Ian & Stuart's Australian Mac: Not for Sale/Another.not.for.sale (Australia).iso/hold me in your arms/Media Lab/MediaMOO Results

Text File | 1994-09-13 | 40KB | 721 lines

               Virtual Professional Community:
              Results from the MediaMOO Project

           Amy Bruckman            Mitchel Resnick
     MIT Media Laboratory            MIT Media Laboratory
      [e-mail]            [e-mail]

Abstract

MediaMOO is a text-based, networked, virtual reality environment designed to
enhance professional community among media researchers.  This paper analyzes
experience with the system to date and highlights the value of Constructionist
principles to virtual reality design.

Virtual Professional Community

Once or twice a year we stand with name badges sipping coffee in a corridor,
exchange ideas over expense-accounted lunches, and maybe attend a few talks.
Friendships are made and projects hatched.  Then it's back home to file for
expenses, perhaps write a trip report, and get back to "real work" and
relative isolation.

MediaMOO is a text-based, networked, virtual reality environment designed to
extend the type of casual collaboration which occurs at conferences to a daily
activity.(1) Visitors to a conference share not just a set of interests, but
also a place and a set of activities.  Interaction is generated as much by the
latter two as the former:

Person A:    Can you tell me how to get to Ballroom A?
Person B:    I'm headed that way now.  It's up this way.
Person A:    Thanks!
Person B:    I see you're at Company X....

Person C:    Is this seat taken?
Person D:    No, it isn't.
Person C:    I'm surprised the room is so packed.
Person D:    Well, Y is a really good speaker....

A text-based virtual environment can provide both a shared place (the virtual
world), and a shared set of activities (exploring and extending the virtual
world).  Like at a coffee break at a conference, there is a social convention
that it is appropriate to strike up a conversation with strangers simply based
on their name tags.  On MediaMOO, you can read descriptions of people's
research interests as well as their names, and this can form a basis for
striking up a substantive conversation.

However, name tags alone are not enough.  The best sorts of interactions occur
when people participate in a shared activity and not just a shared context.
On MediaMOO, this takes the form of constructing and interacting with the
virtual world.  The Constructionist theory of education emphasizes the value
of constructing personally meaningful artifacts [Papert 80].  This theory has
guided design decisions made in MediaMOO.  For example, in most MUDs, the
privilege to extend the virtual world is restricted to a small number of
users.  Everyone in MediaMOO is automatically a programmer with full
privileges to create new objects and places in the virtual world.

This paper has two main goals.  First, it documents experience with the
MediaMOO project to date and evaluates its success as a virtual professional
community.  Second, it explores the application of Constructionist principles
to virtual reality design.

What is MediaMOO?
MediaMOO is a text-based, networked virtual reality environment or "MUD"(2)
running on the Internet.  Its basic structure is a representation of the MIT
Media Lab.  Users connect in the LEGO Closet, and then step out into the E&L
(Epistemology and Learning research group) Garden:

>connect guest
Okay,... guest is in use.  Logging you in as `Green_Guest'
*** Connected ***
The LEGO Closet
It's dark in here, and there are little crunchy plastic things under your feet!  Groping around, you discover what feels like a doorknob on one wall.
Obvious exits: out to The E&L Garden

>out
The E&L Garden
The E&L Garden is a happy jumble of little and big computers, papers, coffee
cups, and stray pieces of LEGO.
Obvious exits: hallway to E&L Hallway, closet to The LEGO Closet, and sts to
STS Centre Lounge
You see a newspaper, a Warhol print, a Sun SPARCstation IPC, Projects
Chalkboard, and Research Directory here.  Amy is here.

>say hi
You say, "hi"
Amy says, "Hi Green_Guest!  Welcome!"

Users from around the world connect to this virtual place to socialize, talk
about their research projects, interact with the virtual world, and create new
objects and places.  People from the Media Lab are encouraged to build their
own offices; users from other places can build their offices as well and
connect them via a "virtual Internet."  The system developers constructed
basic infrastructure and a few interesting and evocative objects and places,
but almost all the building was left to the users.  This is not a result of
time constraints, but is a central principle of the project that will be
elaborated throughout this paper.

The first MUD or "Multi-User Dungeon" was developed in 1979 as a multi-player
Dungeons and Dragons game.  In 1989, a graduate student at Carnegie Mellon
University named James Aspnes decided to see what would happen if the monsters
and magic swords were removed, and instead each user was allowed to help
extend the virtual world.  Aspnes' project, which he called "TinyMUD," became
less like a game and more like a community.  There was no longer a score or
goal, but instead a gathering of people who enjoyed one another's virtual
company and worked together to extend the virtual world.

