Unlike many early peers such as Sierra, who began publishing at least as many third-party titles as titles they developed themselves in the wake of their first success, Infocom throughout their independent existence prided themselves on keeping everything in-house. Every game they released was written right there in their Cambridge offices by their own dedicated little band of Implementors. But Activision’s pressure to release many more games every year in the wake of their acquisition of Infocom finally changed that in late 1986. Infocom desperately needed more Imps to meet Activision’s craving, but weren’t in a position to pay for any more full-time employees. While they could continue to promote eager testers and programmers from other parts of the company, always their most fecund source of new blood for the Imp pool, there was an obvious point of diminishing returns at play there as well: every tester they promoted to Imp meant one less person to test all of those new games that were now coming down the pipe. After promoting Amy Briggs, who was destined to go down in history as the last person ever to become a full-time Imp at Infocom, it was time to beat some other bushes. Like so many companies before and since who couldn’t afford real employees, Infocom filled the labor gap with contractors willing to work remotely for an initial advance against royalties. After all, modems — even expensive, state-of-the-art ones like the two they now hung off their venerable old PDP-10 — were a lot cheaper than employees.
This change represented, I shouldn’t neglect to emphasize, an enormous attitudinal shift for Infocom. They had never before paid anyone whatsoever on a royalty basis; all of the Imps had always worked for a flat salary. In retrospect we can see this instant as marking the beginning of a sea change in what Infocom really was, from a true creative collective making games together to a mere label under which Activision released any game that was even vaguely adventure-like or story-oriented. It would take a few years for that change to reach its full fruition, but the process started here.
Having elected — or been forced — to make this change in their way of doing business, the natural next question for Infocom was who these new contractors should be. One signee was an enterprising Californian named Bob Bates, founder of a tiny would-be adventure developer he called Challenge, Inc.; I’ll be telling his story in a future article. For other contractors, Infocom turned to some old friends, former Imps who had left the fold. With time and resources at a premium, that made a lot of sense: they already knew ZIL, knew how the Infocom development process worked and what would be expected of them as writers and designers.
Infocom thus reached out to Mike Berlyn, who had left the company back in early 1985 to found a design studio of his own, Brainwave Creations, with his wife Muffy. Having already shipped their first adventure game, Interplay’s text/graphic hybrid Tass Times in Tonetown — and through Activision as publisher at that — they were now sniffing around for a home for the new dimension-bending comedic caper they had on the boil. Much as Berlyn’s mercurial nature could make him difficult to work with at times, everyone at Infocom had always liked him personally. If they must work with outside contractors, he certainly seemed like one of the least objectionable choices. They got as far as signing an initial development deal before things fell apart for reasons that to my knowledge have never been fully explained.
Infocom’s other attempt to get the old band back together again would prove more fruitful even as it could also seem, at least on the surface, a much more surprising move. Marc Blank, you see, hadn’t just quit Infocom a year before. After months of squabbling with Al Vezza’s board over Cornerstone and pretty much every other decision they were making, he had been summarily fired.
Blank moved on to, of all places, Infocom’s erstwhile suitors Simon & Schuster, where he became “Vice President of Computer Software Development,” his responsibilities to include “artificial intelligence, expert systems, sorting out new technologies like optical and disk storage.” The brash young Blank, however, soon found he didn’t fit all that well within the conservative old halls of Simon & Schuster. He left what he calls today a “terrible” job within a few months.
Blank’s next stop would prove more extended. He moved to California to work with a company called American Interactive Media, a new corporation with roots in the music industry who were now so closely associated with the Dutch consumer-electronics giant Philips as to blur the line between independent contractor and subsidiary. Philips had initiated a project to bring the brand new technology of CD-ROM to consumers via a set-top box for the living room, and American Interactive was to create games and other content to run on it. In the long run, this would prove another frustrating experience for Blank; Philips’s gadget wouldn’t finally be released until 1991, an astonishing seven years after the project had been started, and for all sorts of reasons would never take off commercially. For the time being, however, it felt fantastic. A guy who loved nothing better than to take a Big New Idea and give it practical form, Blank felt like he was taking interactivity to the logical next step after Infocom, working not with plain old text but with a whole rich universe of multimedia potential at his fingertips.
