I’ve been meaning to write this column for a couple years but, somehow, just never got around to it. I intended to do it back in early 1999, when the game press started quoting MMOG publishers referring to EverQuest and Ultima Online as "first generation massively-multiplayer games." I was somewhat offended by this attitude, as it implied the industry had sprung fully grown from somebody’s navel in 1997, like they hadn’t been using ideas already in use in other games for years. It ignored all the work that had gone on for thirty years, in fact. What about Kesmai with all their games starting in 1982, Mythic (nee’ AUSI) with several games starting in 1984, or the original Bartle/Trubshaw MUD1 in 1978? What about MPG-Net with Kingdom of Drakkar, or ImagiNation Network with Shadow of Yserbius? Or the original AD&D: NeverWinter Nights on AOL? For goodness sake, what about the DIKU MUDs that inspired Brad McQuade and crew to develop EverQuest?

I did turn out a series of three columns called "Happy Birthday, Online Games" in October and November of 1999, motivated mainly by the prevailing attitude among the gaming press that online games had started in 1994 or 1995 with Quake and such games. It ticked me off so much that I wrote the series as a general timeline history of online games. You can find the original 1999 columns in the archives here as .PDF file.

Somewhat to my surprise, those three BTHs have been reprinted more than any other of my columns. I had no idea there would be such interest in the history of online games. Things really took off when Raph Koster distributed them to the MUD-DEV mailing list, then went on to expand on them on his own web site (select Essays and scroll down to Online World Timeline). I’ve always meant to expand mine and separate the entries by generations, as well, just to get back to my original thought and intent.

And it should be ‘generations,’ plural; by my reckoning, we’re in the third generation, heading quickly into the fourth. So what, you might say; does it matter it? Hasn’t the technology changed so much in thirty years that we literally started over in 1997? For all that I rail and cuss these days about how art is taking precedence over the basics of good games today, there was both art and entertainment in some of those games, and they built the foundation of the best of what we see today. Ignoring those seminal games, and the processes and design elements they developed, is like saying that the first generation of movies started with color and what went before is meaningless. I don’t think anyone would seriously propose that Nosferatu, The Grapes of Wrath, Citizen Kane and Casablanca were unimportant to the evolution of films.

Before I could rewrite the timeline in generational terms, however, I realized I needed to set down what defines each of the generations. Writing that column has taken on something of a sense of urgency, because in a couple months I’ll be on a panel discussion at the Game Developer’s Conference titled Building a Third-Generation Online Persistent World Game. Gordon "Tyrant" Walton, Executive Producer of The Sims! Online and one of the most knowledgeable people in the industry, is chairing the panel. Other scheduled participants are Rich Lawrence and Raph Koster from Verant/Sony Online.

Gordon and I (among others) have congenially disagreed for the past three years or so about just what generation marker we should assign to today’s games. In his best mock-stern "Tyrant" manner, he has also cautioned me not to waste the panel’s time with a long argument over the issue. And he’s right; the panel isn’t the time or place for that, really, because there will be plenty of disagreement and wriggle-room in any definition given.

So, I decided to finally write that ‘generations’ column and let it serve the purpose. Here it is, in all its opinionated glory; agree or disagree at your leisure.

The First Generation: 1969 to 1977

I peg the beginning of the first generation in 1969, when Rick Blomme wrote a two-player version of MIT's famous Spacewar for the PLATO service. PLATO (Programmed Logic for Automatic Teaching Operations) was one of the first time-sharing systems, dedicated to experimenting with new ways to use computers for education. Originally built in the late 1960's at the University of Illinois/Urbana, it blossomed into a system that, by 1972, could host about 1,000 simultaneous users. PLATO featured plenty of games over the years, but my research indicates this was the first true multiplayer game on the system. It is also interesting to note that PLATO had graphics capability, as evidenced by Airfight, a two-person 3D dogfighting simulation that first appeared on PLATO around 1973.

Note that Blomme’s Spacewar was done two years before the first commercially available arcade videogame, Computer Space, which was pretty much… Spacewar. 1969 was also the year ARPAnet was founded. You probably know ARPAnet as the homey little system we now call the Internet.

PLATO doesn’t get enough credit for inspiring the online games revolution, or for its effects on computer and videogaming as a whole. For example, the guys who wrote Airfight went on to build Flight Simulator, which was bought by Microsoft in the 1980s and is now Microsoft Flight Simulator, one of the best-selling PC games of all time. The service also had one of the first 32-player versions of the venerable Empire circa 1974, a game that developers still imitate today (think "turn-based and/or real-time strategy"), and Oubliette, circa 1975-1978, which was a dungeon crawl that inspired the Wizardry series that is still with us today. And it’s pretty obvious that Spacewar inspired Nolan Bushnell to create Computer Space in 1971 and then to go on to found Atari a year or two later.

