Ingmar: You’re certainly right that many who go back far enough do remember Infocom fondly. In your mind, what makes Infocom so special even after all these years?
Bob: I think the answer is that there is still power in the open-ended worlds that parser-driven adventure games created. Players felt that they could type anything and have a decent chance that the implementers might have a special message waiting for them.
I know that my biggest joy in writing the games for Infocom, and later for Legend, was the sense that I had a one-to-one relationship with the person playing the game. As a player, you might be faced with a particular situation and come up with an input that is so off-the-wall that you think you’re the only person in the world who would ever type that. If the game comes right back to you with a non-default response that shows the author anticipated the input, that is magic.
I remember with fondness the very late nights when I would be coding at 3:00 in the morning and saying to myself, “Someone is going to be playing this game at 3:00 in the morning, and they’re going to be feeling just as tired and halfway crazy as I feel right now. What are they going to type, and how can I reward them for it?”
That kind of one-to-one relationship between the player and author has passed from our game-making, and we are all the poorer for it.
Ingmar: When you and Mike Verdu founded Legend Entertaiment, what was the philosophy behind the company and how much was Legend still related to Infocom?
Bob: Legend was a very definite break from Infocom. At the beginning, we were petrified that Infocom would claim that we had stolen their parser. We went so far as to formally “black box” our game engine – which is to say that I wrote up a new design for the Legend parser and farmed its development out to a company that had had no contact whatsoever with Infocom. This, even though Duane Beck and Mark Poesch were fully capable of developing our parser themselves. We didn’t want to run the risk that Infocom could claim we had stolen their code.
That said, there were certainly some spiritual ties between Legend and Infocom. Most obvious was that we were doing parser-driven adventure games (albeit with graphics), and that our first several games were written by ex-Infocom authors (Steve Meretzky and myself).
Our philosophy was that we believed there was still a market for adventure games, as evidenced by the continuing success of Sierra and LucasArts. We felt that our games had greater depth because the point-and-click games were already losing the magic of the open-ended-ness that I mentioned earlier – the sense that the player could try anything.
Ingmar: How do you remember the mood at the company in the early days?
Bob: Legend was a company made up of very young people. I was the oldest at 35, and the next oldest was my partner Mike Verdu at age 24. Everyone else was younger still!
So we were young and enthusiastic and ready to work our butts off, which we did. We were delighted with our early successes, but always disappointed that we never were able to unseat Sierra or LucasArts.
Nevertheless, it was an adventure, and we all shared the feeling that we were creating something special with our games. Also, you have to remember that we were a publisher as well as a developer. That meant that we were responsible not only for writing the code and creating the master disk, but also everything that came afterwards – getting the disks duplicated, getting labels onto the disks, writing and printing the manuals, contracting for the box cover art, getting all the finished elements into the boxes, and even shrink-wrapping the boxes. We did it all with just a few people and a fulfillment house. We delivered pallets of thousands of finished boxes to our distributor, who then shipped them off to the retail stores where our games were sold.
So we were creating not just games, but a company. We did our own customer support, we read the comments on the warranty cards that players returned to us. We had real relationships with our customers. In hindsight, as the game industry has become larger, more hits-driven, and ever more impersonal, I realize I should have taken more joy from working with that small group of talented people who took so much pride in our games and in our company.
I should also note that after we shut down, virtually all of our people have gone on to stellar careers in the games industry.
Ingmar: Legend developed a versatile engine for its catalogue of adventures. How much were you involved in its creation? Like many other companies, you had to adapt to the new industry-changing paradigms like the popularization of the mouse and CD-ROM. What was the impact on the engine and the design of your games?
Bob: The lion’s share of the credit for our game engine goes to Mark Poesch and Duane Beck. As I mentioned earlier, I wrote the specs for our original parser and that was built out-of-house. But the parser ended up being only a small part of the overall engine, and Mark and Duane were the guys who built the majority of it. Michael Lindner also played a strong role, especially with regard to music integration.
Legend's multifaceted interface on display in Spellcasting 101: Sorcerers Get All the Girls
My role was to be one of the many voices concerned with the “look and feel” of the games. We wanted them to be parser-driven, but we also needed to respond to the rising use of the mouse. We decided that we would have several modes – an all-text mode for people who didn’t care about the pictures, a horizontally split-screen mode with graphics on the top half and typed input and game responses on the bottom half, for people who wanted to see the pictures but still wanted to type, and then a menu-driven mode for people who only wanted to click.
The key to creating this third mode was when I realized that once a verb had been selected from the left hand column, the list of verbs could go away to be replaced by a list of prepositions. That allowed us to have a fully clickable three token parser for inputs like >Hit ball with bat. I was pretty proud of that one.
Other than that, my engine work was mostly confined to working with Duane to continually refine the “smarts” of the engine to make it better at disambiguating player inputs. We hated messages like “Which ball do you mean, the red one or the blue one?” and developed quite sophisticated ways to figure out what players were trying to do. We also did a lot of work with the pronouns like “it” and “her” for example, to make interacting with the game more natural and to reduce typing.
Later, when we went to the inverse parser that we used starting with Companions of Xanth, I think Michael Lindner was responsible for most of the thinking that went into that interface.
With regard to CD-ROMs, they were probably the biggest contributor to the death of adventure games. I think it cost us about $125,000 to create Spellcasting 101. You could sell 30,000 copies of a game like that and make a handsome profit. With the advent of CD-ROMs and graphic-intensive games like Myst and 7th Guest, the industry started on a graphics binge from which it still suffers. Games started to cost more and more to make. Soon you couldn’t create a competitive game for under $1 Million, and the break-evens started to go to 100,000 units, to 250,000 units and more.
One of the biggest problems with the game industry today is that AAA games are outrageously expensive to make. Artists comprise between 2/3 and 3/4 of a game developer’s staff. The result is the “hits-only” business we see today, where games now cost as much to make as movies, and so few companies can afford to risk making them that innovation is stifled. That trend still continues with the release of ever-more-powerful consoles, and it’s only the rise of free-to-play and directly distributed games that make it possible for “smaller” games to potentially make a come-back.Continued on the next page...