EVERQUEST WAS CREATED in an era when teams were small and everyone pitched in where and when needed. That necessitated hiring developers who were dynamic and flexible. Scott Hartsman was all that, plus he brought a background in the type of game that inspired Sony’s MMO darling: He not only played MUDs (multi-user dungeons), he worked on some of the niche genre’s most popular.

Hartsman and I talked about his experience playing and writing for MUDs, how that transitioned to a role on the EQ team, the rapid pace of development that encouraged a culture of crunch (overtime), and how arguably his most important role, opening channels of communication between EQ’s developers and its community, set the stage for interaction 20 years into the future.

David Craddock: EverQuest evolved from MUDs, and you’re one of a select few on the dev team who, like Brad McQuaid and Steve Clover, played those games in their day. You also developed some. How’d you get started?

Scott Hartsman: I've been involved with online games for nearly as long as online games have existed. The first versions of online games were MUDs where players were playing with each other. It was very Dungeons & Dragons-inspired, the idea of virtual worlds where, essentially, there's a D&D game going 24 hours a day that you can play from home without having to organize friends. To those of us who grew up with D&D, it wasn't a big shock when MUDs went big. It was something all of us were dying to do.

The first game I ever worked on was a game called Scepter of Goth [also spelled Sceptre of Goth]. The maximum number of players was around 32 online at a time, all text, going room by room, fighting monsters and getting loot, just roleplaying. Honestly, the thing I really fell in love with back then was you were making new friends, and they could be anywhere. You actually needed people, and you were building relationships and online communities. That was the first exposure I had to the power of that kind of thing. I decided to go into game development because of that. I've only ever worked on online games because while I am a huge gaming fan, gaming with other people is what's really interesting to me.

I was an old Apple II geek from back in the day. That was back when everybody had to be something of a programmer to even play games, because state-of-the-art wasn't comparable to what we had now. I got an Apple II when I was a kid and played games on that, and then discovered the wonderful world of owning a mode. That's how I dialed into MUDs, and the one I was playing at the time was Scepter. I was around 14 when I first played that. I got incredibly, incredibly hooked on it. I made friends, and a year or two later ended up working for them when I was 15 or 16.

Online communities were much smaller, so everybody could get to know everybody. There were many 1,000 or 2,000 lifetime users on this commercial MUD. The owners, the game masters and designers, would sit online and talk with us. I made friends with those folks, and they brought me on. They were looking for people who could help out with all kinds of things. This was back in the day when there was really only one role on text games: programmer. The roles of designer, customer service, writer--those didn't exist. Programmers built the software, and they would also create monsters, write room descriptions, create spells, and answer customer questions, run credit cards to do billing, moderate forums. I was one of those people. I started out as a writer/content creator/customer service agent/billing operator, I guess.

Craddock: Since MUDs were driven by text, what was the writing process like on those games? Did writers stake out a purview? Did everybody pitch in here and there?

Hartsman: The vast majority of writing we did was for areas. Since it didn't have any graphics, the only way you could communicate where players were and what they were saying was through words. Most of the job was writing areas. You wouldn't just design the areas in terms of layout and how they connected, what monsters went where, and what loot existed; you were describing the area in words.

There was nothing more intimidating than designing a new area, and you've suddenly got 200 new rooms, and realizing, "Holy crap, I've got to do a lot of writing." It's the blank page problem times 200, and that happened anytime we added anything new to the game.

Craddock: Did you emphasize descriptive prose, or did you come to find that less was more?

Hartsman: That's a good question. Keeping in mind that I was 15, back then it was, how much flowing text can you cram into this to give people a sense of the place? Now, I can definitely say less is more. There were seven of us in this role. We were called DMs, and we had seven people constantly filling up new worlds, adding new events and spaces all the time.

It was fun, especially because the people in charge were programmers. They didn't get into much of what we did. I would say to my boss, "Hey, Chris, I want to add an area that's a keep. Here's three sentences of the story." She'd say, "Okay. Go for it." That was the level of oversight.

Craddock: How'd you transition from Scepter to Gemstone?

Hartsman: The Gemstone project, and the company around it, formed in the later days of Scepter as that company was going out of business. That was called Interplay Protocol. For about six months to a year, we'd been running the game on a national network called Timenet before the Internet was the Internet. It was these big networks that others had made, and it was how you allowed people from different area codes to connect and pay them.

