Historical Analysis of MUD Servers

Giovanni Ruffini, gruffin@netcom.com

Research on multi-used dungeons in the recent years has focused on the psycho-social aspects of mudding. The discussion has largely stayed within the narrow range of definitions between mudding as a dangerous addiction and mudding as a hip new form of psychotherapy. Little thought has been given to the historical origins of these phenomena. The discussion takes place completely outside of its own context, as if these MUDs emerged fully formed from the ether. What sketchy historical analysis takes place usually covers two decades in two hundred words, running from Dungeon to Ultima Online without pausing for breath. It is the proposition of this comment that the best way for the mudding community as a whole to understand itself is to engage in a thorough discussion of its own social, institutional, and political history.

There are a number of reasons why this approach is not already common. To begin with, the concept itself can seem a bit nebulous. What are the subjects for such a history, the players, or the characters that represent them? What are the institutions through which they move, ever-altering pieces of code, or the mental concepts which they serve to create? Does the history of a MUD depend on the so-called "real life" externals, such as the geographic locale of the machine and the sexual proclivities of its implementors, or can it be described as if it had an independent cyberspatial existence in its own right? How are we to reconcile the need for an honesty about real life issues with the average mudder's desires to have their lives remain in a quiet anonymity? Finally, how are we to deal with problems of scope? At what point does the history of a mudding community begin to lose sight of meaningful detail, and at what point do the political dramas of one MUD cease to have meaning to the history of the community as a whole? These are difficult questions, depending largely on the biases of the author.

A comment on methodology is in order at this point. What are the tools available for engaging in such research? Obviously, the richest body of knowledge a mudding historian is likely to find is the oral tradition of the community itself. Interviews, usenet posts, session logs, and old emails are now the necessary tools of the modern historian. Most useful, however, and to date the avenue least explored, is what I think of as "code criticism".[1] Simply put, one copy of a MUD's source can tell more than any number of session logs. Compare the code, line by line, with the base source from which it stems. Read the comments of past coders. Look for tell-tale changes in style, deletions, insertions, incongruities and overlaps in approach. Glean evidence of immortal careers by careful perusal of old wizlists or backup copies of area submissions. Even pay close attention to the date of last edit on each specific file.

We can show the utility of these historical approaches to mudding by discussing a concrete example. My mudding career began five years ago on a DikuMUD known as JediMUD. The approaches formulated herein have stemmed from a prolonged stay within the Diku community. Thus, little attempt has been made to place the issue in the context of the wider mudding world. Nevertheless, discussion of the social, institutional, and political history of mudding can be elucidated even by such a narrow example as JediMUD's historical role within the Diku community.

This story is the subject of an unfinished narrative history over a hundred pages in length. I will share here a condensed version of one of the more significant aspects of this work. One of the largest branches of the Diku community is CircleMUD. Jeremy "Ras" Elson's public release code has defined the paradigm of an entire generation of mudders. What is the historical explanation? The development of CircleMUD cannot be understood without the sociopolitical context of JediMUD, where Ras served as head coder for over a year. During that time, he became dissatisfied with JediMUD and sought to release his own code, partially as a form of escape from the political quagmire around him.

This is the paradox of minute cause leading to great effect. To understand why Ras drifted from JediMUD, one must explore the first months of JediMUD's version 4.0. A picnic at Johns Hopkins University is one of the fascinating points of the tale. Rampant factionalism marred the event.[2] Members of KHFC, a well-placed and notorious JediMUD clan that included several friends of Ras, drove some members of the Jedi administration to accusations of sexual harassment. The administration's cyber-vengeance against that clan left Ras stuck between the proverbial rock and a hard place. Barely a week after the picnic's political disasters, Ras announced the public release of CircleMUD, including in the package large amounts of JediMUD code that he himself did not write.[3]

This is a single anecdote that makes no claim to a thorough analysis of cause and effect. It simply takes a step towards revealing the ways in which historical context deepens our understanding of the patterns of the mudding community as it currently exists. The eventual publication of a complete history of JediMUD will do more justice to a tale well worth telling. Other topics suggest themselves readily. What sort of personnel continuities existed within the fledgling Diku world that led to so much conceptual overlap amongst the earliest public release versions of Silly, Sequent, and Copper? How are we to understand the mammoth impact of Sojourn and its spin-offs? Hopefully, these comments will inspire others to illuminate the many rich corners of the Diku community.

[1] See a discussion of the topic on a macro-historical level in Martin Keegan JOMR 2(2).

[2] Documentation of this event in the personal possession of the author includes archives of KHFC clan email, including eye-witness accounts from KHFC members, and the JediMUD implementor named Torg, JediMUD news postings from the summer of 1993, and transcripts of online personal interviews with JediMUD's Ras, Romulus, Kombat, and Naved.

[3] See the "credits" file in the standard releases of CircleMUD 2.2 and CircleMUD 3.0 beta patch 11. JediMUD's Torg in particular is responsible for a number of coding concepts transplanted directly into the Circle release.