To exemplify in more concrete terms what I mean by the term distributed virtual environment, I shall first refer briefly to a type of object-oriented distributed environment generally referred to as multi-user environment[43], object oriented, often abbreviated to MOO. Amy Bruckman and Mitchel Resnick of the Epistemology and Learning Group[44] at the MIT Media Lab, describe their brainchild MediaMOO, one of the longest functioning distributed virtual environments online, as "a text-based, networked, virtual reality environment designed to extend the type of casual collaboration which occurs at conferences to a daily activity.[45] Visitors to a MOO 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."[46]
Actors all over the world connect to distributed virtual environments like Media MOO and interact in real time within, and with, the virtual environment through virtual `representatives', or `agent-objects', negotiating meaning with one another through structured forms of interaction, and constructing, alone or together, new virtual spaces and objects, such as private and public rooms which may be linked to spaces in other virtual environments via a virtual internet and then used as arenas for meetings, teaching, performances, commercial activities etc.. Other types of virtual objects such as notes, recorders, slide projectors, web-browsers, and other tools embodying functions useful for the development and maintenance of the everyday life of the virtual community are being developed and tested in the virtual environment by actors, either alone or together with others, through the mediation of their respective agent-objects.
As an aside: the concept of `agent-object' introduced above is rather new, at least for me, and as such, perhaps not quite fully thought out from epistemological and/or ontological perspective. The actual ontological status of such objects is, for instance, not completely clear. This issue is in turn linked to the degree of autonomy and developmental potential each such object has relative to their respective actor-creators. But it does seem to me necessary to introduce some kind of notion of this kind in such a context as our present one, if for nothing else, in order to account for the fact that any given actor, while linked in various ways to one or more virtual representatives or avatars which, we shall assume, are, or will be, able to operate with varying degrees of autonomy in the socio-semiotic field of the virtual environment, is at one and the same time being represented as both co-present and not co-present for other actors (via the mediation of their own linked agent-objects) in the same environment, since each agent-object is essentially a mathematically or algorithmically generated sign object possessing only a very few characteristic attributes and qualities of its actor of origin, and since the register of possible actions and meaning negotiations which each agent-object is able to perform will be limited in various ways by the relative degree of complexity and functionality of the algorithms describing and maintaining the topology of the virtual environment to which this agent-object `belongs,' and thus also the development of the `social calculus', to use Derek Bickerton's term[47], which serves to regulate, or at least frame, interactions between it and other agent-objects operating there. Since any given actor can, by means of the logging function built into the virtual environment, or into agent-objects themselves, subsequently study the way in which their own agent-objects have actually interacted and negotiated meaning with other agent-objects in the distributed virtual environment, then, in this connection, having a fairly well delineated concept of agent-object seems quite functional from a number of different descriptive, analytical and evaluation perspectives. Some actors also take part in the development of several virtual environments, and even have different agent-objects active in a number of different environments at one and the same time.
In the above figure, it is assumed that Actors 1, 2 and 3 are physically and geographically distant from one another, for instance in different cities or countries around the world. Actor 2 has two independent agent-objects operative in both virtual environments, requiring separate operations for each one in order for them to move or take part in communication or other forms of action and interaction. Actor 3 commands two linked or coordinated agent-objects, one in each environment. Coordinated agent-objects of this kind are designed to make simultaneous monitoring and participation in both environments less complicated, since each agent-object is always `aware' of what the other is doing. If, for instance another agent-object in Environment 1 initiates a conversation with AgO3b, then Actor 3 can participate in this conversation by way of AgO3a, which may be involved in other activities at the same time in Environment 2. Since the two environments in the figure above are physically linked to one another by an Internet portal, they can to all intents and purposes be considered as merely constituting `local' social or cultural fields[48] within the wider global-local field indicated by the - - - stippled circle in Figure 2 above
The nominal message code for text-based virtual environments like Media MOO is written language. Navigation, manipulation and communication in the virtual environment is performed by means of commands - minimal syntactical units. For instance, the command: out will normally move an agent-object out of the virtual locality it is in at the time of the uttering of the command to a linked neighbouring space, while the command say hi! distributes a message from an actor, in this case represented by an agent-object named `player', of the type player says "hi!", to everyone who happens to be co-present in the same virtual space at the time, A short visit to Media MOO may for example proceed as follows[49]:
- *>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 and closet to The LEGO Closet
- You see a newspaper, a Warhol print, a Sun SPARCstation IPC, Projects chalkboard, and MediaMOO Map here.
- Michele, Jade_Guest, Albert, and Guest are here.
- *>"Hello everyone!