At the MIT Media Lab, we decided to see whether this technology which began as
a game could be adapted to a more serious purpose: enhancing professional
community among media researchers.  We chose to build on top of the MOO ("MUD
Object Oriented") software developed by Pavel Curtis at Xerox PARC.(3) System
development began on October 28th, 1992, and MediaMOO was opened to the public
on January 20th, 1993, with an opening celebration called the MediaMOO
Inaugural Ball, scheduled to coincide with Bill Clinton's inauguration as
President of the United States.  As of May 7th 1993, MediaMOO had 327
participants from at least thirteen countries, including Austria, Australia,
Canada, Denmark, England, France, Ireland, Israel, The Netherlands, Norway,
South Africa, Sweden, and The United States.  (See Table 1.)  The population
has grown over time, with peaks at special events and troughs on weekends.
(See Figure 1.)  MediaMOO runs on a Sun SPARCstation IPC where it uses 25 Mb
of RAM.  The database is currently 4 Mb on disk.

Table 1:  MediaMOO People and Characters
People Registered(4)        327
Registered Characters(5)    375
Identified Characters        279  (74%)
Anonymous Characters        96   (26%)
    Multiple Characters          48   (13%)
    Anonymous Character Only    48   (13%)

[Graph of connections versus date omitted from .txt version]
Figure 1:  The MediaMOO Population

A Community of Researchers

While people of a wide variety of ages and backgrounds do MUD, the majority of
players on publicly announced Internet MUDs are college students.  On
MediaMOO, we wanted to attract media researchers.  We advertised selectively
on electronic mailing lists devoted to media studies, and required people to
submit a description of their research interests to register.  While we have
said no to some applicants, it is primarily a self-selection process.

In most MUDs, characters are anonymous.  People who become friends can
exchange real names and email addresses, but many choose not to.  Conventions
about when it is acceptable to talk about "real life" vary between
communities.  In most MUDs, people begin to talk more about real life when
they get to know someone better.  However, in some communities such as those
based on the Dragonriders of Pern series of books by Anne McCaffrey, talking
about real life is taboo.

On MediaMOO, we wanted to promote discussions of "real life" and real research
interests.  Consequently, we offered users the opportunity to get an anonymous
or identified character, or both.  Since there was already a function called
@whereis which returned the location of a character, we implemented a function
@whois which returns their real name, if they choose to be identified.  Real
names are set by the system administrators to lessen the chance that people
will pretend to be who they are not.  We also implemented functions so that
people could carry a description of their research interests on their
character.

It surprised us that almost all users chose to be identified.  To date, 74% of
users chose to have only an identified character, and 13% chose to have both
anonymous and identified characters; another 13% chose to be completely
anonymous.  (See Table 1.)  Two days after the opening, one user who had been
anonymous sent in this request:

"I have enjoyed MediaMOO so far and like the direction it is headed.  With
these forums and such I believe I would like an identifiable character, but
would like to continue with my anonymous character as well."

Users clamored for extensions to the functions provided for finding people.
At users' request, we made the function @whois return the person's email
address as well as real name.  One user asked to be able to reveal his/her
email address, but not real name.  A user whose character is named Daniel
(Daniel Rose in real life) wrote a function @char to find the characters
associated with a particular person's real name (the inverse of @whois).
Users requested that @whois with no arguments return a list of the real names
of everyone currently logged on.  A user named cdr (Ken Schweller) made a
research directory, enabling people to find all the users who mention a
particular word or phrase in their research description.  The idea of having
identified characters and research interests has proved popular and
successful.

A "Third Place"

In The Great Good Place, Ray Oldenburg eloquently argues for the importance of
"third places," places which are neither work nor home.  The book's subtitle
is "Cafes, coffee shops, community centers, beauty parlors, general stores,
bars, hangouts and how they get you through the day."  Oldenburg summarizes:

"Third places exist on neutral ground and serve to level their guests to a
condition of social equality.  Within these places, conversation is the
primary activity and the major vehicle for the display and appreciation of
human personality and individuality.  Third places are taken for granted and
most have a low profile.  Since the formal institutions of society make
stronger claims on the individual, third places are normally open during the
off hours, as well as other times.  The character of a third place is
determined most of all by its regular clientele and is marked by a playful
mood, which contrasts with people's more serious involvement in other
spheres.  Though a radically different kind of setting from the home, the
third place is remarkably similar to a good home in the psychological comfort
and support that it extends.  [Oldenburg 89, p. 42]"

The population of third places are self-selected-people go to a cafe
because they choose to and not because they must.  From this self-selection
process emerges a group of people with some degree of common interests and
values.  Traditional third places draw people from the local geographic area.
On the Internet, MUDs become third places which draw people with common
interests from all around the world.  People from the opposite hemisphere can
become a part of your daily life.  On MediaMOO, those people also share
research interests.  MediaMOO is perhaps best described as an endless
conference reception.  The conversation fluidly moves between personal and
research issues.

