Glenn Wichman on The Cellar, 2021
Glenn Wichman, who wrote the first Rogue with Michael Toy at UC Santa Cruz, talks to Taigen Moon of the YouTube channel The Cellar: how he and Toy met in the computer lab in 1978, curses and the ADM-3A terminals, the players in the stat lab, permadeath and mystery items, the three stages of Rogue's making (Santa Cruz, Berkeley with Ken Arnold, and A.I. Design with Jon Lane), drawing the Macintosh graphics and writing the Atari ST version for Epyx, and what he would do differently.
See also the 2016 panel with Michael Toy and Ken Arnold.
About this transcript. Made from YouTube's automatic captions and lightly edited for readability: punctuation added, false starts, repetitions and filler words removed, and obvious mishearings corrected (names and titles, Wichman, Epyx, Jon Lane, ASCII, M.U.L.E., chess piece). The wording is otherwise as spoken, slips included; words in [brackets] were supplied. The captions do not say who is speaking: the host, Taigen Moon of The Cellar, narrates the opening and closing and asks the questions, and Glenn Wichman answers; a dash marks a change of speaker within a paragraph. Each timestamp links to that point in the video. The opening line is a clip from later in the interview, and the closing music is left out.
Transcript
0:00 "You know, if I had it to do over again, I would have made Minecraft." So, video game genres are a mess, and this owes a lot to the fact that, unlike other forms, games aren't just defined by their tones, moods and themes, but also all of the different ideas that they bring together. This is why so many tags can be applied to a single title, and it's rare that just one descriptor will tell the player most of what they need to know about what a game is. And that's why Rogue is so interesting — well, it's one of the reasons. This 1980 dungeon crawler was not the first, and it definitely wasn't the last of its kind, but its influence would become wide-reaching enough that those that took after it would be dubbed roguelikes, a term that persists today, in a world where Doom clones eventually became first-person shooters. Rogue was originally programmed by Michael Toy and Glenn Wichman, two students at the University of California, Santa Cruz, and today I have the privilege of sitting down with Wichman to talk a little bit about the game's development and his history as a designer.
1:15 All right. — Hi. — Hi, nice to meet you. Thanks for coming on. — I was born in Michigan, yes. — So what was your exposure to computers growing up?
1:23 Pretty much none. So my father was a college professor, and there was a computer at the college where he worked, and so I saw it and played with punch cards when I was a kid, but I didn't ever actually interact with a computer until I got to college. And that was at UC Santa Cruz — University of California, Santa Cruz. Yeah. So there wasn't really a group in that sense. There were four of us who should be considered the creators of Rogue, but all four of us have never been in the same room at the same time.
2:17 So it's not something the four of us created together, but it started with Michael and me. It started with — we met at UC Santa Cruz. We were both freshmen in the fall of 1978, and I had ambitions to be a game designer, which meant board games and tabletop games. I was, again, not familiar with computers growing up. I knew they existed, and I was kind of distantly fascinated by them, but I'd never seen one. And then when I got to college, there was a computer that, again, you didn't actually see. We had terminals — which, coincidentally, are the same ones that they're using in the Loki series on Disney+ — those ADM-3A terminals that people use.
3:23 So that's what we had, and they connected to a PDP-11/70. It was off in a basement somewhere, and nobody actually ever saw the computer unless you were in the IT department. And we had terminals, and it was a time-sharing system, and we used it to write papers and stuff like that. And so I discovered that there was this computer — there were games on the computer. Specifically, there was Adventure, or Colossal Cave; I knew it as Adventure. And so I played that, and I was like, "Oh, wow, I want to make games like this." And so I got a book on BASIC programming — BASIC the language — and started trying to create my own games, but I had no idea, really, how to program. And that was when Michael Toy wandered into the lab. Again, nobody had computers in their rooms or anything like that; there was a lab where there were, at that time, about four terminals set up. And he kind of watched what I was doing and said, "What are you working on?" And that was our first interaction. And I told him, and then we basically started making adventure games for each other to play — so, text adventures.