- You say, "Hello everyone!"
- Michele waves.
- *>look michele
- A grad student who's carrying various books and papers she should probably be reading.
- She is awake and looks alert.
- Carrying: Directions to the Ball
- *>@whois michele
- Michele's real name is "Michele Evard".
- Michele can be reached via email as: mevard@media.mit.edu
- *>research michele
- Research interests:
- General Area: Learning, Technological Tools for Learning. Current projects: kids & LEGO/Logo, kids & telecommunications.
- Albert waves.
- *>@research me is "Testing MediaMOO"
- You set the "research" message of Example (#543).
- *>"How did you wave, Albert?
- You say, "How did you wave, Albert?"
- Albert says, "i did :waves."
- *>"Thanks.
- You say, "Thanks."
The everyday life that goes on in the mediated social environment of a MOO community depends for its existence on a continuity of semiotic systems, all of which contribute in various ways to the instantiation of the social and cultural fields in the virtual environment where systems of norms for action, interaction, communication (and meta-communication) develop over time
One such semiotic system is the syntax of the command code exemplified in part in the snippet of transcript above, used by agent-objects for speaking and acting in (and on) the virtual environment. Communicative and other kinds of functions coded into this system determine to a large extent how, and which, forms of interaction may be initiated, carried out and regulated in the virtual environment by participating actors and their respective agent-objects. In principle, since the potential number and categories of algorithmically based communicative and performative functions which may be built into both agent-objects and all other types of objects that populate and describe the virtual environment is indeterminate, they are in principle open for an infinite degree of future development, as are the programming languages used to create and develop these functions.
Another indeterminate semiotic system influencing in fundamental ways the conditions for further development of the mediated everyday life of the actors, agent-objects and objects of the virtual environment is the low-level machine code on which the host system itself runs. This determines to a large extent just how effectively any of the other semiotic systems operating at even higher levels of complexity and visibility will function within the virtual environment. Again, this system is in principle open for a potentially infinite degree of future development, as is yet another semiotic system which is constituted by the interrelations and interactions between the various types of Internet protocols which regulate the way in which text- and other types of utterances are translated and interpreted as they are piped from around the world into the social space of the virtual community.
All these indeterminate, and in a sense artificial[50] semiotic systems are in a sense `inherent' in the technology of the medium itself, and they all interact in various kinds of ways with the equally indeterminate linguistic, social and cultural semiotic systems which are brought into the social space of the virtual environment by the plurality of migrant actors involved. In doing so, these materially instantiated semiotic systems contribute to the constitution and development of textual and interactional norm systems[51] operative within the shared virtual environment. Here, however, it is important to note that there is also a certain degree of reciprocity, in the sense that there is a migration of systems of norms in both directions, i.e. between the virtual environment and the `real world', something we shall discuss in a bit more detail later on.
[43] or dimension or dialogue
[44] See the home page of the Epistemology and Learning Group at http://el.www.media.mit.edu/groups/el/
[45] To connect to MediaMOO via the Internet, start up a terminal application, and type:
telnet purple-crayon.media.mit.edu 8888 , followed by connect guest when the login prompt appears. The following is also functional: telnet mediamoo.cc.gatech.edu 8888. See also connection information at the MediaMOO home page at http://www.cc.gatech.edu/fac/Amy.Bruckman/MediaMOO/connect.html
[46] Bruckman & Resnick 1995, 1
[47] The term social calculus is, I believe, Derek Bickerton's. See Bickerton & Calvin (forthcoming), online draft version at http://www.williamcalvin.com/bk-lingua/, and especially http://www.williamcalvin.com/bk-lingua/LEMch11.htm and http://www.williamcalvin.com/bk-lingua/LEMch12.htm for discussion of the role of reciprocal altruism as a precursor to argument structure via the development of social calculus from which thematic roles may be derived.
[48] Elsewhere (see for instance Coppock 1996a), I have discussed the applicability of the literary notion of genre to the problem of describing the emergence of systems of textual and interactional norms in distributed virtual environments, especially with reference to the often quite different kinds of conceptualizations of basic purpose and function found across a fairly wide range of such environments.
[49] Commands typed into the virtual environment by the visitor are prefixed in the transcript by *>, and are followed by the appropriate response generated by the system.
[50] For discussion of the role of artificial and natural languages in the space between computation and biology see Marcus1998b.
[51] For discussion of the general notion of textual norm systems and processes of textual norm system change see Berge 1993, and in relation to distributed virtual environments see Coppock 1995, 1996a, 1996b, 1997a, 1998; Coppock & Violi 1999.