« Meta Grinding and Nemo | Main | Sticky NeoPets »

Jun 18, 2005

Comments

101.

At the risk of resurrecting the undead italics...

1) Do you think that the experience of using graphical worlds compared to text worlds is greater or less than the same comparison between movies and novels? Should either of the differences be classified as fundamental?

What exactly is the metric for "greater" or "lesser"? On what axis are we measuring? I almost always find more powerful experiences from books than from movies, personally, but then books are generally much longer and richer in story... they certainly pale in terms of visual experience, but the visuals are not really the chief area that I get the experience from in movies... the visuals are there in service to the story.

So I have trouble saying that it's fundamental or not in those cases... obviously, the visuals are fundamental to the medium of film, but what is the "greater" experience? I have no idea. I frequently find movies to be more manipulative than books--perhaps it's more overtly manipulative. :)

2) On the continuum of complexity (or, perhaps, freedom of expression), is it a fair statement to say that movies offer more complexity than text and that games offer more than movies? (Nota bene: "games" in this context includes text and graphical games)

Again, complexity on what axis? Complexity of interpretation? It's generally agreed that film versions of stories are usually simpler--usually because they are adaptations of novels getting reduced down to novella length. As a result much plot and characterization tends to be lost. Games are typically even simpler in terms of their content from an interpretive point of view.

Complexity of interaction? Arguably, games come on top, followed by text and movies are on the bottom on that particular scale; movies by their structure seem to offer less room for interpretation than text, which is the main point that Richard and other pro-text people seem to be making...

My point in all this is just that there's a lot of ways to measure here, and asserting primacy of one medium over another is almost certainly going to require pinning down the domain of said primacy better. Text and graphical games both have things that they do better than the other.

102.

Oops, poor proofing. s/greater or less/more or less different/

Raph, yes, this is why both questions made for interesting -- and extensive -- discussion and debate. What became clear to me is that since my primary interest is in building and interacting with games and worlds, that the topic is really about perception and interaction. Thus my contention that text interaction is fundamentally different than 3D movement. Certainly YMMV if other areas interest you more.

Also, to be clear, I'm not arguing that "more complexity" == "better", especially after just finishing "The Inmates Are Running the Asylum" by Alan Cooper (a great read, btw).

103.

Cory Ondrejka wrote:

1) Do you think that the experience of using graphical worlds compared to text worlds is greater or less than the same comparison between movies and novels? Should either of the differences be classified as fundamental?

2) On the continuum of complexity (or, perhaps, freedom of expression), is it a fair statement to say that movies offer more complexity than text and that games offer more than movies? (Nota bene: "games" in this context includes text and graphical games)

For some reason this discussion has reminded me of a group of (ancient) greek philosophers discussing which is more important, eyesight or hearing.

However, this latest direction could provide some concrete answers, if I'm allowed to twist the questions...

1) Rather than asking about "greater/lesser" or whether the differences are "funademtal", twist the question into "How does the experience of a graphical VW differ from a textual VW?"

The answers to this a potentially useful to both camps (pro-eyesight and pro-hearing) because it helps both sides understand their strengths and weaknesses.

2) You could reword "complexity" and "freedom of expression" into, "Can movie A vary more from movie B, than game C from game D?" (IE: How much variation is there amongst members of a class.) This too could be rephrased: "How can a graphical world vary that a text world cannot, and vice versa?"

104.

Mike>1) Rather than asking about "greater/lesser" or whether the differences are "funademtal", twist the question into "How does the experience of a graphical VW differ from a textual VW?"

Easy and discussed earlier in the thread. The 'graphical' vw relies on local modelling of the world simulation (some neighborhood near the avatar) to communicate the vw's states of affairs, whereas textual VWs rely on descriptive/propositional content to do the same.

105.

Kirk>I about 80% agree but feel that we need to take it just a bit further beyond just the "observer."

I don't see why. At the level of virtual world if we have

World Model>Client/Avatar>User

then we have the whole enchillada. The world model is both necessary and essential. The client/avatar is necessary but what form the client/avatar takes is not. The presence of a user is not necessary, since it's just a token of an observer, since are conceivable observers that are not 'users'.

106.

Monkeysan>Well, I really hope you don't mean that literally, since NPCs don't have a point of view (at least not yet).

How do you know that anyone else on the planet has a point of view? All you know is that you have a point of view. If you're willing to ascribe a POV to other people, isn't it merely a case of how far you're willing to stretch your definition before ascribing POVs to NPCs?

>Rather the issue is that without the observer, there is no virtual world at that point in time.

But there is: we can point to the code and say there must be.

>Clearly, gamers are not simply asking for a description of the state of the world model. They are asking about other things that lie completely outside the world simulation.

If they're asking about what the players are doing, yes. If the virtual world were sufficiently rich, though, they could be wondering whether or not the Armies of the North had beaten the Armies of the South and their goblin allies.

There's plenty that can go on in a virtual world that players could find interesting without there having to be other players present. Come to that, there may be things that NPCs could find interesting (for certain definitions of "interesting"...).

A virtual world isn't a quantum wave that collapses only when observed, except insofar as the entire universe is. It's basically a highly complicated piece of clockwork that keeps running (just like a clock) when no-one looks at it. Now if you want to argue that clocks only exist when people look at them, OK, you can do that, but then you're pretty well arguing that everything but you only exists when you look at it. This is a valid philosophical point of view, but it means virtual worlds are no more or no less real than the rest of reality.

>apparently that wasn't the end of the italics bracket you closed earlier =]

Thansk. My guess is that a close bracket was typed as an open bracket, so there were two to close instead of just one.

Richard

107.

I have been working on a little text based virtual world - I can it COW - Creative Object World. I'd say the main reasons for using text not graphics (mind you graphics are allowed in an iconic way in COW) are:

1) greater accessibility (people who cant see can have the site read to them)
2) leave it all up to the imagination - try drawing a huge elephant and putting it in a tiny matchbox.

Oh yes, its smaller, faster, can run in any browser without plugins and provides more creative possibilities.

Feel free to check out http://www.wolispace.com/cow

108.

Cory: By the way, Kirk, after spending the end of last week at the GLS conference listening to Henry Jenkins, James Paul Gee, Kurt Squire, and others, I'm again curious about your certainty that learning is technology independent.

Well, at this point I'll just pull a John B. Watson and point out that I'm going a bit beyond my evidence. But so have the "new media" folk and they've been getting away with it for many years.

Don't you think that if visual authenicity was the primary thing that mattered, that we would have had a knock-down victory for instructional film 70 years ago, and instructional television 30 years ago? After all, in terms of graphics, the 3d virtual lens is still just a cartoon compared to what is possible with a camera.

The problem is that learning is more than just presentation. Learning is inquiry, practice, locus of control, improvisation, and social cognition. The media you use is really only a small part of the picture.

Also, by your lack of answer to the appendectomy question, I presume that your choice would be to learn from a book (or that all of the options would equally prepare you for surgery if you only had access to one)?

Well, I don't think this is really fair. I gave you a clear answer, and I'll repeat it here: It depends on what you want to learn, and how you want to learn it. I think the methods that work great for teaching surgery, are not so great at teaching people to be better short-story writers. The former learning task works better with authentic practice, the latter learning task works better with text-based history and reflection.

Pick the tool to match the task. Don't pick a tool and expect it to excel at every task.

1) Do you think that the experience of using graphical worlds compared to text worlds is greater or less than the same comparison between movies and novels? Should either of the differences be classified as fundamental?

In my mind, a bit less. But of course, I think the narrative is more important than the media in both equations. You can create open narratives in all four media, and you can create closed narratives in all four media.