They had that service up and running for around $3.60 per hour of play. For its time, it was a big service, but it wasn't sustaining itself because a lot of that money went to the networks. Online services like CompuServe and Genie existed, and they had much larger audiences, a couple of hundred thousand people. They decided to get into games, and a company called Simutronics formed. Five of us who'd been working on Scepter came over.

A lot of Gemstone was built so people who were Scepter DMs would be able to work within it easily. That also gave us the power of a scripting language. That was my first introduction to online-game programming. Not only did we have to do the loot, items, descriptions, and drilling connections between rooms, we also had scripts. We could do things like dynamic events and exceptional behavior for creatures: We could have creatures respond to you waving at them, for example. We could do that ourselves without having to bother the programmers.

Craddock: Do you feel the core, the driving attraction of a text-based game like Scepter and Gemstone, isn't so different from what makes more sophisticated games like EverQuest appealing?

Hartsman: Oh, yeah, the core is very similar. You look at EverQuest, and that was launched by a group of people who'd been playing MUDs. They went, "We want to do this with graphics, because holy crap, wouldn't that be amazing?" And they were right.

Craddock: What did you learn from working on MUDs, these text-only games, that you were able to apply to EverQuest and other graphics-based games?

Hartsman: The importance of keeping the live game interesting. That's something we were doing as far back as the first MUDs. It wasn't just: launch a game, run it, and let it do its own thing. You are there to provide entertainment day over day, week over week, month over month, year over year.

Craddock: What brought you to Sony?

Hartsman: I fell in love with programming as a scripter in Gemstone, and I was very fortunate in that there were three programmers working on the game who were friends and extremely generous with their time. All three of them took me under their wings in different ways, and helped me become a C programmer. I kept running into limitations of what our system could do, and after months of that I realized I'd need to be a programmer if I wanted to bring even cooler things to life.

I ended up quitting that job to get an engineering degree. I worked part-time in games during college, came out of college, and ended up working in the business world for a few years. One of the interesting things I worked on was, I worked with the team who took all the court system records in LA pertaining to the O.J. Simpson case, digitizing them, and doing online electronic submission. I wrote the client that allowed that to happen. We were doing this all during the O.J. trial, so we got to see everything as it came in. Just a bit of random trivia. [laughs]

Eventually, that got boring. A friend of mine who was the project manager at Genie, which ran Gemstone, called me out of the blue and said, "I'm working at a different company called Interplay." This was Brian Fargo's Interplay. "We're doing an online games division. It's called Engage, and I kind of need your help. Would you be able to write trivia games?" I said, "Well, trivia games aren't hard. Client, server, graphics, Windows--no problem."

They contracted with me while I lived in Wisconsin to write a couple client-server games. After that, they asked if I'd consider coming out to work with them full-time because they needed more games launched. I did that, launched more games, did the Windows and Internet port for WarCraft II, which we had the license to. That was fun. We got to about 11 games up and running, and these primarily ran on AOL and our own platform. I ran that company for a while, and that phase of Internet gaming came to a close: Engage, TEN [Total Entertainment Network], and Mplayer were the big names [in online gaming] in that generation.

After that, I worked with a group of people who were pitching Sierra on doing a Lord of the Rings online game. I don't know if you know the history of those online games, but there have been three or four projects over the years. This was prior to the one Turbine launched. We worked on it for a year before Sierra spiked it. The woman in charge at the time would eventually become the Sony-side producer of EverQuest for its launch. She reached out to Smed [John Smedley] and said, "If you need a strong programmer, you should probably hire Scott.] I talked to Smed, met him, and he brought me on.

Craddock: What was your first experience playing EverQuest? Had you played it before joining the company?

Hartsman: When I worked at Engage, we were primarily a publisher of games. The only games we'd created were the ones I'd made, the trivia games. In the lobbies of the trivia games, they all had chat channels--because everything did back then--and I'd actually created them as a MUD where you could walk through different rooms. We were trying to bring the basics of MUD concepts into genres outside of MUDs.