5:00 So you would type in commands, and you would see the results, and type in more commands, and that kind of back and forth. And after quite a while of this — the thing that was no fun about it is that I could make games for him to play, and he could make games for me to play, but it's no fun to play your own game, because you're basically creating puzzles that you already know the answer to. And so then, the next year, two things happened. One was, we were talking about: can we make a game where the computer comes up with the puzzles for you to solve, instead of having to do it by hand? And then the other was that these terminals were not built to do graphics. They didn't have the wireframe things like the PLATO system had; [it] is purely text. But there was — well, there was Bill Joy, who wrote an editor called vi, which I still use, and it had to run on a lot of different terminals. And so there was a library of routines that could basically place the cursor in specific places on the terminal, so that you could scroll up and down on your text.
6:37 So anyway, long story short: Ken Arnold, who was a student at UC Berkeley, took Bill Joy's code for vi and separated out just the library that did the cursor-addressing routines, called that curses, and made it available. And then Michael got his hands on it and said, "Oh, with this we can draw pictures on the screen," by using ASCII characters to draw lines and dots.
7:11 What I find so impressive about making games back in the late '70s is that games barely existed at the time. Nowadays there's countless books and [an] ocean of YouTube videos, and you can even take college courses on game design. So how do you even go about designing a game back then?
7:31 Yeah, it's a good question. We didn't have a language to use around game design, and we didn't really think about game design as its own discipline. It was all mixed together with programming, with software development. And so the people who were programming games — I guess we thought of ourselves as software developers, or programmers, or engineers, and we just wanted to play stuff, and had an idea about what we would create and how we would play it. And a lot of things just kind of works themselves out, in terms of game mechanics and balance: how many levels do you have, how many monsters you have, how hard should the game be. They weren't really strategically thought through. We just kind of hacked away at things until we liked what we saw.
8:53 So you guys were constantly playing it and testing it? — Yeah, we were. There were about 20 people who just were people who hung out in the stat lab — which, again, is where the terminals were — when they should have been in class, all playing it. And even before we had monsters, all we had was a dungeon you could walk around in, and people were playing that and really having a great time.
9:25 So is that the moment you knew that there was something really compelling about what you'd made? — Yeah, I mean, I don't know if there was a single moment, but that was certainly — in moments like it, we realized we had something pretty special here.
9:49 So did you think it was funny that there were games on the PLATO that could have been considered roguelikes? — Yeah — which I had never seen, by the way. I did not, until years later, know that there was this PLATO system that also had games that would be considered roguelikes. I was completely unfamiliar with that. — Right. So that was just — but you guys ended up getting the name.
10:13 Yeah, well, what's funny is that we ended up getting the name, right? That the genre got called roguelikes — I think that was not by anybody's design. It just sort of is. It was one of those things that came along, that was kind of to our advantage. But we weren't basing it off of any other computer game, right? We were basing [it] off of Dungeons & Dragons, to the point where we had to make some very careful changes to the game when we decided to do a commercial version of it, so that Gary Gygax didn't sue us.
11:11 I don't know if this is going back too far, but do you remember a game called Beneath Apple Manor? — Again, right, I never had an Apple II, right? And it was a time in life where most people my age couldn't afford a computer, and if you could, then there was that one, and that's the only one that you knew anything about.
11:35 So, when reading books about roguelikes, I have seen Beneath Apple Manor mentioned as the first roguelike game. I didn't know about it; I never saw it or played it. But I understand that it's very similar, in terms of having the core things that make a roguelike a roguelike — which is a whole other discussion.
12:08 This is something that a lot of people have debated over the years: what makes a roguelike a roguelike. And Glenn did an interview with Gamereactor just last year, where he had some things to say on the idea of permadeath.