2) On the continuum of complexity (or, perhaps, freedom of expression), is it a fair statement to say that movies offer more complexity than text and that games offer more than movies? (Nota bene: "games" in this context includes text and graphical games)

No, I don't think so. I think that they offer different kinds of complexity and freedom of expression for the artist. Text offers higher degrees of subjectivity, so you can do something like As I Lay Dying where each new chapter is the internal monologue of a different character. Movies can offer pantomime such as Modern Times or color associations such as Blue. Interactive media I think offer a different level of constraints and afoardances.

109.

Cory: What became clear to me is that since my primary interest is in building and interacting with games and worlds, that the topic is really about perception and interaction. Thus my contention that text interaction is fundamentally different than 3D movement. Certainly YMMV if other areas interest you more.

Certainly. And my primary interest involves community. Which I suspect is the core of our disagreement.

monkeysan: I don't see why. At the level of virtual world if we have

World Model>Client/Avatar>User

Well, the way that I would describe it is:
World Model <-> User

With that arrow being some medium through which the user changes the world, and is changed by the world. One of the expectations I have of the real world is that dirty dishes remain in the sink until I wash them. I have similar expectations of my virtual worlds that my actions change that world to a modest degree.

Ideally, the medium of interaction should be transparent and unnoticed.

110.

Kirk> Pick the tool to match the task. Don't pick a tool and expect it to excel at every task.

I agree.

*But*, as Johnson spends some time doing, we have a history of analyzing new media forms in terms of the old, and that this approach often leads us astray. Similarly, I think it is safe to say that often try to create new media in terms of the old, which can be limiting. Like the examples in Christensen, new and disruptive technology is often *worse* at many tasks that existing tech has mastered, but the new tech displaces the encumbents anyway once people fully exploit new design options that it makes available.

I bet even Richard and I could agree on the following statement:

"Today, graphical virtual worlds provide less complex narratives than text worlds."

(OK, Crawford would say that they both suck for actual storytelling, but that's another thread!)

I'm just not so certain of the statement if you remove "Today" :-)!

111.

Oh, I hate insomnia, and those ideas where I just can't let go.

Here is my previous two-axis model of how I see these issues exploded into bad ASCII art.


|Low-Sociability| High-Sociability
_________|_______________|_________________
Language | alt.flame | LambaMoo
_________|_______________|_________________
Graphics | Quake | Second Life
_________|_______________|_________________

So the question I have is, to what degree do Graphical environments with high sociability require different design structures and online politics than Linguistic environments with high sociability?

I don't think that increasingly greater fidelity in simulation and representation are necessarily the key. American history is littered with attempts to build utopias that fizzled and turned ugly. I need to find an article from Communities magazine that claimed something like less than half of intentional communities made it past the 10-year mark.

*But*, as Johnson spends some time doing, we have a history of analyzing new media forms in terms of the old, and that this approach often leads us astray.

I think though that treating new media as "revolutionary" rather than "evolutionary" tends to lead to different types of blindness. Knowing that forms of abbreviated "chat text" have been around since the beginning of the telegraph would certainly quell some fears about online synchronous discourse.

112.

Your affiliate form at http://secondlife.com/corporate/affiliates.php is broken :)

113.

So, by your reasoning, quake isn't fundamentally different than alt.flame?

American history is littered with attempts to build utopias that fizzled and turned ugly. I need to find an article from Communities magazine that claimed something like less than half of intentional communities made it past the 10-year mark.

Sure, but the economics of meatspace are fundamentally different than digital worlds -- or, at the very least, the results of negligible marginal costs of reproduction, storage, etc combined with incredibly cheap communication profoundly change production. How many utopias crashed and burned because the utopians starved or ran out of money?

114.

Richard>How do you know that anyone else on the planet has a point of view? All you know is that you have a point of view. If you're willing to ascribe a POV to other people, isn't it merely a case of how far you're willing to stretch your definition before ascribing POVs to NPCs?

Well, I'm not going to recreate the argument against solipsism and for the existence of other minds here.

Richard>But there is [a virtual world w/o observer]: we can point to the code and say there must be.

I don't know what you mean by "point to the code and say there must be." The code itself isn't a virtual world. The world model that is instantiated by runnning the code is the issue. You seem to think that a virtual world is identical to the world model. I disagree. See the lengthy discussion between myself and Raph to see why.

Richard>If the virtual world were sufficiently rich, though, [when players ask about what's going on in some vw] they could be wondering whether or not the Armies of the North had beaten the Armies of the South and their goblin allies.

That misses the point. No matter how rich your world model gets, there will always be events, relations, and features of the virtual world that have no representation in the world model.

As for what NPCs find interesting, we'll know that when NPCs become conscious agents. At that point I'll just ask. But you and I are almost certainly not going to get that chance in our lifetimes.

Richard>A virtual world isn't a quantum wave that collapses only when observed, except insofar as the entire universe is. It's basically a highly complicated piece of clockwork that keeps running (just like a clock) when no-one looks at it.

I never said a virtual world is a quantum wave. I said that a virtual world is like Schroedinger's cat in that an observer is required to make it's existence determinate. The argument for saying that I have supplied in several places throughout this thread, in particular in my exchange with Raph.

As for a virtual world being like a clock that keeps telling time even when no one looks at it, that begs the question at issue. Your analogy only works if we assume that virtual worlds are identical to their world models, since that's precisely what's at issue, you are begging the question. I fully agree that a world model can keep running even when 'no one's looking'.

115.

Richard>How do you know that anyone else on the planet has a point of view? All you know is that you have a point of view. If you're willing to ascribe a POV to other people, isn't it merely a case of how far you're willing to stretch your definition before ascribing POVs to NPCs?

Well, I'm not going to recreate the argument against solipsism and for the existence of other minds here.

Richard>But there is [a virtual world w/o observer]: we can point to the code and say there must be.

I don't know what you mean by "point to the code and say there must be." The code itself isn't a virtual world. The world model that is instantiated by runnning the code is the issue. You seem to think that a virtual world is identical to the world model. I disagree. See the lengthy discussion between myself and Raph to see why.

Richard>If the virtual world were sufficiently rich, though, [when players ask about what's going on in some vw] they could be wondering whether or not the Armies of the North had beaten the Armies of the South and their goblin allies.

That misses the point. No matter how rich your world model gets, there will always be events, relations, and features of the virtual world that have no representation in the world model.

As for what NPCs find interesting, we'll know that when NPCs become conscious agents. At that point I'll just ask. But you and I are almost certainly not going to get that chance in our lifetimes.

Richard>A virtual world isn't a quantum wave that collapses only when observed, except insofar as the entire universe is. It's basically a highly complicated piece of clockwork that keeps running (just like a clock) when no-one looks at it.

I never said a virtual world is a quantum wave. I said that a virtual world is like Schroedinger's cat in that an observer is required to make it's existence determinate. The argument for saying that I have supplied in several places throughout this thread, in particular in my exchange with Raph.

As for a virtual world being like a clock that keeps telling time even when no one looks at it, that begs the question at issue. Your analogy only works if we assume that virtual worlds are identical to their world models, since that's precisely what's at issue, you are begging the question. I fully agree that a world model can keep running even when 'no one's looking'.

116.

My bad I seriously apologize.

117.

Kirk, btw: I think my short hand use of '>' instead of bidrectional arrows confused you. I mean bidrectionality when I use '>'. I just got tired of rewriting all the '<->' from earlier posts.

I think that clears up our misunderstanding.

118.

Ran across an interesting article: Steven Yantis, a researcher at Johns Hopkins University, has obtained some results suggesting that people can't pay full attention to both visual and audio information simultaneously.

