Ah, another year, another Interactive Fiction Competition. It’s the seventh year of this competition. The idea is simple – sometime in September people start writing text adventures – short affairs, nothing too elaborate – and at the end of the month submit them to the organizers. At the beginning of October, the entries are posted on the IF Competition web page and votes are submitted by November 15. The winner is announced shortly thereafter. The prizes run the gamut from cash to Cheerios coloring books, and all are donated. In addition, there’s always the fame and prestige that comes along with the riff-raff on newsgroups remembering your name. Really the competition is more a nifty long running idea than a bloodthirsty match.

Interactive Fiction (IF) is the name Infocom gave text adventures back in the 1980’s, and it’s the one that stuck. IF consists (usually) of a text parser. You are presented a text description of your surroundings and you type in what you want to do, i.e. take the knife, kill the troll, go north, etc. A holdover from a less technologically inclined time and one of platform ambiguity, it exists today for estoteric reasons, as well as romantic ones. IF is to the interactive industry what books are to the entertainment industry – seen by many as old fashioned and useless, viewed by an elite mass as neccessary art.

In the mid 1970’s, four students at MIT, Marc Blank, Bruce K. Daniels, Tim Anderson, and Dave Lebling, got together and attempted to put together a “Computerized Fantasy Simulation Game” using a language called MDL. The idea was to take the elements at hand and see if a “game” could be authored using them. Originally titled Dungeon they renamed it Zork and showed it to others on their system through the wonders of mainframe. It was a huge hit, which then struck the authors with an idea.

It was a given that they wanted to form a software company out of college. Former MIT alums had formed the company Lotus (which was then on top of the spreadsheet market with Lotus 1-2-3), so why couldn’t they? However, there were a few barriers to personal software at the time. First and foremost was the platform issue. Today when a company writes a piece of software the platforms at hand are Windows, Macintosh or Linux. However in the late 1970’s-early 1980’s, there were tons of players on the field. The Apple II, IIe, IIgs, Commodore VIC-20, 64, 128, Amiga, Atari ST, etc. Machine languages (like Assembly) were the order of the day. Portable languages like C were a new thing and didn’t take into account real world concerns, plus they were slow. When you wrote a program, you had to choose carefully. However, these guys had a different idea – create a ficticious computer, the Z-machine, and then write a program for each system that emulates it. Then write your program to run on the Z-machine and your one program will work on all the platforms that there is a Z-machine interpreter for. This notion was the precursor to Java.

Therefore, coming out with business software for all the platforms of the day and having one product work across all platforms seemed to be a brilliant idea, were it not for the other problem – money. These guys needed the time to make this idea work for a massive application, but they didn’t have any capital to work with. However, they did already have a game, and a popular one at that. That’s when it hit them – release Zork as a game for all the platforms of the day. What better way to test the Z-machine notion and generate revenue at the same time?

Zork‘s size was too large for the personal computers of the day, so they split it up into three games, Zork I, Zork II, and Zork III under the label of Infocom. They were all smash hits. This of course led to more games, too many to list here, but notable titles included The Hitchhiker’s Guide to the Galaxy (actually written by Douglas Adams, no less), Leather Goddesses of Phobos, Trinity and Planetfall.

Infocom is, of course, no longer around. So what went wrong? Well, as detailed in this paper, the games mainly served to fund the “true endeavor” of the Infocom founders, business software. They wanted to make it big in business software like their Lotus comrades, both because it seemed to be a bigger pie and because of the fact that they were somewhat embarrassed about making their money in the game industry. I’m not certian what the status is now – I’m sure it has improved – but at that day and time the game industry was the “black sheep” of the software industry. To only have made your money off of games and call yourself a software developer was akin to being a porn film director and comparing yourself to Speilberg.

Infocom decided to start working on a business database product called Cornerstone. It employed the same Z-machine technology which had made the games so portable and popular. However, it fell flat. At the time the lone competitor was dBase, still a player today. dBase had basically nil in so far as portablility and ease, so Infocom spotted a window of opportunity. The only problem was speed. Cornerstone was so woefully slow that some reviewers gave up on it during testing. The business world needed fast reliable software, and Cornerstone was too slow. The Z-machine worked fine for text games where speed is of less concern, but it choked on more intensive applications. Also, the platform ambiguity was less of a selling point in 1989, when the market had already gone a long way to whittling itself down to the duolopy it is today.

So Infocom merged with Activision and renamed the whole venture Mediagenic. Mediagenic was under the auspices of Tom Snyder (yes, that Tom Snyder) and once he sold it off it changed its name back to Activision. By this point everyone who was left at Infocom had left the company. Couple the decline of Infocom with the increasing popularity and sophistication of computer graphics, and it spelled the end of the text adventure in the interactive entertainment industry.