Among the game teams at Engage, we absolutely wanted to make bigger, connected, online fantasy games, but that just wasn't what the company was about. It was primarily a publisher. I had started a company with a group of people whom I'd worked with at Engage, and one of those guys said, "Have you heard about this EverQuest game?" It had just been announced. I said, "No, but I've done my time in online fantasy games. I know what those do to me: I won't do anything but play them. I should probably stay away."

I went to E3 that year, and another friend said, "You really need to check out EverQuest. They're demoing it, and holy shit--it's the game we talked about years ago." Again, I said no. After work, another coworker, a guy I worked with at our startup, bought it for me. It was on my desk waiting for me in the morning, and I said, "Well, I guess I've got to try it now." I played it very, very hardcore. We had a group of friends playing, met new friends online, we formed a guild, figured out how to recruit good people and make friendships, made it through the game and the raids. Then I slowed down on playing and ended up working for them.

And it's so funny. If you look at players from that generation, so many of us are working in prominent positions at different companies now. One of the guys in my guild, I later hired him to be a designer on EverQuest II. He's now the executive producer of Rift at my former company, Trion, which we just sold. His brother is the exec producer of Age of Conan at Funcom. There are so many people from that generation in various parts of the games industry.

Craddock: How would you describe the EQ team's culture when you joined?

Hartsman: I came on about two years after launch. The team at that time was around 50, maybe 55 people. The engineering side was around five people. It was small. The culture was one of, "Get the job done, whatever it takes." It was my first real encounter with a crunch culture. It's something that I'm glad I had to live through because I was crazy-passionate about the game, so I didn't mind. But there were times when we'd work 12-hour days, 16-hour days, and come in on weekends also. We sustained that for quite a long time.

I'm glad I lived through it, but also glad it's in the rear-view mirror. Beyond that, I should say it was the single most passionate team I've had the pleasure of working on. Everyone was there because they loved this game. There were vastly easier jobs, even in the same company, if you wanted that, because the studio had other games going on. I threw myself into the fire because when I came in with experience doing bigger projects, they had wanted me to help get their game called Sovereign on track. That was the MMORTS they'd been making.

I was a huge EverQuest person and saw what the team was going through. We had people on Sovereign working normally to slightly longer hours, and then people on the EverQuest team getting ground to dust. I asked repeatedly if I could go help EverQuest, and they finally said, "Yeah, we could use what you do."

Craddock: I wondered about that. I remember EQ expansions popping out at a pace of, like, two a year, at least one a year. What was that like internally? Was there ever a cool-down period between those projects?

Hartsman: No, no. What would happen is, you'd get to launch, and then things would get harder. As launch stresses happened--bugs and other problems--you'd fix those, make more content to keep the live game going, and when it dials down from incredibly insane to only slightly insane, that's when you began peeling off people to start filling out the next expansion.

The only teams that had what you would call a demarcated production cycle was the art team. Art has a long lead: You have to plan it out far ahead of time. When the art is done, that's when design implementation begins. You don't have live, emergency art bugs in the same way you do coding bugs or design bugs. The artists worked crazy overtime as well, but theirs was at least somewhat more organized into something you could call a cycle.

Craddock: What did your role as technical director entail?

Hartsman: We didn't have the role of live producer back then. A live producer these days is someone who's organizing live updates, doing shipping, making sure patch notes are in order, making sure things get through QA, and so on. The team didn't have that role at the time, so I kind of took it on.

When I came in, the QA team was four people. The QA that was happening was: developers would ship a game, and the expectation was QA would play the game, send bugs, and someone would fix them. There wasn't any concept of doing pre-submits to our QA team to get passes or fails back to fix things before setting them live. I did a lot of instituting processes like that, as well as managing the engineering team and programming gameplay myself.

One of the least technical parts of that job was I started toning down some of the egotism that would happen when we'd present ourselves to the outside world. The original EverQuest slogan was, "You're in our world now." I was the first person to post a set of patch notes saying, "Thank you for making our world yours," which was a 180-degree turn from a lot of the more antagonistic [messaging]. I wanted the game to feel like less of a player-versus-developer experience, which had been the case for many years, and start putting more emphasis on, "We're here with you. We're here to entertain you, and we need to earn your business every day."

In the era when I was there, we went through a management turnover. Brad [McQuaid] was gone. Other producers had left. There was a vacuum, and we needed people to step up and take on things way outside their roles. A lot of us did that.