The research is mentioned as Multitasking: You Can't Pay Full Attention to Sights, Sounds in a Johns Hopkins news release.

Is the audio/visual shift that Yantis describes best seen as a simple difference between hearing and seeing? Or is it a conceptual shifting between linguistic and spatial perception, and thus something related to text vs. graphics?

--Flatfingers

119.

Cory: So, by your reasoning, quake isn't fundamentally different than alt.flame?

To the extent that both involve human participants competing with each other for status through a computer medium, no.

Sure, but the economics of meatspace are fundamentally different than digital worlds -- or, at the very least, the results of negligible marginal costs of reproduction, storage, etc combined with incredibly cheap communication profoundly change production. How many utopias crashed and burned because the utopians starved or ran out of money?

I'm thinking about the economics and politics inside the space. So for example, I've seen spaces fail because a small group was hogging all the resources, leaving no room for novice participation. I've seen groups undergo the equivalent of a "civil war" with reprecussions that made the space feel uncomfortable to novices for months, and even years after the original conflicts.

So you raise a chicken and egg question. Some communities fail because they ran out of money, however in many cases they ran out of money because of internal political problems.

120.

monkeysan: I think that clears up our misunderstanding.

That does, thanks.

121.

Is the audio/visual shift that Yantis describes best seen as a simple difference between hearing and seeing? Or is it a conceptual shifting between linguistic and spatial perception, and thus something related to text vs. graphics?

It's interesting to note that this study apparently only tested visual information that was also textual, while they simultaneously provided aural input that was identical. From this, I don't see how they draw the conclusion that this explains why people can't concentrate on driving and being on a cell phone at the same time.

My own hypothesis is that people cannot divide their attention effectively between the spatial-causal model of the world and their interaction with it that they build in their brain and descriptive input like voices or text.

122.

monkeysan> From this, I don't see how they draw the conclusion that this explains why people can't concentrate on driving and being on a cell phone at the same time.

I had a similar reaction, but I wanted to see if others felt the same way.

It seemed Yantis et al. were focused on how the human brain receives information rather than how it processes information: sight vs. sound, rather than language vs. spatiality. That was probably done to minimize variables -- they consistently presented linguistic information, but altered the form of presentation.

So their test was more like comparing books to audiotapes. But driving isn't about processing linguistic information; it's about processing spatial information. If the typical human brain is capable of performing both these types of processing simultaneously, then the results cited by Yantis could be misleading, and the driving-while-talking problem might turn out not to be a problem after all.

But what if we're not so good at doing both linguistic and spatial processing simultaneously? More to the point of this conversation, what if some people tend to be much better at one type of processing than another type? Isn't there already some evidence (from testing) that women tend to outperform men on linguistic reasoning, while men tend to outperform women on spatial reasoning?

What would brain studies in this field say about how games should be designed? If a linguistic/spatial difference can be confirmed, does that imply that developers should aim linguistic games (like MUDs) at women, while 3d graphical games should focus on attracting men as their natural constituents?

Or is that how things work already?

It's just one data point, but a recent "Astromech Stats" from SWG reveals that female personae are chosen approximately 1/3 as often as male. If that's an accurate reflection of the actual sex of the player, is it reflective of the player base of most graphical MMOGs? Is there demographic information available on MUDs that shows a more equal distribution of the sexes, or even greater popularity of MUDs among women?

And does this have any bearing on Kirk's question: "to what degree do Graphical environments with high sociability require different design structures and online politics than Linguistic environments with high sociability?"

--Flatfingers

123.

Rozak>How does the experience of a graphical VW differ from a textual VW?

Perhaps it is best I explain this as experiments to ponder…

Experiment #1:Take a space, like an apartment - a living space. Create a model of that living space, and allow it to be accessible through a corporeal proxy available via text or graphics interface. The text interface -describes- what the corporeal proxy is seeing. The graphics interface allows one to see what the corporeal proxy is seeing. Select two friends who have not experienced the real space. One friend uses the text interface, and another friend uses the graphical interface. Give each of them tasks to accomplish in the virtual space. When they have completed the tasks, introduce them to the real apartment. Have them perform the same tasks in the real space. Is there a difference in their performance? Speak with them afterwards and ask them, what was different in performing tasks between the virtual and the real world? Compare the two answers. Do the experiment again, with two new people, but blind-fold them before they enter the real space. Is there a difference in the performance?

Experiment #2:Get three people who have not experienced either the real or virtual spaces, then have each respectively interact with each realization, that is, the first person has real, the second graphics, and the third text. Ask each of them to go into the guest bedroom and interact with it and look around. Ask each of the subjects to describe in text what they experienced. Ask each of the subjects to draw a picture of the room they experienced. Ask others not involved in the exercise if they can tell which one experienced the real, the graphics, and the text based on the text descriptions. Than ask again with the pictorial description. Do they answer correctly? How do the answers change between the two?

Experiment #3: Get four people. Allow both a text and graphics interface to interact with the same virtual space. All users are able to interact with each other in the same space, two using a graphics interface, and two using a text interface. When one executes an action, the others are able to observe it, regardless of interface. What processing demands are required of each interface? What differences in implementation are required of each interface for the users to interact? Provide tasks for the users to perform alone. Provide tasks for the users to perform together. Do some users and pairings perform better or worse at some tasks than others?

Although graphics interfaces have greater potential towards visual expression, I find that the singular modality (ie. text only rather than graphics, text, sound) of text allows for a greater potential of overall expression than the singular modality of graphics. I can describe other senses through graphics, like smell, touch, taste, hearing, and emotion but I can be more readily precise and expressive with such descriptions in text. Conversely, I can be expressive with visual descriptions and spatial relationships in text, but with graphics I can be more readily precise. Graphics have a greater potential towards spatial and visual fidelity than text. That is, I am more able to accurately reproduce the effect of visualizing and moving through a space in graphics than in text. Further, in text, I am much more able to control subjective significance than in graphics. For example, in text, I can say that a car looks hip. In graphics, I can show them a car I think is hip in hopes the viewer finds that it is hip. Experiment #2, by the way, illustrates subjective significance and relates to how we remember and organize our experiences. This puts more of an emphasis on how an experience is remembered than on how it is actually experienced.

124.

Monkeysan>Well, I'm not going to recreate the argument against solipsism and for the existence of other minds here.

OK, but at least answer the question: if someone you don't know exists observes a virtual world, does that make it a virtual world?

>I don't know what you mean by "point to the code and say there must be."

What I mean is that it's possible to deduce from the existence of the code (plus the data) that the virtual world exists. You don't have to observe it - indeed, it could never have been observed by anyone - for you to know it's there.

>The code itself isn't a virtual world. The world model that is instantiated by runnning the code is the issue.

There are three levels here:
1) The code plus the initial data. This combination describes all possible future virtual worlds for this world model.
2) The code and data as it is as this very instant. This is one of the possibilities defined in 1), and is what I mean by the "virtual world".
3) The construct in your head that you have built having observed the virtual world. This includes any relationships, emotions or other interactive data you may have accumulated (even though this may be wrong - that other player you think you have a rapport with could be a mindless bot that has thus far struck lucky in its interactions with you).

>You seem to think that a virtual world is identical to the world model. I disagree.

You seem to think that a virtual world is a construct inside the head of the observer. If we're accepting that other minds exist, then I disagree.

Is a novel merely words on the page, or the words plus the imagery and emotional responses of the reader, plus their imagined dialogue with the author? Well, when I go into a shop and buy a novel, I only get rhe words on the page. The rest is what I make of it. When you play a virtual world, you only get to interact with the code. The rest is what you make of it.

>That misses the point. No matter how rich your world model gets, there will always be events, relations, and features of the virtual world that have no representation in the world model.