There were of course other companies who made IF but they all phased it out eventually. However, some found it was more interesting to write IF than it was to play it. Mark Welch authored a commercial piece of software called AGT, the Adventure Game Toolkit. From 1987 to 1992 he marketed the product, eventually making it freeware when he no longer wanted to maintain it. To encourage people to buy and use it, he sponsored an annual contest which ran for four years. AGT could produce games which were more advanced than Infocom offerings in some ways and less advanced in others, but ultimately less intricate due to the abstaction of the interface.

In the early 1990’s, around the same time that AGT was being discontinued, rumblings began on newsgroups to the effect that people wanted to be able to play the IF of the 80’s on the computers of the 90’s.This event was mainly spurred on by 1992’s re-release of the old Infocom games in two packages by Activision called Lost Treasures of Infocom. DOS based games would still of course run in Windows, but people wanted to branch out to other platforms such as Linux and to be prepared should legacy platforms be done away with (i,e, for when Windows 2012 finally kills off all DOS support). Therefore people wanted Z-machine interpreters for the platforms of the day. This would of course mean that the Infocom story file format (the “programs” the Z-machine ran) would have to be reverse engineered.

In 1993 a man named Graham Nelson delivered a 1-2 punch. Not only had he reverse engineered the Z-machine and authored a document detailing how to write an interpreter for the games on any platform, but he also unveiled a new programming language called Inform for the creation of new IF. Heavily based on C/C++ and with a hefty manual in tow, it single handedly sparked a renaissance of IF programming, and there’s been no turning back since.

Today Z-machine interpreters exist for every platform imaginiable, from Windows to Linux, including PalmOS and the Game Boy (though it’s really hard to use that one). There have been thousands of games written using Inform. Inform came full circle when a modern day Zork title, Zork: The Grand Inquistor, was prototyped entirely using Inform and as a bonus the original Zork creators got back together to author another IF sequel, Zork: The Undiscovered Underground.

In 1995 someone suggested a revival of the old AGT contest, only this time no boundaries on what kinds of systems could be used to author games. The result was the newly-christened Interactive Fiction Competition, which is now in its seventh year. Inform entries make up over half of the entries this year. Inform’s Designer’s Manual is in its fourth version, and its sixth major version looks more or less final (which it probably should be – IF doesn’t exactly innovate).

IF is fun to play – I fire up these games on my Visor all the time (though it’s kinda dippy to be sitting there making your own maps on paper). As a genre, it is freed from a lot of the shackles of video games – a need for better graphics and more action – and is instead able to concentrate on other elements. Games like Quake are hindered by unrealistic AI, stiff character movement, and nil character interactions (Carmack noted they dodge this issue a lot by using Zombies in the new DOOM game and such), as well as a need for higher quality graphics and a bleeding edge engine. The engine of IF will never need more work, and instead the AI is finely developed and the character interactions are not hindered by what you can do with a mouse.

In 1995 or so an individual started up an FTP server for IF, called the IF Archive. For many years it has been the central repository of all things IF. In 2001 the server he hosted it on went away, but by that time it has been reloacated to http://www.ifarchive.org/, where it is accessible via HTTP and FTP. If you go there, first find an interpreter for your platform, then pick up the games.

In the course of its lifetime Infocom released 35 IF games. Activision released the two previously mentioned Lost Treasures of Infocom collections, the first containing 20 games, and the second containing 11 games on the floppy version and 14 games on the CD version, representing all games except for Leather Goddesses of Phobos, which was available via coupon to help alleviate confusion with the graphic sequel, Leather Goddesses of Phobos II. They followed this up in 1994 with five themed collections, each of which had 4-6 titles, plus Planetfall (which was gearing up for a now-cancelled sequel) and one of the five Zork titles (the original trilogy also had two sequels, Beyond Zork and Zork Zero). These collections contained 31 games, excluding LGoP again, as well as the more graphical IF titles Infocom did, Shogun, Arthur, and Journey – games Infocom did towards the end of their run which have since given users many headaches through their usage of DOS EGA graphics and who are incompatible with Z-machine interpreters (though Zork Zero uses this engine as well). Finally, Activision released a compilation CD in 1996 called The Masterpieces of Infocom, containing 33 of the 35 IF games Infocom released. (by this point Shogun and Hitchhiker’s Guide‘s rights had expired). It is available sporadically from their store, and if you have a software store near you with slightly older stock this is your best bet on finding the games.

So, if you’re an anal retentive completionist like me, that’s 33 games available on the Masterpieces CD, which excludes Shogun, which can be downloaded here, The Hitchhiker’s Guide to the Galaxy, which can be played here (as well as on the Sci-Fi Collection), and their non-IF games, which can be found (along with some of their IF ones) here, along with Quarterstaff, the one game by Infocom (instead of just under the Infocom label) that never made it to the PC (it was a Macintosh only title).

But if you’ll excuse me, I’m off to go read the Inform Designer’s Manual, 4th Edition and learn just enough Inform to be dangerous.