As Oldenburg points out, conversation is the primary activity in third places,
and MediaMOO is no exception.  Most of those who chose to respond to an email
survey of randomly selected MediaMOO users stated that they had had
interesting conversations and met new people in their field through MediaMOO.
Paul (Paul Dourish) writes:

"I've met a number of people whom I've talked to about my research and theirs,
although I think there are fewer (probably just the one or two) whom I've
actually talked to enough to refer to as "new professional contacts".

One I met while he was a Guest; we started talking after he read my research
description. The other I met early on when I was stumbling around asking all
sorts of people for help on doing things just after I got my character."

The process of helping new players, often called "newbies," is an important
part of MUD culture.  For Paul, the process of being new and reaching out for
help has led to his most meaningful professional contacts- once while he
himself was new, and once while he was helping another new player.  The
context of the MUD in these instances provided a shared context and shared
activity which promoted social interaction.

A number of users commented that their most meaningful interactions on the MUD
were with the "regulars"-the people who use MediaMOO the most and are most
likely to be logged on at any given time.  Oldenburg emphasizes the importance
of "the regulars" to a third place-they give a place its character.

Some question the value of the sort of interaction which takes place on
MediaMOO.  One user wrote that "frankly it strikes me for now as a schmooze
place for people with nothing better to do, not a place where more productive
things will happen than already happen in other communicative modes."  We do
not share that perspective for a variety of reasons.  Is spending time at a
conference reception a waste of time?  Most veteran conference-goers attest to
the fact that the conversation at coffee breaks and receptions is usually more
valuable than the sessions attended. The exchange of ideas and networking
which take place on MediaMOO are similarly productive.  One might question
whether these benefits can be obtained with less time commitment through other
media-but this analysis ignores other benefits of the MediaMOO.  For media
researchers, coming to understand this medium may be interesting in itself.
There are also personal, emotional benefits that come from participating in a
"third place," such as relaxation, friendship, good conversation, and a sense
of belonging to a community.  In short, if you enjoy it, then you will reap
benefits from it which go beyond that pleasure.  Some people do not enjoy it,
and that is certainly a valid point of view.  It is a strength of the medium
that the community is self-selected-everyone who is there wants to be there.

It's worth noting that to determine whether an activity is "productive"
requires a definition of what it means to be "productive," and this quickly
leads to questions of an ethical and philosophical nature.  MediaMOO
challenges the boundaries between work and play, forcing one to rethink what
counts as productive.

Online Colloquiums and Forums

The kind of networking and casual collaboration which occurs at conferences
does appear to be taking place on MediaMOO.  It seems natural to extend the
conference metaphor to literally have meetings on MediaMOO.  This has been
tried several times with mixed results.  Alan (Alan Wexelblat) writes:

"I've been to the Ball and a few other gatherings.  I thought they were
OK, but a bit jabbersome."

Conversations in MUDs are often multi-threaded.  When large numbers of people
all talk at once, each of those threads can become very short-rather like a
tailor's scrap bin.  Users have experimented with a variety of
computational ways to improve the quality of discussion.  Wade (Wade Roush)
organized a forum on Cyberspace and the Humanities, which more than 40 people
attended.  The forum took place in two different "conference rooms," each with
a different method for focusing conversation.  In one, users are all allowed
to talk at once and tag their comments with a relevant subject line.(6) In
another, a limited number of people can talk at a time and there is a queue
for those wishing to speak.(7) This increases the coherence of the conversation
but takes an already slow medium and slows it further.  Oracle (Randy Farmer)
writes:

"I attended the Cyberspace and the Humanities get together for awhile.  It was
clear that the wide-open discussion didn't work very well, it got a bit nicer
in the moderated room.  I don't think much came out of the meeting, 'cept
maybe that many more experiments in group discussion need to happen.  I'll
gladly help!"

There seems to be interest and potential in exploring new ways to improve the
quality of forums and group discussions.  However, it is possible that the
problem resides in too literal an application of the metaphor of a conference
to virtual space.  A virtual meeting is not a conference or colloquium; to
call it that is a metaphor borrowed from an existing medium and applied to
this new medium which we do not yet completely understand.  Many early movies
were like recorded plays.  The camera was pointed squarely at a stage and
never moved.  It took years of experimentation to develop new techniques such
as close ups and zooms.  Early automobiles were called "horseless carriages."
We are still at the horseless carriage stage of cyberspace.  Initiatives for
improving online gatherings must not just stretch ways the technology can
create a conference, but stretch the metaphor of "conference" to create a new
form of social and intellectual event for this new medium.