Suppose we had a world so rich that interesting and exciting things happened in it, but that no-one could play. People could control their own personal, invisible cameras so they could watch what was going on with mounting levels of engrossment, identifying with particular NPCs or rooting for a particular kingdom. They might set up web sites to discuss what was happening, and offer their predictions of how things were going to pan out. These would be external to (what you call) the world model, though. Would that world model therefore exactly correspond with your notion of a virtual world? Or would the fact that people became emotionally engaged in it stop it from being a virtual world - even though players had no ability to interact with it whatsoever?


>As for what NPCs find interesting, we'll know that when NPCs become conscious agents.

And by "conscious agents" you mean..?

If a world model is being used only by a chimp being trained to farm adena, is the chimp at a sufficient level of consciousness to prevent the world model from being equivalent to a virtual world in your scheme? If so, what if it were something less clever such as a parrot trained to peck at keys? Where do you draw the line?

>a virtual world is like Schroedinger's cat in that an observer is required to make it's existence determinate.

But no single observer can possibly observe more than a fraction of it at once. Does the rest not exist?

>Your analogy only works if we assume that virtual worlds are identical to their world models, since that's precisely what's at issue, you are begging the question.

I was trying to use the clock analogy to show that virtual worlds are identical to their world models (as you call them - I call that level their "instantiation" or "instance", but that's just a terminology thing). What is it about a virtual world that means is isn't just an incredibly complicated clock?

Richard

125.

Cory>"Today, graphical virtual worlds provide less complex narratives than text worlds."

I'd agree with this, although I'd also say that narrative has only a minor place in virtual worlds; they're not "about" narrative.

>I'm just not so certain of the statement if you remove "Today" :-)!

You're determioned to make this thread the longest ever on TN, aren't you? (grin)

In order to answer (or maybe not!) your question, I need to ask you a question of my own: what narratives does the real world provide?

Richard

126.

Flatfingers>Is the audio/visual shift that Yantis describes best seen as a simple difference between hearing and seeing?

The problem I have with that research is that the seeing and the hearing are both going to the same place - the brain's language centres. If you were listening to music while reading letters, or looking at a landscape while hearing letters, the results might not be the same. In my case, I know the results would be different, but I may be a little unusual here...

>Or is it a conceptual shifting between linguistic and spatial perception, and thus something related to text vs. graphics?

In that case, wouldn't we expect to see a difference between the reactions of men and women, given that study after study has shown that "most men" are better at spatial tasks than "most women", and "most women" are better at linguistic tasks than "most men"?

Richard

127.

I almost (almost!) hate to chime in on the other side of this discussion, but...

Richard> when I go into a shop and buy a novel, I only get rhe words on the page. The rest is what I make of it. When you play a virtual world, you only get to interact with the code. The rest is what you make of it.

But isn't a virtual world that is explicitly designed as a multiplayer system a different kind of thing than a book?

In fact, isn't a virtual world that is functioning as designed required to be "about" people interacting with other people, rather than being about individuals interacting with a system such as a single-user game or a book? (Note that while a book can be experienced by multiple people simultaneously, that's not the same thing as people interacting with each other.)

Why then is it unreasonable to conclude that what's in the heads of users of multiplayer code isn't also part of the system? If the whole point of the code is to serve as a mediating entity, how can that code be a complete world without users (and what's in their heads)?

> If you were listening to music while reading letters, or looking at a landscape while hearing letters, the results might not be the same. In my case, I know the results would be different, but I may be a little unusual here...

If so, then as long as we're admitting unusualness I suppose I have to raise my hand as well, because for most of my life I've been able to demonstrate two of the three traits you described.

I've tried to explain to family and friends, but they've never quite seemed to understand how important it is for me to be able to concentrate, and how difficult that is for me when there are words being spoken nearby. Back in college I started my collection of "music without words": Mike Oldfield, Tangerine Dream, Jean-Luc Ponty, Jean Michel Jarre, J.S. Bach, Vangelis, some movie soundtracks, and most things off the Windham Hill label. And I still listen to these artists' works today because they still serve as pleasing "white noise"; they don't distract from whatever linguistic thing I'm concentrating on creating.

Interestingly (maybe), there are a few artists I can listen to who do use words, but the words don't seem to distract: Enya, Clannad, and Sigur Ros come to mind... and I suspect it's not coincidence that these artists spend some or all their time singing in a language that's not English. (My Gaelic is virtually non-existent and my Icelandic is worse.)

And this is the first time I've ever run into anyone else who can do the "white noise in the ears" trick at will (or even understands what I mean when I try to describe it). Although for me it's not about cancelling ambient conversation; I tend to use it solely to temporarily block painfully loud sounds (jet aircraft engines, fire alarms, etc.). The only way I've ever figured out how to describe it to other people is to say that it's what you hear at the end of a really big yawn.

Never did learn to switch off my hearing, though. (Although my wife claims I do something like that all the time....)

But to come back to the subject:

> wouldn't we expect to see a difference between the reactions of men and women, given that study after study has shown that "most men" are better at spatial tasks than "most women", and "most women" are better at linguistic tasks than "most men"?

Yes, I would expect to see evidence of such differences in behavioral choices in many aspects of life, including what games we choose to play.

A study that measures and compares preferences among a wide range of game types might be worthwhile here. Has something like this been done already?

--Flatfingers

128.

Richard: You seem to think that a virtual world is a construct inside the head of the observer. If we're accepting that other minds exist, then I disagree.

Well, to push back on this, there are quite a few spaces out there in the "real world" that are intersubjective, a possibility that we have not really discussed. A cafe is a cafe because there is a shared social consensus, mediated by the objects in that space, that there are roles to be played in that space, and people to play those roles.

There is a lot of "worldliness" that centers not on fidelity, but on participation and engagement with objects and spaces.

Suppose we had a world so rich that interesting and exciting things happened in it, but that no-one could play. People could control their own personal, invisible cameras so they could watch what was going on with mounting levels of engrossment, identifying with particular NPCs or rooting for a particular kingdom.

I'd call this television. ;) Certainly you could argue that things like television that give the participant no control over what happens in the setting are virtual worlds, but I don't find them very interesting without interactivity.

I'm wondering how much of this discussion is understood if we knew where the participants were coming from. I see "virtual worlds" as online communities with a richly detailed sense of place. I'm wondering if others are seeing it as a novel or game with interactivity and community just bolted on.

In order to answer (or maybe not!) your question, I need to ask you a question of my own: what narratives does the real world provide?

I go to the cafe, meet a waiter, order coffee, the coffee is served, I pay for it and leave. Not a very deep or profound narrative, but still a narrative.

129.

Richard>OK, but at least answer the question: if someone you don't know exists observes a virtual world, does that make it a virtual world?

As long as they actually do exist, sure.

Richard>it's possible to deduce from the existence of the code (plus the data) that the virtual world exists. You don't have to observe it - indeed, it could never have been observed by anyone - for you to know it's there.

That deduction is only available if you assume that the world model is the virtual world. But that's precisely what's at issue.

Richard>The code plus the initial data. This combination describes all possible future virtual worlds for this world model.

It can't possibly do that, since the code and the initial data can't be used to predict the actions of players in the virtual world, many of which actions affect the state of the world model.

Richard>The code and data as it is as this very instant. This is one of the possibilities defined in 1), and is what I mean by the "virtual world".

How can a 'virtual world' be an instantaneous state? If there's one thing that worlds require, whether virtual or otherwise, it's persistence over at least some time interval.

Richard>You seem to think that a virtual world is a construct inside the head of the observer. If we're accepting that other minds exist, then I disagree.