While he was about inventing the future for Philips, though, he wasn’t above doing some more work for his Infocom friends from the past. And Infocom was very eager to work with him again. The people he had pissed off enough to get himself fired were largely gone from the newly slimmed-down, games-only edition of the company. Truth be told, most of the people still there had agreed with every sullen argument and veiled jab he had ever delivered to Al Vezza and his cronies. They called, Blank said yes, and Border Zone was born.
It would prove a classic Marc Blank project. Never a gamer, he claims that to this day he’s never played a single Infocom game, other than those he wrote himself, to completion. Nor does he have much intrinsic interest in writing or game design as disciplines unto themselves. During his time with Infocom and even before, when working on the original MIT Zork, he preferred to see himself as the wizard behind the curtain, crafting the magic behind the magic, so to speak, that enabled people like Dave Lebling and Steve Meretzky to do their thing. It was Marc Blank who tinkered endlessly with the parser in that original Zork, taking it from a clone of Adventure‘s primitive two-word jobber to one that wouldn’t be fully equaled by anyone else for well over a decade. It was Blank who came up with vehicles you could ride in and characters you could talk to. It was Blank who sat down with Joel Berez and figured out just how Zork could be chopped up and delivered onto microcomputers via a cross-platform virtual machine, an event that marks the beginning of the real story of Infocom as a maker of computer games. And for his pièce de résistance, it was Blank who radically upended people’s very ideas of what an adventure game could be with his interactive murder mystery Deadline, not in the name of art or literature but simply because he found doing so such a fascinating technical exercise.
When Blank wrote and designed a game, he did so essentially as a demonstration of the one or more Big New Ideas it contained, with the thinking that, new technology now to hand, better writers and designers than him could make something really cool with it. Selling his own skills in those departments short though he may have been, Blank manifested no innate need to create in the sense of crafting a single unified work and stamping his name on it as its author. He was perfectly happy to just help others with the interesting technical questions raised by their own would-be creations, as when he built the system for Mike Berlyn’s Suspended that let you play by issuing commands not to a single avatar but to six different robots, each with its own unique outlook and capabilities. Tellingly, Blank authored — or rather co-authored, with Dave Lebling — his last game during his tenure as an Infocom employee, Enchanter, more than two years before he left. Once other Imps were readily available to implement games, he was content to let them while he did other interesting things.
If Blank was suddenly eager now, three years after Enchanter had been published, to write a game again, it could only mean that he had another very compelling Big Idea which he wanted to put through its paces. This time it was real time.
On the surface, it was far from a new idea. As far back as 1982’s The Hobbit, games from other companies had incorporated a timer such that, if you sat too long at a command prompt without doing anything, the program would process a turn as if you had entered a “wait” command, presumably in the name of keeping you on your toes and adding a dollop of urgency to the experience. In addition to The Hobbit and its descendants, Synapse Software’s BTZ engine (“Better than Zork,” although it really wasn’t) had also used this mechanic — a somewhat odd choice for a line which otherwise strained to promote itself as even more cerebral and “literary” than Infocom, but there you go. On the whole it had proved little more than an annoyance, here and everywhere else it had turned up. The actual games it sat atop did nothing of real interest with it. They weren’t actually real-time games at all, rather turn-based games with a chess timer grafted on.
Blank’s idea, which he worked with Infocom’s systems programmers to build into the new version 5 Z-Machine, was to do something much more sophisticated and thoughtful with real time. He had always been deeply interested in creating more dynamic, realistic environments, in pushing back the boundaries of Infocom’s games as simulations. Consider what made him find Deadline so exciting back in 1982:
You’re in this world where all these things are going on. People are doing things. They stop here, talk to someone, go here. Some things would change depending on what you did, but you could sit in one place and watch people come and go. I loved that the world was alive. Instead of exploring a dead world, you’re in a dynamic world with other things going on that you can impact, and what people then do will change, and that will then resonate out, etc. I had no idea how to do that. I had to make it up as I went along. That to me was the fun part.