It seems likely that online meetings will be more successful when networked
video and audio systems become more generally available.  Text is a
tremendously expressive medium and has advantages over graphics and audio for
some applications.  For example, in less than a moment one can write that "at
the top of the hill is a gnarled peach tree," and this conjures up an image in
the reader's mind.  It would take considerably longer to draw such a tree,
and less would be left to the imagination.  Audio and video will not replace
text-based VR, just as television didn't replace radio or magazines but
simply changed what they are used for.  Online meetings are one application
where the greater bandwidth of audio and video may be needed.

Contributory Objects

Building amusing and/or useful MUD objects is a means of creative expression
for the designer, and completed objects promote social interaction for the
community.  One design paradigm which has proved particularly successful is
the idea of a contributory object.  For example, in the dressing rooms of the
MediaMOO Ballroom, it is possible to design new costumes for the clothing
racks:

>northwest
You step through the velvet curtain into the women's dressing room.
Women's Dressing Room
The dressing room is a clutter of gowns, hats, and gloves from all different
eras.  Type 'examine rack' for more information.
Obvious exits: east to Ballroom Foyer and south to The Ballroom
You see women's clothing rack and a gold plaque here.

>list rack
Outfits on the rack:
1: a classic black cocktail dress and snakeskin pumps by Amy (#75)
2: a floor-length, strapless, pink-taffeta, ball gown and long, dangling
rhinestone earrings by Amy (#75)
3: a red silk dress with a black, patent-leather belt and high heels by Amy
(#75)
4: a marigold, lower calf length dress, with an off-shoulder top and an
elegantly pleated skirt by Lenny (#115)
5: a three quarter sleeved, cotton knit dress, in basic blue by Lenny (#115)
6: a ribbed, cotton sweater-dress, cut to mid-thigh, in white by Lenny (#115)
7: a halter wrap dress, cut to mid-thigh, colored in pastels with a distinctly
tie-died look to them by Lenny (#115)
8: a knit, royal-purple trankdress, belted at the waist, with a full, sweeping
skirt by Lenny (#115)
9: a black evening dress with long, delicate sleeves, made entirely of lace by
Lenny (#115)
10: a ribbed cotton turtleneck and clean, neat, denim jeans by Lenny (#115)
11: a three-quarter's length black leather trench coat, with black stockings
by Lenny (#115)
12: an off-the-shoulder black velvet dress with handmade wide white lace all
along the slightly veed neckline by Michele (#120)
13: a gorgeous red dress with a slight sweetheart neckline, buttons down the
side, a flared, mid calf length skirt, and a slit to the hip by Lenny (#115)
14: black leather one-piece jumpsuit with glistening alloy lapels by Guest
(#113)
15: frayed jeans and a rather faded Phish t-shirt by Paul (#559)
16: a floor-length green-sequined evening gown, matching sling-back heels, and
dangling diamond earrings by Amy (#75)
17: a narrow-cut, black silk dress with a slit almost up to the waist, and
spike heels by Amy (#75)
18: clean but well-worn jeans, a faded Phish t-shirt and a comfortable old
pair of Birkenstocks by Drave (#176)
19: black shoes, black socks, black jeans, black shirt, black jacket black
earrings by Amber_Guest (#700)
20: a floor-length royal-purple velvet dress with a scalloped neckline, fitted
bodice and flowing skirt, with matching purple suede shoes and opaque purple
nylons by Robyn (#575)
21: a shimmering jester's costume, in mauve and lavender, with a headdress of
orchids and dove-feathers, and turquoise pendant earrings, set off by
turquoise high-heels by Mauve_Guest (#702)
22: a slinky, classic little black dress by slim (#913)
23: A glowing violet Indian Sari with Paisleys. by Guest (#113)
24: absolutely nothing by Guest (#113)
25: a sheath-style ball gown made of Shantung silk in deep, hunter green.
This lovely dress comes with matching pumps by Tarot (#1749)
26: black leather thong with matching bustier, whip and pink high-heels by
Ivan_the_Unstable (#1769)
27: black leather, crotchless teddy by Guest (#113)
28: an ensemble in off-white: two-piece, calf-length dress, high boots, long
cape, fake fur muff and hat by Guest (#113)
29: a pair of Guat pants and a Grateful Dead t-shirt by Chartreuse_Guest (#696)
30: a t-shirt reading: Duke- Back to Back National Champions.  Blue basketball
shorts.  Shoes by Nike.  Jacket by Starter. by Ochre_Guest (#689)

>wear 1 from rack
You slip into a classic black cocktail dress and snakeskin pumps.

At the MediaMOO Inaugural Ball, people spent as much time in the dressing
rooms as in the ballroom itself.  The costumes on the rack are effective
conversational props.  More important, however, is the fact that it is easy to
contribute a new costume to the rack.  One can simply type "design 3Cyberconf
T-Shirt and mirrorshades for rack" and it is added to the collection of
available costumes with the designer's name attached.  Contributory objects
offer a lower threshold to participation than actually programming a new
object.  The user has a sense of having taken a first step towards mastering
the computational environment, and a sense of having contributed something to
the community.