No, I don't think a virtual world is a construct inside the head of an observer. Far from it.

Richard>Is a novel merely words on the page, or the words plus the imagery and emotional responses of the reader, plus their imagined dialogue with the author?

A novel is a set of descriptions of states of affairs. The reader uses those descriptions to build an internal world model where those states of affairs obtain, and the reader evolves that model as she reads the novel. But the novel itself is a set of descriptions, not a model.

Richard>When you play a virtual world, you only get to interact with the code.

I don't know anyone other than developers who've ever interacted with 'the code'. When the rest of us interact with a virtual world we do so via interactions with representations. These representations are instantiated by running code, but they are not 'the code' itself. I also often take myself to be interacting with other players when I'm playing with a virtual world, and many of these players are clearly not 'the code'.

Richard>Would that world model therefore exactly correspond with your notion of a virtual world? The fact that [the players] can observe the world model and have control over their avatar's behavior (even if that avatar is just a disembodied point of view) is enough for me. Even in this case there are features of the virtual world that aren't represented in the world model--chief among these, of course, is the 'what-it's-like-to-be-in-that-virtual-world' feature. The world model doesn't have a representation of that. In fact, it's not possible for it to do so.

Richard>And by "conscious agents" you mean..?

Agents that are capable of recognizing the mapping between the behavior of the world model and some imgained world that behavior is purported to represent. The coordination of the two yields the virtual world.

Richard>But no single observer can possibly observe more than a fraction of it at once. Does the rest not exist?

That's a good question. That really is the tree falling in the forest question for virtual worlds. It's a question that's much easier to answer for the real world. Nevertheless, I'd say yes, even in the virtual case.

Richard>What is it about a virtual world that means is isn't just an incredibly complicated clock?

Well, the primary difference is that one is an object and the other is a (synthetic) place.


130.

Sorry, since handles have been criticisized of late, I should point out that 'monkeysan' is identical to Aaron Ruby. =]

131.

This whole discussion feels silly at this point. We're going to end up at a philosophical impasse.

Aaron aka Monkeysan is basically saying that the wonderful magical thing that happens when a human agent interacts with a world model is what a "virtual world" really is. To do so, he's reifying a highly abstract concept, a particular sort of perceptual model. This is beyond the position that someone like Kirk takes, which does a much more conventional reifying of community.

The more mechanistic among us are saying that the world model is sufficient; we're probably influenced by the fact that we build the damn things, so we'd rather talk about concrete matters.

It's probably pretty safe to say that all of us agree that all these levels exist. Debating where the soul lives, in the heart or the head or not at all, is approaching casuistry.

So my question is, again, what are we trying to solve (sure doesn't seem to be Cory's original question)? Or are we hardening into dogmatic positions here?

Perhaps we should label the parties the mechanists and the interactionists and start another narratology/ludology war? :)

Lastly, FWIW, I too can do the white noise thing, and have been doing it since I was seven. :) I have always suspected it's something common that most people don't mess with. I don't USE it for anything though.

132.

I think that the discussion about what a virtual world came up in the effort to understand what we are saying when we ask, "Is there a fundamental difference between text and graphically based virtual worlds.

As for the implication that what you call mechanists are some how more engaged in more concrete matters, I'd take issue. I'm not reifying a highly abstract concept any more than we reify the mind when we say that the mental supervenes on the physical.


Also, Raph, the world simulation (model), where you locate "the virtual world", is is also just as abstract as what I'm suggesting. The world simutation is not just the code or even the code plus some data. The world simulation is the execution of the code plus the mapping between the physical state of the machine runring the simulation and the states of affairs of some world you've designed the model to represent.

I think outside of that debate, it was agreed that text worlds differ fundamentally from graphic worlds with respect to mode of presentation but do not differ in kind as long as both incorporate a world simulation/model. That is, they both lay equal claim to being virtual worlds.

At least that's what I took home from the discussion. =)
Aaron

133.

Raph>This whole discussion feels silly at this point. We're going to end up at a philosophical impasse.

I would agree there are a few arguments that are potential red herrings floating around in this discussion that tend to digress the conversation rather than progress it.

I would like to point out, though, that I agree with both schools of thought because I view virtual worlds as a continuum. Both schools of thought are discussing virtual worlds, but at different points on the continuum. When Koster and Bartle speak along the minima, for example, the aspects of virtual worlds at the state of creation, others seem to argue further along the continuum that it is much more. Koster and Bartle do not seem to be saying that it is not much more, but defend that it can be much less, and I agree. The others seem to be repeatedly arguing that it is much more, but do not seem to be saying that it cannot be much less. Such an argument cannot be considered opposition unless those arguing it is much more say that it cannot be much less. Thus the confusion on what, actually, we are arguing. There is such a minimum set of characteristics which allows one to identify a construct as a virtual world. Additions to that minima are also virtual worlds. In the current argument, both schools are correct. It is much more and much less.

I own a cafe. When it is open and bustling with customers, it is a cafe. Before the grand opening to the public, it was a cafe. When it closes at night, it is a cafe. When I only have one customer, it is a cafe.

The description vs. model argument keeps taking us back to the beginning. I can describe a model with text or graphics. When someone is interacting with a model of my apartment via a text interface (see experiments above), is it somehow no longer a model of my apartment because it is text? I make maps of both text and graphic worlds. It helps me understand the relationship of the parts to the whole. It is not a simple conversion of text to graphics. Such relationships can be expressed in text or graphics. Some relationships are more easily expressed in text, and others in graphics.

134.

Eric wrote: The others seem to be repeatedly arguing that it is much more, but do not seem to be saying that it cannot be much less. Such an argument cannot be considered opposition unless those arguing it is much more say that it cannot be much less.

I'm not sure that's true. I, for example, certainly have supplied such arguments. I'm even pretty sure that several others in this thread have as well.

135.

Eric>Such an argument cannot be considered opposition unless those arguing it is much more say that it cannot be much less.

Monkeysan>I'm not sure that's true. I, for example, certainly have supplied such arguments. I'm even pretty sure that several others in this thread have as well.

I've seen many arguments for it being much more, such as yours. I've seen many arguments saying that it is much less, such as Koster and Bartle. I did not note any arguments that say it cannot be much less. I did not note any arguments that say it cannot be much more. I've seen arguments that say it cannot -only- be much less, such as yours. I've seen arguments that say it cannot -only- be much more, such as Bartle. When someone talks about one side of the continuum, the other side seems to say - but what about this side.

From your reply, and the context of my statement, are you saying that you indeed have argued it cannot be much less?

Monkeysan>do not differ in kind as long as both incorporate a world simulation/model.

I see this as you now noting that it -can- be much less. At some minimum, a world simulation/model is represented. From my perspective, this is exactly what Bartle and Koster were saying.

136.

Eric: Well, I would argue that it cannot be "much less" if by "much less" you remove the potential for interactivity from the picture. Television and cinema offer rich views of a world simulation/model. But I don't think that when we talk about a "virtual world" that we are talking about a three-sided set. Likewise, computer simulations/models are used a lot in scientific computing and visualization, but I don't consider my Monte Carlo statistics to be a "virtual world" in any meaningful sense of the word.

Your cafe is a cafe, because I can interact with it as a cafe. The rules for interaction may change depending on state, and the notion of time-dependent state is central to how we interact with many places/objects. If it is under construction, I will look at the "opening soon" sign and peek in the windows to see progress. If it is closed for the night, I'll look for an sign displaying the hours and a menu.

In the city where I live in, many businesses are located in what used to be residential houses. The existance of a sign that communicates "cafe" makes these "opening soon" and "after hours" behaviors more acceptable, than they would be on the porch of a residential house.