Now, in Border Zone, the world would live and change, often completely outside of your view, even as you read, thought, and typed your commands. Blank would remove the artificiality of the turn-based structure and create a truly living world behind the words that you read on the screen. You might be in a train compartment trying frantically to hide some key piece of evidence to avoid arrest. As you do so, a guard is following his own schedule, moving from compartment to compartment in the train, getting ever closer, all unbeknownst to you until he bursts into your cabin. If you do manage to get everything sorted in your compartment before the guard turns up, you’re left to wait — literally to wait, sitting there watching the seconds tick by on the clock on your real-world wall, knowing some sort of security check must be coming, wondering if you hid everything well enough. Nothing quite like this had ever been done before. It absolutely teemed with complications, ran contrary to some of the most bedrock assumptions in Infocom’s development system. It would be a massive technical challenge to get working correctly. But then, massive technical challenges were what Blank lived for.
As usual for Blank, the fictional premise, plot, and puzzles were chosen after the Big Idea, in answer to the question of what sort of fiction would demonstrate said Idea to best effect. Thankfully, and again as usual for Blank, what he came up with proved far more compelling than one might expect from a designer so eager to declare himself so uninterested in game design. He settled on spy fiction. Stu Galley had actually already tried to craft a spy thriller for some six months between implementing Seastalker and Moonmist, but had finally given up on it as just too complex to bring off with Infocom’s technology at that time. Now, though, Blank thought he might have cracked the code. Spy fiction should make an excellent fit for a game of nail-biting real-time tension. And it certainly didn’t hurt that it was enjoying considerable commercial success at the time: countless readers of writers like John le Carré, Robert Ludlum, and Frederick Forsyth were enjoying a final spot of classic Cold War intriguing in a world that was soon to change in ways that absolutely no one could ever have predicted.
There’s a bit of the typical Infocom in-jokery, now getting more tired than not, in “Frobnia,” the name of the fictional Eastern Bloc country where much of the action of Border Zone takes place. The principal feelie, a tourist brochure and phrase book for Frobnia, also plays for laughs of the “in America you break law, in Soviet Russia law breaks you!” stripe, complete with poorly translated English, and that’s okay because it’s actually pretty sharp and funny stuff. But otherwise Blank plays it straight, and in the process does a good job evoking classic spy thrillers like Day of the Jackal. Border Zone is, like Nord and Bert, a segmented game, telling a single story in three parts from three points of view; breaking it up like this helped to keep the complexities of this real-time, player-responsive world from becoming overwhelming. While Blank didn’t lift much if any text or code directly from Galley’s previous stab at the spy genre, a game that was to be called Checkpoint, the plot of Border Zone‘s opening sequence in particular bears a marked similarity to Galley’s outline: “You, an innocent train traveler in a foreign country, get mixed up with spies and have to be as clever as they are to survive.” In the first part, then, you play the role of an ordinary American businessman who’s entrusted with some vital documents by an American agent on a train that’s about to cross the border from Frobnia into the ostensibly neutral but Western-leaning (and equally fictional) nation of Litzenburg. In the second, you play the American agent himself, who, having palmed the documents on the businessman, must still escape his KGB pursuers. And in the third, you play a KGB agent with secrets of his own on the scene of the attempted assassination of the American ambassador to Litzenburg — an assassination pointed to by the documents. It all feels appropriately morally murky, and is about as intricately plotted as you can reasonably expect from a work with a fraction of the word count of the novels that inspired it.
Still, and as Blank would no doubt agree, the most interesting aspect of Border Zone is what it does technically and conceptually within its fictional premise. Going yet one step beyond those early mysteries, this is the most complex and responsive world simulation Infocom would ever manage, and, like Plundered Hearts, one of their relatively few games that comes close to delivering on their marketing’s promise of interactive fiction as being “like waking up inside a story.” Most interactive fiction, then and now, is fixated on things, on rooms and their contents, to a degree that can feel downright strange to the uninitiated. Border Zone steps away from that fixation to focus not on what things are in the world but on what’s happening there. The old rooms-and-connections model of geography is still there below the surface, but it’s radically de-emphasized. Many rooms have no set-piece descriptions at all to separate them from the story that’s happening to you and all around you, giving the text a sense of urgent flow. Consider, for example, this extract from the second part, from which you could remove the command prompts to end up with something that would read pretty well as an avant-garde second-person novel (a little Italo Calvino, anyone?).