There are a variety of contributory objects around MediaMOO, including
bartenders that you can teach new drinks, a talking picture that you can teach
to come alive when it is looked at, statues of famous sociologists and
historians of science that you scribble on (designed to promote discussion of
their work), and a projects chalkboard for ideas for new objects and places.
The majority of users asked to name one of their favorite objects on MediaMOO
cited one of these items.

It is interesting to note that attributing the contribution to a person is an
essential feature-it allows the person to take pride in what they have done,
and discourages virtual vandalism.  Even though guests are effectively
anonymous, there have been few inappropriate contributions to the costume
racks (as you can see in the unedited list which appears above).  However, at
one point it was possible to add messages to the bartenders without
attribution.  One might add something like "Bill starts rearranging the kegs
of root beer behind the bar" to the program for Bill, the bartender in the
Root Lounge.  Unfortunately, people added messages that were trivial or
obscene and even deleted other people's contributions since the program
allowed it.  The software will be rewritten to provide attributions for
messages added like the other contributory objects around MediaMOO.

Contributory objects are an example of the application of Constructionist
ideas to virtual reality design.  Seymour Papert, designer of the Logo
programming language, believes that people learn best when constructing
personally meaningful artifacts.  Papert seeks to empower children by helping
them to develop a sense of mastery over computers and computational ideas.
The Logo language is designed to have "no threshold and no ceiling"; it is
easy to begin participating, and the system is powerful enough to continue to
challenge a more experienced user.  A contributory object is an easy route to
initial participation which can open up into the greater power of mastering
the underlying programming language.

Constructionism involves two types of construction.  First, it asserts that
learning is an active process, in which people actively construct knowledge
from their experiences in the world.  (This idea is based on the theories of
Jean Piaget.)  To this, Constructionism adds the idea that people construct
new knowledge with particular effectiveness when they are engaged in
constructing personally-meaningful products.  They might be constructing sand
castles, LEGO machines, computer programs, or virtual objects.  What's
important is that they are actively engaged in creating something that is
meaningful to themselves and to others around them.

Constructing a new costume is an act of self expression which benefits the
individual and the community.  Langdon Winner remarks that "Social activity is
an ongoing process of world-making" [Winner 86].  In MUDs, this is literally
true.  Contributory objects facilitate this literal sense of world
construction.

Virtual Representations of Real Places

In addition to being able to contribute to objects, MediaMOO users can design
and program new objects and places.  The center of MediaMOO is a virtual copy
of the MIT Media Lab.  There is a psychological power in the ability to
construct a representation of a real place in the virtual world which is your
own.  The challenge was: how do we let people build offices in California and
Australia without trying to build everything in between?  We developed a
"virtual Internet" as a technique for spatial ellipsis.  Users from other
places can build their own offices by connecting them via the virtual
Internet.  You step inside a computer and dematerialize into a collection of
packets, and can then travel through a tree-like structure, going down to root
and back up through the hierarchy (actually a much more orderly arrangement
than the real Internet!)  Each user can add his or her own Internet site, if
it is not already there.

Daniel (Daniel Rose) writes of his first experience on MediaMOO:

"I logged in first as a Guest, and came out in the E&L Garden.  I had never
been to the Media Lab IRL [in real life], so I felt a bit lost....  Then I met
Michele, and when she found out I was from Apple, she said that someone had
constructed our building, and she'd take me there....  When we stepped out
into the Apple R&D building atrium, I felt this incredible shock of
recognition...  More than that, I felt a sense of relief that there were
places here that were familiar and home to me just as the Media Lab was to
you.  And all of this was from a couple of lines of textual description."

The actual text a user sees when they arrive at apple.com in the virtual
Internet and step out looks like this:

apple.com
You are in a maze of twisty little passages.
You see the back of a computer screen here.
Obvious exits: down to com and out to Apple Computer R&D Atrium

>out
Your packets gather in a glob, and then flow into the screen!  You feel
yourself rematerializing.
Apple Computer R&D Atrium
You are in a glass atrium, four stories tall. Offices look out from the walls.
Beyond the glass wall to the east, there is some arcane construction taking
place.  A walkway exits the atrium to the west.

To most people, this is a rather unremarkable description.  The idea of
rematerializing might appeal to Star Trek fans, but the room description
itself is bland-it sounds like an office.  However, to Daniel this provoked a
"shock of recognition" and a sense of belonging.  Representations of the real
give users a sense of comfort and make the medium more appropriable: if your
office is there, then you belong there.