12:23 Well, yeah, because it's another thing like roguelike. I think there has been some research done on how the name roguelike came into being, but I don't know — I tried to find anything I could find out about it. I don't know who came up with the term permadeath, or how it became this term associated with roguelike games, or associated with this idea that you can't just save your game and then go back and replay it from a save point. But we certainly, when we were creating the game, didn't think of it as permadeath or anything like that, right? And again, as I mentioned in that interview, if you're going to have mystery items, then you need to have to live with the consequences, right?
13:30 So it's really the equivalent of taking your hand off the chess piece when you're playing chess. It's a turn-based game, and once you've committed your turn, then you have to live with the consequences of your turn, whether those are bad consequences or good consequences. And one of those consequences could be you dying, but that's not singled out as a permanent consequence where all the rest are temporary, right? All of the consequences are: you can't go back and redo. So, again, the whole game doesn't work if you don't have that mechanic in there, because there's no reason not to just try everything, and then, if it turned out to be a bad thing to do, you just go back and don't do it that way.
14:30 So you said yourself that a lot of the decisions were just things that ended up working out during development, and now, afterwards, people have taken that and dissected those decisions. Do you think that's strange at all?
14:47 No. I think, for the most part, where the genre went after [it] got into other people's hands, right — [those] are amazing, astonishing things. The things were taken so much farther than what we had envisioned.
15:16 So I think that the things that have been done with procedural generation, and other of those concepts that we came up with, have been overall great things.
15:40 Going back to something you said earlier: you wanted to be a game designer, but specifically tabletop gaming. — Well, yeah, and really more even board games. I was creating board games as far back as I can remember, since I was old enough to draw.
16:04 So I don't know; I can't point to "Oh, this is the genesis of that." From as far back as I can remember, I always wanted to either make board games or make cartoons. I wanted to have one of those jobs.
16:21 And then I played lots of different kinds of games, like Avalon Hill games, serious board games, because that's what the geeks in high school were playing, and I was a geek in high school, so that's what I played. And then Dungeons & Dragons kind of took over life for quite some time. But yeah, once I started figuring out about computers, I kind of never really looked back. It's like, "Oh, the computer can do all of the bookkeeping stuff that's such a drag, and you can just focus on the strategy." So once I figured out I could make computer games, that's all I wanted to do.
17:28 So a lot of people would say that games, by their nature, have more limitations than pen and paper, but not really, in your eyes?
17:35 Yeah. I agree that, in terms of doing role-playing and stuff like that, you can have a lot of freedom. And on that front, for my group of friends that were all playing D&D, after a while we completely stopped using books and rules and everything like that, and just basically treated it more like a never-ending-story kind of thing, where one of us would set up a scenario and the rest of us would say what we would do in that scenario. And so, if we were going to be creative, then I guess I'd put no constraints on it, and we would just make up stories. But if you're going to have a game with rules and constraints and stuff like that, then
18:38 Yeah, I like the capabilities that having the computer handle the boring parts of dealing with the game gave you. — So then avoiding that bookkeeping is what gets you to the fun part? — Yeah, I don't know.
19:12 I think that is a way of thinking [of] it. The other thing is that I loved writing code; that was the thing that I discovered. I didn't set out to be an engineer. I set out to make games, right? To be a game creator, whether you want to call it a designer or implementer or developer. And to do that, you had to write code, and then it turns out that writing code is an incredibly fun thing to do. So I guess that's the other part of it.
19:44 And you two bounced off of each other as coders, I understand? — And Michael had gotten his first computer in high school, and he built his own, so he was far ahead of me in terms of coding experience when we met.
20:00 So from the point of view of coding, at that point in our relationship it was really much more that I was apprenticed to him, figuring out how to write code. But in terms of game design, we were bouncing ideas off of each other. But again, we didn't really make the distinction clearly at the time. It was more in retrospect, looking back at it, [that] a lot of what we were doing would now be considered game design, not coding. But it was all mixed together at the time.
20:45 Of course, I wanted to know if there were any funny stories from development. — Oh, yeah. See, that would have been a good thing to have asked me in advance, so I could have some time to think about it. So I guess the answer is no, right? Because if there was, I would have it at my fingertips. There were — I mean, it was all more than 40 years ago.