So at the very least, I do expect for that model to have some sort of interactivity. Otherwise, we might as well just be talking about television.

137.

Eric: I did not note any arguments that say it cannot be much less.

I've made a few. Here's a quick one: "The code" (with or without "some" data) cannot be a virtual world on it's own because the code itself merely dictates state changes of the machine that executes it. In order to have a world model, there has to be a mapping of the state changes of the machine onto the behaviorsl/properties of the object it represents. That mapping is external to the code itself. In this sense, it "cannot be much less."


Here's an even quicker one: A virtual world cannot consist solely as a world model or simulation be cause world models are objects and virtual worlds are places. As such, a virtual world simply cannot -be- a world model.

Anyway, I won't press the issue anymore.

Aaron

138.

To me the discussion about being much more or much less is about defining the common denominators/factors of a virtual world. Kirk argues that the community element should be one of the common denominators. Others disagree.

I am of the opinion that a virtual Pluto can and is a virtual "world", (text or otherwise), but without life and living activity the world is "dead". I am also of the opinion that a virtual world resides in the mind. The rest are objects and symbols are reference materials.

Greg Costikyan in his website has a very good analysis of the transitioning of 2D to 3D symbology/view of Heroes of Might and Magic series. His analysis is more about implementation, but does add to the discussion.

Hey, blind people who can't see text or 3d models still can live in the virtual world. It matters not for them.


139.

Kirk> I would argue that it cannot be "much less" if by "much less" you remove the potential for interactivity from the picture.

I do not remove the potential of interactivity from the picture. That is the point. Interactivity need not be happening at that moment for it to be considered, at that moment, a cafe. It is a cafe. People come and go. Perhaps people never show up. I still constructed a cafe. One which never experienced its potential.

Kirk>If it is under construction, I will look at the "opening soon" sign and peek in the windows to see progress. If it is closed for the night, I'll look for an sign displaying the hours and a menu.

But, I imagine, you would still consider it a cafe. Peeking in the window, you can clearly see it is constructed to be one. When you look at the hours, you may even estimate a future use of it.

Then again, I could screw with people's minds and place a big sign on the window saying "Ceci n'est pas une cafe"... I apologize beforehand if this initiates another argument. I couldn't help myself.

140.

I've followed the arguments and did read and attempt to understand everything you are saying. You supplied many arguments around the basis of:

Aaron>I don't think that 'virtual world' is equivalent to 'world simulation'

I took it to mean, "I don't think that it is -only- that", which is to say "It's that, and more." I would agree. It can be so much more. I would also agree that it was perhaps a sensitive and difficult semantic difference, on my part, to argue, so I'll concede that to maintain focus. Norrath is still a place, though, even when the servers are down. A cafe is still a cafe, even when it's closed.

Aaron> In order to have a world model, there has to be a mapping of the state changes of the machine onto the behaviorsl/properties of the object it represents.

I don't know what this means, but I'd like to, so I'll offer a construct that we can use to illustrate our points to each other.

There exists a program, which, when executed, allows 0 to 25 users to simultaneously interact with a virtual model of my apartment.

Q1: Which would not be considered a virtual world?

1.The program is not running.
2.The program is running, but no one is connected.
3.The program is running, another program acting as a user is rearranging a sock drawer.
4.The program is running, a human user is in the kitchen taking an inventory of the soup.
5.The program is running, two human users are discussing Foucault in the living room.
6.The program is running, two humans observe a program dust the picture frames in the living room.

Q2: Consider a chronology of events:

T0.The program is running, and multiple users are connected.
T1.All the users disconnect
T2.The program terminates.
T3.The program is not running.
T4.The program is executed.
T5.Multiple users connect.

Does a virtual world cease to exist?

Q3. Which one is not like the other?

1. The program is running, but no one is able to connect. The network interface card on the server is down.
2. The program is not running.
3. The program is running, but no one is connected.
4. A human user is unable to connect. The connection attempt times out. The problem is unclear.

CONCLUSION
Q1. None
Q2. No
Q3. They are all the same.

Execution is a matter of access, not existence. When the programmer created the program, they created a virtual world.

But a virtual world cannot change when it is not running, one might argue. If I disconnect from a virtual world, then re-connect later and found no difference in the world between the two connections, is it still a virtual world? Would it have mattered if the program was not running during the time I was disconnected?

There are clear measures in which one can say that one virtual world is more world-like than another. This is the continuum. Simultaneity of access, persistence, sociability, history, economics, culture. These make the virtual worlds more world-like, but their absence does not deny the existence of a virtual world.

141.

Just to be picky, virtual worlds do in fact change when they are not running. :) It requires intervention to accomplish it, but it happens all the time--addition of world data by content creators, for example.

142.

Eric>Execution is a matter of access, not existence. When the programmer created the program, they created a virtual world.


Execution is not access. It is existence. Execution describes the process whereby a computer follows a ruleset that directs it to make certain changes in it's physical state. At least that's what I take it to be. And when that ruleset is designed to model certain features of some imagined world over time, we have a world model. The code, however, is not the world model, nor is it the virtual world.

Don't believe me, here's the proof. What is the code? It's a description. It's made out of propositions, for chrissake! How can the code, on its own, serve in any way as a model?

Execution is when the code is employed in a manner that determines the behavior of a computer so that it generates a representation of some imagined world. If that representation works by directly instantiating properties and relations (spatial features, phsical laws, etc.) of the imagined world, we can say that the representation is a world model. If it doesn't do that, then it isn't really a model at all.

The world model, by virtue of being a model, by definition, cannot consist of a set of descriptions. (This is feature of models--that they cannot consist of descriptions is consistent with nearly all conceptions of model I'm aware of.) Since the code is a set of descriptions/linguistic items, it cannot be a model. Therefore I don't see why it should be called the world model.

Even more difficult for the 'codist' position is that fact that world models are dynamic, and descriptions and linguistic items are certainly not dynamic. Since the code is not dynamic, then, how can it be the world model.

The answer is that it can't. It's the instantiation/representation of the behavior of an imagined world over time that does the job. And that's why it deserves the name 'world model', or 'world simulation' if you prefer.

And if the code isn't a model, then it certainly can't qualify as a virtual world.

Yes, there is likely a continuum, but the minima are nowhere near as minimal as 'the code'. Nor are they even as minimal as 'the world model/simulation'.

I guess I just fundamentally disagree, and I've said why already, and I don't think it's worth pushing much more.

But, as the dear governor of California is prone to say, "I'll be back."

Aaron

143.

magickback: Kirk argues that the community element should be one of the common denominators.

Not really. I'm arguing that interactivity should be a common denominator. Simulation/modeling alone is not sufficient because there are much richer simulations and models of real-world phenomena that are not interactive to any great degree.

Eric Random: Norrath is still a place, though, even when the servers are down. A cafe is still a cafe, even when it's closed.

How is there no "interactivity" going on. Wardialing a server in order to get a connection, seems like it would be an interaction. And in fact, many online spaces have other spaces just to talk about whether the server is up or down at any point in time.

Looking through the windows, or reading a sign is also an interaction. And where you have interaction, you have interactivity.

Of course, I agree with you that a cafe is a cafe when closed, and Norrath is a place when the servers are down. But there is a social element involved in those designations. If you don't pay your taxes or mortgage, that cafe could become a public auction. If someone gets prickly over illegal behavior using the servers, Norrath could become evidence in a criminal investigation.

And for that matter, there are spaces that are entirely dependent on a group of people engaged in a ritual behavior. A church basement can be a day care, a potluck dinner, a classroom, or a Boy Scout meeting at any point in the week.