Author Comments and Community Comments: A Lesson in Constructionism

One of the first additions we made to the MOO software was to allow comments
to objects they created.  We envisioned that this would serve as a kind of
documentation-if a programmer hoped others would use an object, he or she
could put useful information on the comments message.  (Reuse of code is
encouraged and facilitated by the object-oriented nature of the programming
environment.)

Months later as part of early preparations for a MUD for children, we decided
to develop a "Generic Commentable Object."  If an object is commentable (made
to inherit from Generic Commentable Object), then anyone can type "comment on
<object name>" and anyone can read all the comments that have been made on an
object.  It wasn't until after Albert (Albert Lin) implemented this feature
that it occurred to us that it was similar to the comments message, with one
important difference: the comments message is controlled by the owner of an
object; the commentable object creates a dialog among members of the community
about the object.

The second design better embodies the participatory philosophy of MediaMOO.
It assumes that the community of users have worthwhile things to say; the
privilege to comment is no longer reserved for the author of the object alone.
Valuing participation and respecting the contribution of each individual are
principles inspired by Constructionism.

Future Directions:  A MUD for Kids

The MediaMOO Project was conceived in part as preparation for a MUD for kids.
We believe that this technology can provide an authentic context for children
to learn reading, writing and programming.  In these virtual worlds, writing
and programming become means of self-expression to a community of peers.  MUDs
are a Constructionist playground.

Developing good MUD objects is as much creative writing as programming.  One
hypothesis of this research is that divisions between the humanities and the
sciences are often too sharply drawn and counter-productive, and a more
holistic approach has advantages for many children.  A second hypothesis is
that the social and contextual nature of these worlds may help young girls to
be more comfortable with computers.

If kids are really to make good use of MUDs, however, it will be necessary to
improve the programming language and the interface.  We are developing a
multiple-window client program which we hope will make the system more usable.
We are also currently developing a scripting language to make MUD coding
easier.  We hope to apply lessons learned in the development and use of the
Logo language to make a MUD language more accessible to kids.

At the conclusion of Mindstorms, Seymour Papert describes his vision of a
technological samba school.  In samba schools in Brazil, members of a
community gather to prepare a performance for Carnival.  Everyone is learning
and teaching-even the leads need to learn their parts.  People of all ages
learn and play together as a community.  Papert believes that computers can
create a kind of technological samba school, and we believe MUDs may begin to
realize that vision.

Conclusion:  Constructionism and Virtual Reality

In much current research, virtual reality is like Disneyland: artists design
wondrous creations for users to experience.  If this technology is
"interactive," it is in the limited sense that hypertext systems are
interactive: there are multiple paths through the material, and the system has
a limited ability to react to the user.  However, the ways in which the system
reacts are designed by the artists and engineers who constructed it and not by
the users.

If the power of this technology is to be unleashed, users need to be the
creators and not merely consumers of virtual worlds.  We believe that
Constructionist principles are of central importance to the design of VR
systems.  MediaMOO is an exploration of this idea.

Acknowledgments

The authors would like to thank Pavel Curtis for his wonderful software.
Albert Lin prepared statistics for this paper and has been a key contributor
to the development and daily running of MediaMOO.  Wade Roush's work with
the virtual STS Center on MediaMOO helped bring in enough new, interesting
people to get the community really on its feet.  Jeff d'Arcy, Michele
Evard, Remy Evard, Lenny Foner, and Greg Hudson have all contributed to the
project's development and maintenance.  Doug Alan, David Blank-Edelman, and
Randy Sargent have helped to keep purple-crayon, the computer MediaMOO runs
on, functioning smoothly.  Paul Dourish, F. Randall Farmer, Daniel Rose, Wade
Roush, and Alan Wexelblat commented on drafts of this paper.  Sherry
Turkle's work helped to inspire the project.  Most of all, the authors
would like to thank:

Marc(#102), Vortex(#116), Abbe(#118), Albert(#119), Harbin(#127), Seq(#128),
Emily(#129), LaoTzu(#141), Warren(#144), Torc(#145), DAn(#146), Dr_Jeep(#150),
Alan(#151), Jill(#179), Greg(#206), Chris(#214), Jim(#252), Blimp(#253),
Jonathan(#294), Uri(#295), Wave(#297), Christopher(#298), Rancher(#299),
Tom(#300), Jeremias(#352), Hiro(#353), Curator(#393), Seth(#498),
Killer(#532), Henry3(#533), Kris(#534), Wade(#535), Aslan(#536), djs(#537),
Moulton(#538), dickens(#539), ruby2(#540), Eri(#541), Jon(#542),
Example(#543), Tombstone(#544), Reluctant_Meliorist(#545),
Rex_Recursion(#548), Howard(#549), Ben(#550), Mara(#551), yduJ(#552),
Oracle(#553), JackDangers(#554), Pavel(#555), cmk(#556), John(#558),
Paul(#559), |>oug(#560), Tim(#561), Dave(#562), Judith(#563),
Moonshadow(#564), Pattie(#565), Stephen(#566), 'Fred'(#567), Kid_Gloves(#568),
Vaniar(#569), andru(#570), tex(#571), Henry(#572), Gracia(#573), PHreD(#574),
Robyn(#575), Sean(#576), Donna(#577), emr(#578), Adumus(#579), Aaron(#580),
Michele(#120), Randy(#73), Alice(#869), Automoton(#870), Kendy(#871),
Ford(#872), Orwant(#873), Howard_of_AF(#875), Andre(#876), slk(#877),
Antigonus(#878), CarltonJ(#879), Daniel(#880), Rick(#881), Bob(#882),
Melanie(#883), Shevek(#884), Debby(#885), Cricket(#886), jal(#887), Dex(#888),
Margaret(#889), Butoh(#890), Linnaeus(#891), Wayne(#892), Farber(#893),
Max(#894), Networker(#895), CHRONOS(#896), Pardo(#897), M(#898), Scott(#899),
Quantum(#900), Sam(#901), trevor(#902), Larry(#903), Matal(#904), Terry(#905),
Brenda(#906), Reverend(#907), Erik(#908), Jay(#909), Jacques(#910),
Mandren(#911), balder(#912), slim(#913), dobbs(#914), Peter(#915),
Sherry(#916), shkoo(#917), Barry(#918), bambam(#1336), HTTP2(#1337),
Der_Wanderer(#1338), Andrew(#1339), Jimande(#1340), Tenzig(#1341),
China_Blue(#1342), Shohin(#1343), E-Beth(#1344), Ric_Aubern(#1345),
Gareth(#1346), Wilhelm(#1347), Ivan(#1349), Peter Paul(#1350), Mirror(#1351),
Kevin(#1352), Inui(#1353), tina(#1354), Django(#1355), cdr(#1356),
Wilson(#1358), tonami(#1359), Benoit(#1360), Tailchaser(#1361), trust(#1362),
Jamie(#1363), Martin(#1364), Croyd(#1365), Remy(#1366), Gilbraith(#1367),
Mimi(#1368), Shimon(#1369), French(#1371), Ken(#1372), PaulB(#1373),
Red_Tail(#1374), JaysHouseMOO(#1375), Nicholas(#1376), Tuke(#1378),
DavidSan(#1379), Wlad(#1380), Morpheus(#1381), Brian(#1382), Ron(#1383),
Alex(#1384), Gene(#1725), Ice(#1726), Sherrie(#1727), Greybeard(#1728),
Gabe(#1729), Vermeer(#1730), TheJester(#1731), Vision(#1732), Deanie(#1733),
Janet(#1734), Nightwing(#1735), Pierre(#1736), Myriad(#1737), Tolstoy(#1738),
Grant(#1739), Crunch(#1740), Starsinger(#1741), les(#1742), VBanks(#1743),
Penguin(#1744), Thelonius(#1745), Polly(#1746), Wompom(#1747), Thierry(#1748),
Tarot(#1749), bookworm(#1750), Ninja_Librarian(#1751), Code_Walker(#1752),
Steven(#1753), Egnazio(#1754), Arthur_Dent(#1755), Mandrake(#1756),
Wyoming(#1757), Fin(#1758), TimMac(#1759), Moose(#1760), Sandro(#1761),
Rai(#1762), Clark(#1763), Servant(#1764), Randolph(#1765), wytchy(#1766),
Pooh_Bear(#1767), SRI(#1768), Ivan_the_Unstable(#1769), Obie(#1770),
Riddley(#1771), Zeno_the_Scottish_Terrier(#1772), Scooter(#1773), JJ(#1774),
WhiteRabbit(#1829), Master_Quickly(#1830), Paulie(#1831), jimc(#1832),
Phil(#1833), Doc(#1834), Creighton(#1835), Arno(#1836), Bonnie(#1837),
Ralph124C41(#1838), thinknut(#1839), Robert(#1840), Ozmath(#1841),
Mike(#1842), kilroy(#1843), Pitts_Librarian(#1844), Bill(#1845),
Mumford(#1846), Pat(#1847), Phaine(#1848), mday(#1849), D_Policar(#1850),
garbage_chick(#1851), Nomad(#1852), Zobot(#1853), ANTSMAN(#1854), Eric(#1855),
eeyore(#1856), Jian(#1857), Strata(#1858), Marcia(#1859), Spiker(#1860),
Lori(#1861), Woodie(#1862), Arie(#1863), Red_Scharlach(#1864),
TeknoTroll(#1865), Amber(#1866), palmer(#1868), Mouse(#1869), Mazdak(#1870),
Wandering_Clown(#1871), Bud(#1872), MichelleM(#1873), Sclove(#1874),
Balin(#1875), Nikki(#1876), Wintermute(#1877), Daishi(#1878), Rik(#2154),
FreeSide(#2155), Steve(#2156), MC(#2157), Calypso(#2158), Adam(#2159),
Handle(#2160), GregS(#2161), Merlin(#2162), Biff(#2163), Mollusc(#2164),
J(#2165), Rez(#2166), Tower(#2167), Jan(#2168), hiphop(#2169), Marilyn(#2170),
Russell(#2171), Diana(#2172), Peg_Syverson(#2173), ELF(#2174), jt(#2175),
Werner(#2176), Glyn(#2177), rip(#2178), Dagon(#2179), Molly(#2180),
Sparkle(#2181), GreenDog(#2182), Gideon_Tanner(#2183), Scotty(#2184),
Fang(#2185), Evohe(#2186), Flatlander(#2187), Shepherd_Moon(#2188),
Beckster(#2189), Fish(#2190), Indiana_Jones(#2191), sgs(#2192), Javin(#2193),
Druid(#2194), Victor_L(#2195), Sri_Tex(#2196), Slurpee(#2197),
Seduction(#2198), Jon_A(#2199), G.W.(#2200), Arthropod(#2201),
adamfast(#2202), Kane(#2203), lizard(#2462), Scamper(#2463), Neuron(#2464),
Baramere(#2465), Hollis_Weaver(#2466), JR_Bradley(#2468), Catfood(#2469),
Muddy(#2470), Mharie(#2471), Zoro(#2472), Puddleglum(#2473), soundman(#2474),
Dorothy(#2475), Zeednay(#2476), Molly_Grue(#2477), Jordan(#2478),
Aileen(#2479), Beth(#2480), Ed(#2481), Arthur(#2482), kirstnla_of_Dwin(#2483),
Viz(#2484), Simon(#2485), Flame(#2486), (#2487), Buddha(#2488),
Phaedrus(#2489), (#2490), Lee(#2491), (#2492), Adrianus(#2493),
elBloom(#2494), Glenn-El(#2495), Listog(#2496), (#2497), Almond(#2498),
(#2499), (#2500), (#2501), Buzz(#2502), Else_Fogg(#2503), blackadder(#2504),
TomWhore(#2505), Julia(#2506), Locke(#2507), Glenn(#2508), Person(#2509),
Ellen(#2510), and Mandy(#2511).