21:13 So my memories of a lot of it are pretty vague. And there's kind of three different episodes of creating Rogue. There's what we did originally, where it was Michael and I in Santa Cruz, when we were supposed to be in class. And then Michael left Santa Cruz; he went to Berkeley, where Ken Arnold was, and then Michael and Ken were writing Rogue, and I was in Santa Cruz, and I had my own version that was kind of orphaned there, where I did some work on it.
22:00 So there was that initial one. Michael and I were roommates; we had [a] terminal in our apartment. We had a 300-baud modem with the mouse ears; we'd hook it up to the computer on campus and work on the game. So that was one life. And then later on, Michael and Jon Lane decided to create a company to try to make a commercial version of Rogue. And so it was Michael and me, and then there was Michael and Ken, and then a couple years after that, there was Michael and Jon. And so Michael and Jon did the PC version, and were initially basically manufacturing and boxing and selling it themselves out of Jon's garage. And then Epyx picked it up and decided to distribute it. At that point they wanted a Macintosh version and an Atari ST version and an Amiga version — well, initially it was just a Macintosh version. And I wasn't involved in any of that, but I wanted to get back into it. And so, for the Macintosh version, they needed — that was the first commercial version of Rogue that had actual graphics in it.
23:43 Right — up until then, the PC version still used — I forget the name of the character set now; it's beyond ASCII. It had the smiley faces and the double lines that you could use to draw rooms and stuff like that, but it still was not actual graphics. So the Macintosh version, they needed actual pictures, and so I designed all the graphics for that. And I was paid by being given the used Macintosh that I used to draw the graphics; I got to keep it. So that was my initial re-introduction to Rogue, after Michael and Jon had done the PC version. And then Epyx decided to do versions for the Amiga and the Atari ST — that's the Atari ST version behind me there — and I didn't do those graphics, but they brought me back, or I came back, and I wrote the code for the Atari ST version. So, proof that I can write code.
25:13 You did the graphics for it, so where did you learn how to do art? — That's a good question. I actually did take art classes in college. I took drawing, and in high school and college took drawing classes. And I had, like I said, been a cartoonist from a young age, so drawing simple, funny pictures is something that I had been able to do for a while.
25:40 So I guess that's the only answer there. I didn't take — nobody taught any classes. There was not Adobe Illustrator or anything like that; there was MacPaint. All you could do was make pixels black or make them white. So it wasn't a whole lot of technical skill involved.
26:22 Well, if I had — I don't think there's anything — let me think about that. Yes. I think the game that Rogue ended up being was very just hack-and-slash and kill everything you see, right? Which — if I had it to do over again, I would have made Minecraft.
26:47 And I think that if we had kept working on Rogue, and just kept it evolving with the times, I think that's kind of where it would have ended up: something similar to that. But I always had in mind — what I really wanted to be able to do was make it so that the player really could just be dropped in a world and do whatever they wanted in that world.
27:27 So, yeah, it ended up being too much about just killing monsters or getting killed by monsters, rather than other ways of using magic, or using potions, and all of those kinds of things. So I don't know exactly what I would have done differently, but I guess — the other kinds of games that I have worked on through the rest of my career have been a lot less violent.
28:19 So this is interesting, because if you look at the genealogy, Minecraft was inspired by Dwarf Fortress, which itself was inspired by other ASCII games like Rogue.
28:34 Yeah, it's my understanding that Notch played Rogue and liked Rogue, although I have not heard that directly. But I think that there is a connection there. It's just the direction things were building. There's a lot of different pieces that came there, but in terms of a procedurally generated world that you can explore, right, that's definitely — Rogue was among the first examples of something that attempted that.
29:20 So how do you feel, seeing how far games have come, and being able to be a part of that?
