the Infocom releases were just that game split into three—I had no idea about all the new material! I'll check them out.

I guess it depends on how you view it.

Originally there was a Zork game, then a company was formed and they decided to repurpose their existing game as their first project/product because they had no other inspiration.

The existing content was too large to fit into the small computers available at the time, so only parts of the prior-art were used in the "new" game.

Zork [1] was popular so the idea of a sequel was raised, and of course the content which was left out was then used the second time around. And repeat.

I think it's fair to say there was no cohesive intentions to make three games initially, and you can kinda see that from the "plot".

The folks who wrote the original Zork were AI Lab folks who were generally more interested in the technology than the game design. They weren't really that interested in games as the end state. Which led to Cornerstone but that's another story.

And, yes, computer resources were a big issue. I think you can find at least one presentation from Dave Lebling (maybe?) at GDC where he talks about some of this.

If you play a version of the original MIT Zork (often called "Dungeon"), you'll find it isn't exactly the three games put together. There really isn't much plot in the original -- like Zork I it is just basically a treasure hunt (as was the Colossal Cave Adventure which inspired it). There are additional puzzles and plots in Zork II and III that have no basis in the original.

MC Frontalot - It Is Pitch Dark:

https://www.youtube.com/watch?v=4nigRT2KmCE

>Front geeked around for an evening in a basement in Massachusetts. It got filmed in HD. Then Jason Scott made a whole video out of it. The song and the video are in service of Jason's upcoming documentary about text adventures, Get Lamp, but you get to enjoy it now. Peek the cameo by Steve Meretzky.

"Get Lamp" documentary:

http://www.getlamp.com/

Trailers:

http://getlamp.welcometointernet.org/trailers/

Full archive:

https://archive.org/details/GET_LAMP_The_Text_Adventure_Docu...

Interviews:

http://www.getlamp.com/cast/

I was working on a text adventure myself. It was before llms became a thing. My solution to the commands was to use simple words, not phrases. Pick key, walk north, look fireplace...

Instead of just a game I realize as I writing a game creator as well and the score kept getting bigger, so I decided to take a break. That was 4 years ago.

I do want to fishing it one day

Here is the case study of Infocom

https://archive.org/details/infocom-paper-2000-12-15/page/n1...

They were using virtual machine technology to be writing the only portable software at the beginning of the 1980s and they were gone by 1989

For those interested in retro text adventures, there's a podcast which has been going for about a year - they cover a couple of games each fortnight now. Worth a listen:

https://retroadventurers.podbean.com/

One thing I'd like to see AI used for: draw the Zork scenery based on the in-game text description.

Playing Zork with AI-generated imagery [video] (youtube.com)

133 points by mr_walsh on Jan 8, 2023 | hide | past | favorite | 54 comments

https://news.ycombinator.com/item?id=34300765

https://www.youtube.com/watch?v=ZpCrBBj6AWE

I think people have been experimenting with integrating stable diffusion into text adventure games (like AI Dungeon) for a while now.

Prompt adherence is always an issue particularly in a text adventure game where position matters - e.g. (The crystalline amulet is sitting on top of the old wooden table. A fireplace at the end of the room gently crackles. To the east lies a door, cracked with age. On the west wall you can see a painting of the lords of the manor. etc)

Flux could probably get close - though it clearly doesn't have any examples of an amulet in its training data. :)

https://imgur.com/a/giBqQ1L

That's a fun idea. The lack of continuity will be amusing, especially given the sparse descriptions of everything.

The later Infocom games — I’m thinking especially Moriarty’s evocative Wishbringer, Trinity, Beyond Zork — have much improved writing in general; it might be interesting to see what an LLM actually does with those.

A Mind Forever Voyaging has very good writing as well (and less in the way of puzzles). Of course, Hitchhiker's has great writing too.Not sure I ever played Wishbringer or Beyond Zork but Trinity is good.

That sounds like the perfect use for LLM image generation!

Somebody did this recently for Advent of Code (some year, not 2024 necessarily). I don't remember any details, but the images were entertaining.

We really need some innovative LLM-based games that provide a similar experience as the original infocom games but with a much richer parser and perhaps more detailed texts.

https://news.ycombinator.com/item?id=15080221

DonHopkins on Aug 23, 2017 | parent | context | favorite | on: The Enduring Legacy of Zork

“The MIT machines were a nerd magnet for kids who had access to the ARPANET,” Zork is how and why I got on the ARPANET as a nerdy kid. And I wasn't even a Russian Spy! [1]

Connecting to the ARPANET and getting an account on DM was an adventure in itself, almost like the beginning of the game itself.

At the time there were no passwords or anything but security through obscurity on the ARPANET TIPs. And the MIT-AI Lab was kind enough to hand out free after-work-hours "TURIST Accounts" [2] to anyone who asked nicely with the right magic words.

Some dude named Bruce who had a BBS (Bruce's NorthStar Horizon in Northern Virginia) told me how to do it step by step:

1) After 8PM EST, dial up the NBS TIP at (301) 948 3850 [3] at 300 baud, typed "E" to get the banner, then "@L 134" to connect to AI. (NCP host ids were only 8 bits, before TCP/IP's vast 32 bit address space!)