Bibliography

Benedikt, Michael, ed. (1991).  Cyberspace, First Steps.  Cambridge, MA: MIT
Press.

Bruckman, Amy (1992-3).  "Identity Workshop: Emergent Social and Psychological
Phenomena in Text-Based Virtual Reality."  To appear in the proceedings of
INET ‘93.  Available via anonymous ftp from media.mit.edu in
pub/MediaMOO/Papers/identity-workshop.{ps, rtf}.

Curtis, Pavel (1992).  "Mudding: Social Phenomena in Text-Based Virtual
Realities."  Proceedings of DIAC ‘92.  Available via anonymous ftp from
parcftp.xerox.com, pub/MOO/papers/DIAC92.{ps, txt}.

Oldenburg, Ray (1989).  The Great Good Place.  New York:  Paragon House.

Papert, Seymour (1980).  Mindstorms: Children, Computers, and Powerful Ideas.
New York: Basic Books.

Turkle, Sherry (1984).  The Second Self: Computers and the Human Spirit.  New
York: Simon & Schuster.

Vygotsky, L.S. (1978).  Mind in Society.  Cambridge, MA: Harvard University
Press.

Winner, Langdon (1986).  The Whale and the Reactor.  Chicago: University of
Chicago Press.

Notes

(1) To connect to MediaMOO from the Internet, type:
telnet purple-crayon.media.mit.edu 8888
connect guest
If you have difficulty, send electronic mail to
[e-mail].

(2) For an introduction to MUDs, see [Curtis 92] and [Bruckman 92-3].

(3) The MOO software was developed by Pavel Curtis at Xerox PARC and
generously made available to the public for free.  It can be obtained via
anonymous ftp from parcftp.xerox.com in /pub/MOO.

(4) Some people have more than one character.

(5) This does not count the 33 guest characters and 5 utility characters.
There are a total of 413 character objects in the database.  Of these, 289
have connected within the last month.

(6) This conference room was programmed by Barry (Barry Hayes).

(7) This conference room was programmed by Daniel (Daniel Rose).

Presented at 3CYBERCONF, The Third International Conference on Cyberspace in
Austin, Texas on May 15th, 1993.  Comments welcome via electronic mail:
[e-mail]