29:35 Well, I'm proud of what we did. I think if we hadn't done it, it still would have come about, if somebody else had. It's astonishing, all of the ways that technology has moved forward, exponentially, and all of the ways that it hasn't, that I thought that it might. So I feel — I'm not really proud, but I feel happy with the contribution we made. I feel like we made a significant contribution. I feel grateful that that contribution has not been just lost to history; I think a lot of other people's contributions have. One of the reasons I agreed to be interviewed by you is because of the things on your channel highlighting M.U.L.E., specifically — some of the other things that happened early on. So I feel grateful that they're remembered, and I feel happy to have been part of something that I think has been a good thing. I feel like computers overall can make the world a better place, I feel that games overall can make the world a better place, and I'm happy to keep working on doing that.
31:28 Where did you think technology was going to go?
31:36 Oh, gosh, I might want to get back to you on that, because I know I have good things to say there, but now I'm drawing a bit of a blank. I really don't think I had a vision for just how much faster computers would be, and how much more memory that they would have. Rogue ran in 30K of RAM, and that seemed like a lot at the time. So the idea that just a regular person would have a computer with four gigabytes of RAM — I don't think I'd have ever imagined that. I'm not sure there were four gigabytes of RAM on the planet when I started. So the idea that you would watch movies on your phone, or things like that — I don't think I imagined any of that stuff. It's like science fiction. It really wasn't, if you look at the hard sci-fi movies of the time, like 2001 or Star Trek, right? It never occurred to Star Trek that people could look at their communicators, right? They just talked to them.
33:22 So, final question: if you were your own boss, and time and money were no worry, what kind of game would you make? The dream project.
33:39 Yeah, that's a good question. To be really honest, I don't think that I — I like being a part of an enterprise. I have ideas, and occasionally I pitch ideas, but I don't necessarily want to just take something and pursue it on my own. I like being part of a team, and I like working with other people. At times in my life I've been kind of a visionary, and I certainly have visions for how I should do the product I'm working on, and occasionally I have ideas for games that I would like to pursue. But there isn't a passion of, "Oh, as soon as I get the time, this is the game I'll work on." It's really more like, every now and then, I'll think, "Oh, this would be a fun idea for a game; if I had time, I would work on that." But the next day I might have a completely different idea.
34:56 So I will — like games that I've seen where I feel like — when I first saw 2048, I was like, "Oh, that is a beautiful game. I wish I had thought of that game," because every time it's your turn, you can do one of four things: up, down, left, right. That's it, right? So there's this simplicity to it, but there's this beauty and replayability to it. And so if I could come up with something like that — just one of these simple little ideas, but that really grabs people — that's what I['d] do. And then I['d] do another one the next day, and another one the day after that.
35:52 So it's more about the process, then? — Yeah. And I really do like the process a lot. I have enjoyed my life creating games; I've enjoyed it a lot. And very few of them have really been my ideas. Occasionally I have gotten to be the central designer, but again, even then, that's misleading, right? Because games are not singular visions — almost never. They're always a collaborative medium. There's always inputs coming from everybody who's involved, and more so as the industry has matured, and you can have dedicated game designers
36:49 And dedicated animators, dedicated artists. Everybody's making their contribution, and the final product is never — you can never really say, "Oh, that's so-and-so's game," right? It's all of ours. — I think that's a good note to end on. — All right. — All right, thanks.
37:14 I want to extend a very big thank you to Glenn for agreeing to come on to my very small YouTube channel. He was extremely nice, and very excited and willing to do this interview, and it was super cool for me as well. I hope everyone watching enjoyed this, and I want to thank everyone also watching for all of the support the channel has gotten in the last few weeks. It's been amazing. We are about to hit 500, I think — subscribe — 400. It's like one of those — I'm bad at the numbers. But thank you, thank you very much.
38:08 And if you're interested in playing Rogue, you can find it real easily on My Abandonware, and you can play it in your browser if you don't feel like learning how to use DOSBox. And I believe it's also available on the App Store and/or Steam. You can find it easily. Go play it; it's a classic. It's still really fun to this day — surprisingly, actually, really fun, if you assume that it's just boring because it's [an] old game. But it's actually still really good; it holds up really well. I don't have an outro piece.