During this first generation, there was plenty of experimentation with multiplayer games on various platforms, operating systems and languages. Most of it took place at major university mainframe computers at places like MIT, the University of Virginia and Essex University in Colchester, England, where the Second Generation begins.

The Second Generation: 1978 to 1995

I don’t think many in the industry today would argue that the original 1978 Bartle/Trubshaw Multi-User Dungeon (MUD) was a key turning point in the history of the industry. This was the first true attempt to create a real-time, many-player game with character growth, a semi-persistent world that reset several times per day, but always came back up with the same general terrain, monsters and puzzles. The true persistence came from the character growth and expansion, a la TSR’s Dungeons and Dragons. If you’re interested in all the details, you can find them on Dr. Bartle’s site.

It was also the first instance of a true real-time, many-user game that could be installed anywhere there was a PDP-10 computer - and was it ever distributed. Sometime in the early 1980s, when Dr. Bartle was sharing the code with other universities for education purposes, someone took it upon themselves to release the code to the ARPAnet and it was all over. By the mid- to late 1980s, students at universities all over the world were building variations. Sure, there were other efforts going on in 1978, but this is the game that pretty much evolved the commercial side of this industry.

The key differentiators that make this the second generation are:

• With MUD1, the industry broke out of the need to own a proprietary, single service terminal, such as PLATO required; anyone with a computer and modem could play most of the games;

• The ‘persistence’ model of game design showed that some people would maintain characters not for one or two sessions, but for years.

• The explosion of games onto the primitive commercial online services; people actually started paying money, and lots of it, to play multi-user games online. It truly started in 1983 with MegaWars I on CompuServe, continued with MUDI as a commercial game in 1984 and really started to open up in 1985 and 1986 on GEnie with games such as Air Warrior, Stellar Emperor, Gemstone II, A-Maze-ing, Multiplayer BattleTech and Dragon’s Gate. Throughout the second generation, the average cost for game access was over $4.00 US per hour;

• The first tentative steps into the market by professional game publishers and developers. Activision, Interplay, Sierra Online, Stormfront Studios, Virgin Interactive, SSI and TSR all dipped their toes into the waters by trying out games on GEnie, Prodigy, AOL or CompuServe;

• The first examples of dedicated game networks, beginning with Sierra’s failed The Sierra Network in 1991, later renamed the ImagiNation Network, which was bought by AOL in 1996. Other examples include MPG-Net, TEN, Engage and Mplayer.

The Third Generation: 1996 to Today

This generation begins with the launch of Meridian 59 in Fall of 1996. The game was developed independently by Archetype, founded by the Kirmse brothers and for which Mike Sellers (The Sims Online) and Damion Schubert (UO2, now with his own company, Ninjaneering) worked. The company was bought by 3DO, who published the game. It wasn’t nearly as successful as the next competitor in line, EA/OSI’s Ultima Online in 1997. But then, M59 didn’t have the advantage of having sold millions of Ultima series games, making it one of the most recognizable RPG brand names in computer gaming. 3DO couldn’t make up its mind about a price point either, and at the time, raising prices in the face of competition wasn’t a good idea.

And of course, we all know of the success of EverQuest, Lineage in Korea and Dark Age of Camelot. The third generation has been more evolutionary than revolutionary, with each new entrant into the field seemingly forced by a pact with Satan to relive the mistakes of its predecessors, back unto two generations. One wonders how many subscribers the industry would have now, had those mistakes not been reinvented by today’s publishers and developers. We must make allowances, I suppose; after all, most of them are newbies to the industry.

The key differentiators for this generation are:

• The first widespread use of low, flat monthly rates for MMOGs opened up the genre beyond the early adopters who could afford to pay an hourly rate;

• The first widespread use of the decentralized Internet for playing MMOGs, which made the games available beyond the proprietary online services and opened them up to the masses on a worldwide basis;

• The introduction of professional-sized development budgets and development teams to online games. One could argue that News Corporation’s buyout of Kesmai and Interplay forming an online service division in 1994 started it, but we didn’t see the true effects until 1996, when games such as M59 and Mythic’s Rolemaster: Magestorm were published;

• The beginnings of a true methodology for developing and managing persistent worlds and online games in general. This something we’ve been missing, but with both EA and Sony investing hundreds of millions of dollars, we pretty well had to start coming up with some processes to at least lower how much of that money ended up wasted.

So, you might ask, what will the Fourth Generation look like? Ask me after the panel in May; I plan to write a column about it then.