Craddock: How did that player-versus-developer culture manifest in the game, or through communication?

Hartsman: There was an antagonist tone between the community folk at the time, and the players. It was something that grated on me, and on a lot of the developers, because they didn't want that. Not just the tone, but also, the accuracy of what was being communicated to players was not where it should be. I think that hurt the team's credibility.

For instance, the company would make statements to players about the state of certain things in the game: whether they're working, whether they're not working, and so on. Those statements ended up not being true. The people saying them thought they were true, but they would say things like, "Alchemy is working 100 percent, and you players just haven't figured it out. Maybe you should play better." Meanwhile, players were submitting reports like, "No, this system has been broken for years. You guys need to fix this." The company would come back and say, "No, it's working fine. Get good." Of course it had been broken the entire time.

That kind of thing: repeatedly, constantly, month over month. That kind of tone: "We're right, and you, mere players, are wrong." As a person who not only makes these games but plays them, it grates on me when you treat your players as less than you. I don't think that's the right approach.

Craddock: Where do you think that egotism came from?

Hartsman: Community management is a skill. It's not something just anybody can do. It takes practice, empathy, and learning your limitations. Those were early days for defining what was a community manager in the first place. The person who was doing that job at the time eventually went on to be a very good programmer, but was a horrible community person, and he'd be the first person to say that.

It's just that we as an industry didn't understand community management. Relating to human beings was a valued skill, and needed to be on those games.

Craddock: What were some other ways you tried to facilitate better player-developer interaction?

Hartsman: As a big gamer and a fan of EverQuest myself, I would read bloggers constantly. I had a roll of at least 100 MMO bloggers I was reading at the time. Generally, there wasn't a lot of credibility or credence to what some of the bloggers were saying, but some of them were incredibly smart and offering great insight into the state of the game and what we could do better.

Steve Danuser is at Blizzard now, and his alias back in the day was Moorgard. He was one of the first bloggers who made me say, "This guy is really sharp. We should listen to him." Sony ended up hiring him to work on EverQuest II as our community director. A lot of my job was finding ways for smart players to be heard. I used to hang out on guild raider forums--every guild had their own set of forums--and listen to their complaints so I could bring things that bothered them to greater visibility on our team.

I wish we'd had dozens more people so we could have got even more work done.

Craddock: Every game, especially online games, reach a peak. EverQuest is still going strong, but I wonder what you saw as the height of its popularity.

Hartsman: That's something I'm proud of: We were the team that took it to its peak. This is a fun but long and complicated answer.

One of my biggest regrets of being on that team was, we had around 50 to 60 people and constantly had to be cranking out expansions so we were always selling things. At the same time, we were working on the first international versions that were being made. This game was, as a piece of software, never meant to be localized. It was meant to be a single-language game. All the English was hard-coded into the game, so the engineering team had to spent the better part of a year making possible what our bosses were asking us to do: "Ship it in Korea, Japan, China."

The amount of time and resources that took was extreme. We should have had four to five times as many people as we did, because then we'd have had more people working on the live game.

That answer sets the stage for: We had not been paying attention to the live game as much as we should for nearly as long as I was there. We had a particularly bad month. We all saw reports every day. We knew the subscription numbers and so forth. The game was dropping off, and dropping off pretty badly. We got the OK to do stuff to make players happy: the features they've been crying out for, things we could be doing better.

We were in heaven, because that's what every single one of us wanted to be doing in the first place. That's why you get into this job, right? Not to ignore your players and go work on other projects because corporate masters demand it; to make them happy.

We did that. We fixed bugs. We added features like global chat. That update, and the events surrounding it, had almost no paid marketing. It was known about largely because of community hype: Us telling people, "We fixed this thing, we added this thing." It skyrocketed the game to its peak, around 480,000 subscribers in a given month. This was just a patch. We didn't even name it.

What made me happy about that was we finally had the ability to say: "By making players happy, we grow." But in the face of an international corporate behemoth... The people at SOE all the way up to Smed truly understood that if we wanted to grow, we needed to make players happy. The issue was further up the chain.

Players would report a bug, and if it was live, the chances of someone being able to go back and fix it was basically nil because of our shipping schedule. Just having the ability to put developer time into live issues bothering players was viewed as mana from heaven.