Execution is a matter of access, not existence. When the programmer created the program, they created a virtual world.

That depends a lot on the program. There is a heck of a lot of software out there that does simulation/modeling. However, I think it's reasonable to provide a nice way to distinguish non-interactive simulations of climate change over time, from interactive simulations of life in the Star Wars universe.

144.

For your discussion, here is a very simple virtual world according to the simulation/model definition:

#!/usr/bin/python
import random
random.randint(1,6)

Enjoy.

145.

Me>if someone you don't know exists observes a virtual world, does that make it a virtual world?
Monkeysan>As long as they actually do exist, sure.

This would imply that believers in a single, all-seeing god would regard what you call a "world model" as identical to what you call a "virtual world", because there are no such worlds that aren't being observed by their deity?

>That deduction is only available if you assume that the world model is the virtual world. But that's precisely what's at issue.

But your definition of what makes a world model into a virtual world is dependent on someone observing it. I want to know why you feel they have to observe it - why can't they merely know it's there? Observation is itself only mere deduction based on information supplied to and by the senses; why can't we add some intellectual deduction to the mix?

>It can't possibly do that, since the code and the initial data can't be used to predict the actions of players in the virtual world, many of which actions affect the state of the world model.

No, the code can be used to predict it - it defines the boundaries. I know no-one is going to chop down the tree because there's no code for chopping down trees. The code plus initial data defines all possible futures that the virtual world ("world model" to you) can have. There are an infinite number of them, and only a finite number will ever exist. However, it does define them (in the same sense that "numbers which, when you wholly divide them by 2, leave no remainder" defines all possible even numbers).

>How can a 'virtual world' be an instantaneous state?

It can be in an instantaneous state.

>A novel is a set of descriptions of states of affairs. The reader uses those descriptions to build an internal world model where those states of affairs obtain, and the reader evolves that model as she reads the novel. But the novel itself is a set of descriptions, not a model.

And as far as the player knows, the same applies to a virtual world. The virtual world generates descriptions which the player builds to construct a mental model which evolves as the player encounters more descriptions.

Me>But no single observer can possibly observe more than a fraction of it at once. Does the rest not exist?

>I'd say yes, even in the virtual case.

OK, so what if the observation isn't near-instantaneous? What if there's a 10-second delay between how the world is and how you see it? Maybe some NPC casts a spell on you or something. If you're the only player, does the virtual world exist? Which defines it better - the model on the server or the model inside your head?

What if the world model plays a sensory trick on you? Perhaps it displays information to the client that isn't true and never was true. There may be invisible characters there that you can never see nor ever interact with, but which you can deduce exist because you see the effects of their past actions (turn your back for a moment and they steal the treasure). There may be illusions, so that you see a character who isn't there, it's just something sent to your client. What is the "virtual world" here? What you have in your head or what is running on the server?

>Well, the primary difference is that one is an object and the other is a (synthetic) place.

And why can't places be objects?

Richard

146.

Flatfingers>But isn't a virtual world that is explicitly designed as a multiplayer system a different kind of thing than a book?

Well yes, it is, in many ways. However, being a product that enables people to build models in their heads is not one of them.

>In fact, isn't a virtual world that is functioning as designed required to be "about" people interacting with other people, rather than being about individuals interacting with a system such as a single-user game or a book?

It's "about" people interacting with a system if you're an explorer; it's "about" people interacting with other people if you're a socialiser.

As a designer, ultimately it's "about" the interaction between designer and players at an artistic level. However, to interact with players, you have to create an artefact - in our case, a virtual world. You create that world in such a way that it addresses player needs, but also (hopefully) your own as a designer.

Now, if your artefact has no players, you're not going to get much of a dialogue (in the artistic sense). You still have an artefact, though. If the artefact does have players, you do get the dialogue. Are the players themselves part of the artefactm though? To my mind, no, they're not: they weren't created by the designer; the designer created the world for them.

This is why I say that a virtual world with no players is still a virtual world, in the same sense that a novel with no readers is still a novel. It's an artefact that has been created for a purpose, and (as you say) that purpose by necessity involves having players. The players aren't part of the artefact, though - they're not part of the art. Monkeysan argues that the art is the virtual world; I argue that the artefact is the virtual world.

>Why then is it unreasonable to conclude that what's in the heads of users of multiplayer code isn't also part of the system?

Why is it unreasonable to conclude that what's in people's heads shouldn't be excluded from the definition of the system?

>If the whole point of the code is to serve as a mediating entity, how can that code be a complete world without users (and what's in their heads)?

This would imply that a virtual world is the sum of the code plus the data plus the contents of the head of every player past and present (which, because people have contradictory views, would be inconsistent).

The whole point is not to serve as a mediating entity - that's just part of the point. The point is for the designer to express aspects of their personality through the medium of the virtual world. In order to do that, they have to create an artefact which mediates between individuals. This artefact is what I call a virtual world. If there are no individuals actually mediating through it right now, to me it's still a virtual world - just a temporarily empty one. To the players it may be something else, but to me as a designer it's a virtual world.

[Off topic - feel free to skip this!]
>I've tried to explain to family and friends, but they've never quite seemed to understand how important it is for me to be able to concentrate, and how difficult that is for me when there are words being spoken nearby.

That's just what happens to me. I have to stop writing if someone puts a TV on in my hearing, which means that the time when most people would be writing their best-selling novel part time, I'm unable to do so. I have to play computer games instead..!

>Back in college I started my collection of "music without words"

Me too, but I focused on classical (excluding opera). I wish there were more instrumentalist music about, but there doesn't seem to be much of a market for it - people prefer their stars to be singing stars.

>I suspect it's not coincidence that these artists spend some or all their time singing in a language that's not English.

These still distract me. If I can recognise it as words, that derails me. I can't even listen to Karl Jenkins' music without picking up on the nonsense words (to the extent that I can singalong to them).

>And this is the first time I've ever run into anyone else who can do the "white noise in the ears" trick at will (or even understands what I mean when I try to describe it).

Yes! I knew I couldn't be alone in that!

>I tend to use it solely to temporarily block painfully loud sounds (jet aircraft engines, fire alarms, etc.).

Yes, this is really handy. I don't know why it should be that it stops the noise from hurting, but it works for me.

>The only way I've ever figured out how to describe it to other people is to say that it's what you hear at the end of a really big yawn.

I hadn't noticed that, but now you mention it, yes, you're right. That'll help me explain it in future - thanks!

If you get distracted by language when you're doing linguistic tasks, does that mean (fingers crossed) you can also do the "set something off" thing, too?

Richard

147.

Monkeysaaron>What is the code? It's a description. It's made out of propositions, for chrissake! How can the code, on its own, serve in any way as a model?

It can't, but along with its data it can. I think maybe you're being misled by sloppy programmer langauge here - sorry!

When programmers talk about "code" they tend to take "and data" as a given. So you're right, code on its own can't serve as much of a model (although there are such things as procedural or functional models - you could model a plumbing system that way, for example). Code plus the instantiated data structures it needs is a model. Interestingly, some of those data structures could themselves contain elements that could be interpreted as code, which could access other data elements, and so on indefinitely.

>If that representation works by directly instantiating properties and relations (spatial features, phsical laws, etc.) of the imagined world, we can say that the representation is a world model.

Yes, I agree, although at least the way I teach it in class, the world model is more like a template from which specific instances (virtual worlds) can be struck. An interesting consequence of this (from the point of view of our discussion) is that although the same code runs on the servers for all shards, each shard has its own unique instantiations of the data, therefore each shard is effectively its own, unique virtual world.

Richard

148.

Richard Bartle: This would imply that a virtual world is the sum of the code plus the data plus the contents of the head of every player past and present (which, because people have contradictory views, would be inconsistent).