2) Make up an account name (I chose A2DEH).

3) Try to log in with that name, like ":LOGIN A2DEH".

4) If it asks for a password, somebody already has that account. In that case, think of another name and try again. (RMS's password was famously "RMS", after they forced everyone to use a password over his objections).

5) If it doesn't recognize your user name, it asks "Do you want to apply for an account?" Answer YES. When it asks "Why do you want to use the MIT-AI Lab's PDP-10?" answer "Learning LISP." (Which, as it turns out, is a long incremental process pursued over a lifetime, since there are so many implementations of LISP on the inside with names like MDL and JavaScript on the outside.)

6) When the account is approved, now all ITS systems know about you (ITS had network file and account sharing long before NFS and YP), and although you still can't log into DM directly, you could log into AI to learn LISP (and EMACS).

7) The MIT-AI Lab staff would kindly and patiently go out of their way to help you learn LISP and EMACS. (Many thanks to KMP for writing TEACH-LISP and answering my clueless tasteless questions like "how to you set the value of a variable?").

8) To play Zork, dial up the TIP after 8PM and connect to DM with "@L 70".

9) Log in as "URANUS" with password "RINGS".

10) So as not to look suspicious (3 kids from all over the country [4] logged in as URANUS, URANU0, URANU1 at the same time all playing Zork or watching each other play), change your user name to your own with ":CHUNAME A2DEH".

11) Only two people could play ZORK at once, so hang out chatting with other people waiting to play ZORK, or spying (in a socially acceptable manner) on whoever's playing ZORK via ":OS PDL" (for "Output Spy Paul David Lebling"), or snooping around trying to find the Zork source code [5], which was well hidden.

12) There was no file security, so you could snoop around Marvin Minsky's home directory and hurt your brain trying to understand what appears to be line noise, but is actually the Universal Turing Machine he implemented in TECO. [6]

13) When somebody from USER-ACCOUNTS sends you a "nice private message" telling your they know what you're up to with ZORK, and that you should really learn LISP like you said you would because it's such a great language, instead of demanding you commit "seppuku" and "dumping you off the net and be done with it", you simply start learning LISP instead of acting like an entitled dick [7] by whining about how the people who gave you a free account that you bragged about in BYTE magazine are a bunch of communists and threatening to get some Proxmire type to start inquiring into its operations by seeing if your "Pentagon friends can upset them. Or perhaps some reporter friends. Or both., Or even the House Armed Services Committee."

[1] https://www.youtube.com/watch?v=hVth6T3gMa0

[2] http://www.art.net/~hopkins/Don/text/tourist-policy.html

[3] https://www.saildart.org/TIPS[P,DOC]3

[4] https://archive.org/details/getlamp-rgriffiths

[5] https://github.com/itafroma/zork-mdl

[6] https://news.ycombinator.com/item?id=13514918

[7] http://www.stormtiger.org/bob/humor/pournell/story.html

kabdib on Aug 23, 2017 | next [–]

Oh, you are A2DEH. "Hi" from 1979 or 1980, from ZEMON. I saw you online a lot, playing Zork on MIT-AI, and I probably :os'd a few of your sessions. I too was using the NBS TIP. Later, I actually worked at NBS and became "legal". I first learned Emacs at 300 baud; I'll spare you the whole story, but it involves a lot of assembly language and some soldering...

DonHopkins on Aug 24, 2017 | parent | next [–]

Of course I remember your cool UNAME standing out in all those :WHOJ's! ;) Do you remember Rob Griffiths, aka ROBG? I really enjoyed his full interview from Get Lamp -- he really nailed what it was like at that time, making a pilgrimage to 545 Tech Square as a 15-year-old kid!

https://archive.org/details/getlamp-rgriffiths

He and you are a couple of the people who I was thinking of when I described kids from all over the country hanging out chatting and spying and waiting to play Zork!

kabdib on Aug 24, 2017 | root | parent | next [–]

My best friend in high school went to MIT and I ... didn't (it's okay, the state college I wound up going to was about my academic speed, and I would have been toast in a couple of semesters at MIT). I also did a pilgrimage to MIT and saw the DEC-10s. Printed out a school project on the LGP, played around with a Lisp Machine for a few hours.

MIT's friendly, unparanoid attitude towards people using their systems and basically just digging their technology was very formative in my career. Zork was the hook. I came to play adventure games, I stayed to learn Emacs and a bit about networking and PDP-10s, and LISP. I don't use PDP-10s anymore, but I work in the games industry, use Emacs every hour of my working day, and wish I could write more production LISP (though if you squint at Javascript just right...)

It may be to your amusement that it is entirely possible to run an ITS KS10 directly on the unfiltered Internet without major issue. I did not even have to run PWORD. It seems that modern automated internet scanners are completely confused that DDT does not produce anything resembling a login prompt, so they either sit and wait for one that never comes (and get killed off by the gunner) or disconnect.

The machine is running right now, but is presently isolated from the internet while a hardware issue is worked on.

  SE ITS 1652  Peek 632   1/21/2025 07:38:37  Up time = 1:10:37:08

Reply on news.ycombinator.com