You are standing at the back door of the hut, which can be circled to the northwest and the southwest. On all other sides lies the forest. A small window in the door gives a view into the house.
>n
You walk around to the north side of the hut.
Two men, presumably from the automobile parked at the end of the roadway south of the clearing, are in quiet conversation with a man, presumably the owner of the hut, who stands leaning against the closed front door. They seem to be lecturing him about something, for he speaks little and nods often.
You watch as the lone guard returns to the group. He appears relieved that he has nothing to report.
The dogs are no closer, but now they seem to be off to the south.
>n
A branch falls from a nearby tree, startling you briefly. You turn back and press on through the forest.
>n
You can hear a pack of dogs off to the south.
You continue through the forest, until you come to the edge of a wide clearing to the north - this is the border zone. From atop three guard towers standing in a line from east to west, searchlights play across the zone, brightly illuminating everything in their path. On either side of the towers are tall
fences, running parallel to the border, making a direct assault all but impossible.
>n
You run across the open field at a good clip, though you are hampered by slick- surfaced shoes. You're past the halfway point, but wait! The light from the rightmost tower is heading right at you! You freeze, and consider turning back, but it's too late. The searchlight is upon you now, and before you can react, the night is filled with the sound of wailing sirens.
**** You have been arrested ****
It’s not that geography isn’t important, but the scale is shifted. As you duck from hiding place to hiding place trying to avoid a searchlight’s beam, the details of each piece of snowy tundra where you crouch aren’t so important, but where you are in the bigger picture — specifically, in relation to that questing beam — certainly is. Map-making is, as one would hope, completely deprecated. Where necessary, the feelies provide maps that are good enough to orient you to your environment, and the game itself also strains, within the limitations of its text-only presentation, to give a visual overview of the situation in the status line when you’re doing things like dodging guards and searchlight beams.
But what’s most important of all is the other people in the world, especially the ones with machine guns trying to hunt you down. The “puzzles” in this game aren’t really puzzles at all, but rather grounded, realistic situations that you need to come to understand and manipulate to your advantage. Suffice to say that there are no riddles or sliding blocks in this one, folks.
All that said, there’s an unavoidable irony about Border Zone: although this approach was inspired by the desire to make a scenario that would be a good match for the real-time component, just about everything it does could have been done just as easily — and, I would argue, just as successfully — using a conventional turn-based approach. In short, I’m not sure how much real time really adds to the experience. As with so many technically esoteric or ambitious touches in games, the real time in Border Zone feels ultimately more interesting for the programmer than it is for the player. Certainly there remain quite a number of unsolved problems. Border Zone‘s visual presentation, for example, leaves a lot to be desired. For all the new capabilities of the version 5 Z-Machine, its display is still limited to two “windows”: a static top window, generally used for the status line and other persistent information like Beyond Zork‘s room description and automap, and a scrolling bottom window for the main body of a game’s text. The ideal setup for Border Zone would have the command prompt in a static window of its own below the scrolling text — notably, this is the layout used by Synapse for their pseudo-real-time games — but this was apparently one step too far for Infocom’s programmers. Instead the command prompt is still in-lined with the rest of the text, meaning that when things happen around you outside of your direct prompting your command is rudely interrupted to tell you about it; then the partially completed command is printed again and you can finish what you were trying to do. It works, but it’s pretty ugly, not to mention disconcerting until you get used to it.
Border Zone doesn’t quite know how to make the command prompt work together with real time in an aesthetically pleasing way.
While that could be fixed easily enough with more advanced screen-layout capabilities, other problems with Border Zone feel more intractable. Given the careful reading that interactive fiction requires, you’re likely to be chronically short on time the first time through a scenario, then twiddling your thumbs waiting for things to happen on subsequent tries. And, believe me, you will be playing through each of these segments much more than once. Much like Beyond Zork, Border Zone promises one type of experience only to deliver another. The heavy emphasis on simulation and dynamism would seem to imply that this is an emergent experience, one where you can have a different experience every time out. Actually, though, that’s not the case, largely because, for all the world’s complexity, there’s no randomness to it at all. Everyone will follow the same patterns every time — complex patterns, yes, but patterns nevertheless — unless you interfere with them, which in turn will only set them on another deterministic course. This leads to a mode of play that isn’t as different from the interactive mysteries Blank earlier pioneered as many a player might wish.
In other words, this is another try-and-try-again game. Essentially you start playing by doing what seems the natural thing for a character in your circumstances, until you die or get captured or otherwise fail. Then you analyze the situation, come up with an idea as to how you might avoid the negative outcome, and try again. Rinse and repeat. Border Zone isn’t as punishing as the mysteries can be because each of its segments is so compressed, limited to no more than ten or fifteen minutes of real — i.e., wall-clock — time. It’s here, however, where the real time component can also become actively annoying. When you know the series of steps you need to follow to get to a certain point, you want to be able to “wait” in the game for each decision, not be forced to literally sit around waiting in real life. (It is possible to “wait for” a specific number of seconds, but that can be tough to plan out, not to mention deadly if you get it wrong.) And when you come to one of the junctures that require really precise timing, where you need to hit the enter key to submit your command at just the precisely right instant, it can be extremely frustrating when you, as another superspy would say, “miss it by that much” and have to start again.
Each sequence in Border Zone feels like a single bravura action sequence from a good James Bond flick, which is, one senses, exactly the effect intended. What with all the learning by death, playing can feel oddly like choreographing the delicate ballet that is such a sequence, trying again and again until you get all the drama and thrills just right. One notable side effect of the extreme time compression is that Border Zone, even with all three of its sections taken together, is one very short adventure game. Its rather expansive 175 K story file, which would seem to promise a much longer experience, is padded partially by an in-game InvisiClues-style hint system and partially by the complexities of the real-time system itself, but most of all by the overhead involved in implementing a world of depth rather than breadth. Everything was a trade-off in game development in Infocom’s day. In this case, Blank has dramatically increased your scope of possibility and the number of moving parts in the world around you at the expense of game length. Most of the things you might think to try in Border Zone work logically and have logical consequences in the context of the world and its other actors, even if most of them must inevitably be the wrong things, things that ultimately lead to failure. And yet, even duly accounting for the many replays that will be necessary to finish each sequence, it’s very difficult to spend more than three or four hours on Border Zone. That was a major problem for a commercial computer game selling for $30 or more. It’s not hard to understand why gamers would be put off by the entertainment-per-dollar ratio at play here, not hard to understand why publishers usually opted for longer if sketchier experiences over an intricately tooled Swiss watch of a game like this one.
How much its extreme brevity had to do with Border Zone‘s poor sales reception is, given everything else that was going so wrong for Infocom at the time, hard to say. Certainly they tried hard to make it accessible to as many customers as possible. Despite running under the new version 5 rather than the version 4 Z-Machine of Nord and Bert, it became the second in Infocom’s “LZIP” line of larger-than-usual games that were nevertheless shoehorned into the Commodore 64. Still, sales were bad enough to give Border Zone the title of worst-selling all-text Infocom game in history: less than 12,000 units. Thus it wound down Infocom’s demoralizing 1987 just as it had begun, by setting a sales record of the wrong type. It doubtless didn’t help Border Zone‘s cause that it was released in the immediate wake of the much higher profile and more enthusiastically promoted Beyond Zork.
One of Infocom’s least remarked and, one suspects, least played games today, Border Zone deserves better than that fate. I’m not sold on the case it makes for real-time interactive fiction, and little surprised that that avenue has gone all but completely unexplored through all of the years of non-commercial experimentation that has followed Infocom’s demise. But Border Zone is much more than just a failed technical experiment. Even if real time were taken out of the picture entirely, it would stand as an experience able to get the pulse pounding and the juices flowing, something that textual interactive fiction isn’t exactly known for. It’s also a very solvable game, and one where every challenge truly is part and parcel of the story you’re living. That, again, is something not a whole lot of interactive fiction, vintage or modern, can lay claim to. I highly recommend that you give it a play, and experience yet one more utterly unique Infocom game for yourself.
(Sources: As usual with my Infocom articles, much of this one is drawn from the full Get Lamp interview archives which Jason Scott so kindly shared with me. Also useful was the April 1986 Questbusters.)