Well, as a designer, I think that when you design for human interaction, that the contents of the head of the audience become just as important as a software library. I've yet to see a design built for humans that did not, to some degree, depend on the contents of the user's head. As the old saying goes, the only really intuitive interface is a nipple.

And in many cases, the design explicitly plays on the fact that different people will develop radically different interpretations.

149.

Kirk>Of course, I agree with you that a cafe is a cafe when closed, and Norrath is a place when the servers are down. But there is a social element involved in those designations.

It seems we agree, as both the cafe and Norrath, although not accessible, are still a cafe and virtual world because of their potential.

Eric> Execution is a matter of access, not existence. When the programmer created the program, they created a virtual world.

Kirk> That depends a lot on the program.

And thus, in my questions, I said, there exists a program. I then stated specifically what that program is. That is particularly my point. The virtual world depends on that particular type of program. Without that program created that particular way, it is not a virtual world.

Raph>virtual worlds do in fact change when they are not running. :)

There are many such ways for it to change without running. I completely agree. My point being that a program running and not changing need not be different from a program not running.

I think some of the other arguments are attempting to apply metaphysical concepts of the real world onto concepts of virtual worlds. Some concepts fit, some don't. Many concepts that work well in virtual worlds don't apply well to the real world. It's like God dangling a plushy toy in front of a human's face and saying "Now you see it!", then willing it out of the material plane, and saying "Now you don't! Where'd it go? Where'd your plushy toy go?" Then willing it back "Oh, there it is!". Games like that tend to cause blank stares in the real world.

150.

Eric: The virtual world depends on that particular type of program. Without that program created that particular way, it is not a virtual world.

What is meant by "that particular type of program," and "that particular way?" That is I think the key question. My feeling is that we need to include such terms as participation, interaction and semiotics in addition to simulation. Otherwise, there is no reason to exclude trivial simulations such as random.randint(1,6).

Of course, once you include participation, interaction and semiotics, you are no longer talking about something that just exists somewhere on a hard disk. When you throw in a pointy-eared humanoid into a game and call it an "elf", you are making a link to a cultural library that is just as important to the function of your software as the function calls that allow you to load data from disk.

151.
But isn't a virtual world that is explicitly designed as a multiplayer system a different kind of thing than a book?

In fact, isn't a virtual world that is functioning as designed required to be "about" people interacting with other people, rather than being about individuals interacting with a system such as a single-user game or a book? (Note that while a book can be experienced by multiple people simultaneously, that's not the same thing as people interacting with each other.)

I'd disagree myself. To take one example, Star Wars was very much a virtual world to me long before Raph came along and helped build one particular computer model of it. For some it may have been a virtual world since "Tie Fighter" for others it might have been a virtual world since they first played with the action figures.

There are many world's I would love to see instances of in an interactive computer medium. They are no less worlds to me. I think the more worldy a virtual world is the more mediums it will move through.

152.

Kirk>What is meant by "that particular type of program," and "that particular way?" That is I think the key question.

Yes! The key question is about the program! We're getting somewhere...

Kirk>#!/usr/bin/python
Kirk>import random
Kirk>random.randint(1,6)

I do not consider that code a virtual world.

This is an excerpt of a definition I wrote a while back on MUD-Dev...

**
VIRTUAL: being such in essence or effect
WORLD: the sphere or scene of one's life and action

In it's most fundamental manifestion, it is a SPATIAL REPRESENTATION
which can be INHABITED by a CORPOREAL PROXY.
**

The program I described conforms to this definition.

Eric>There exists a program, which, when executed, allows 0 to 25 users to simultaneously interact with a virtual model of my apartment.

When someone says "I was playing in a virtual world" or says "I build virtual worlds," I know fundamentally, this is what they mean. From here, it can have many facets of functionality, realism, sociability, accessibility, persistence, contiguity, containment, vastness, dynamism, dramatics, and immersion that make it become more world-like.

153.

Eric Random: When someone says "I was playing in a virtual world" or says "I build virtual worlds," I know fundamentally, this is what they mean. From here, it can have many facets of functionality, realism, sociability, accessibility, persistence, contiguity, containment, vastness, dynamism, dramatics, and immersion that make it become more world-like.

Woah, nelly. For many projects, it works the other way around. The spacial representation with a corporeal proxy is just something bolted onto a network that started with a set of idealized norms of behavior.

154.

Kirk>The spacial representation with a corporeal proxy is just something bolted onto a network that started with a set of idealized norms of behavior.

To reduce confusion, I work in specific examples. If you can supply me with a specific case which illustrates your point, I could better understand. I don't know what you mean when you say that.

This definition works in a meaningful way to intersects multiple disciplines. The game designer, programmer, sociologist, simulation engineer, educator, player, writer, etc. all converge upon this fundamental aspect, and work out from there. One can say, it started as a game, a simulation, an educational tool, a chat channel, and became a virtual world. What changed? Where did all this converge? What is the common thread?

Consider this example:

I've seen it argued many times that instant messaging is not a virtual world, and I would agree. Put a bunch of avatars (corporeal proxies) in a series of visualized rooms (spatial representation) and have them talk to each other. All of a sudden, people start to argue this is a virtual world. Functionally, it is exactly the same. What changed?

Have a program which has two warriors battle each other to see who wins. Some would argue this is a simulation, not a virtual world. Give those two warriors avatars (corporeal proxies) and let them fight each other in a castle (spatial representation), and people start arguing it's a virtual world.

Consider this argument:

Do the people who developed Morrowind III have nothing in common with those who developed World of Warcraft? Both applications are virtual worlds with different modes of access. They may not be the same, but they share many commonalities; commonalities that can merge and intersect.

With this definition, I can explicitly define a wide range of variations of virtual worlds, and there are many, and all share many of the same obstacles in implementation and operation. This is not simply drawing lines in the sand. This is creating a framework in which one can draw lines. I can now define a virtual world as persistent or intermittent, perpetual and temporary, access as simultaneous or sequential, point at the rules which can facilitate socialization and discuss the effects of their absence. Discuss dynamic states in the world, or the lack of them. I can discuss how a player visualizes a virtual world even when not connected, and discuss a virtual world before it's program is executed. Discuss how a virtual world can exist among multiple programs, and how multiple virtual worlds can exist as one program. I can discuss the effects when different types of virtual worlds converge, and can even talk about a virtual world before it is bolted onto a network.

So when people are arguing over whether something is red or orange, I can at least have a mechanism which describes the exact electromagnetic frequency of the color in question.

155.

Where is Barsoom?

Let's say there's no model of Barsoom, and no code+data instantiating a virtual world that incorporates most of the elements of the ERB novels.

Does Barsoom exist? If so, as what does it exist? Is any part of its existence based on having some representation inside anyone's head?

Next question: Where is Norrath?

Does Norrath exist "more" than Barsoom? Why?

Would Norrath still exist if you pulled the plug on all its servers? Would it still be a virtual world (assuming you thought it was one to begin with)?

(Incidentally: Why in the world hasn't someone created a major MMORPG based on the John Carter of Mars novels yet? Six-armed green Tharks with bad attitudes, vicious banths, nubile Red Martians to save, flight and weapons technology already described, along with a classic backstory -- I can't understand how this hasn't been massively implemented yet.)

Flatfingers> Why then is it unreasonable to conclude that what's in the heads of users of multiplayer code isn't also part of the system?
Richard> Why is it unreasonable to conclude that what's in people's heads shouldn't be excluded from the definition of the system?

Not to be flip, but I would answer: Because expansive views of systems are more interesting than limiting views of systems. (More pieces = more surprising behavior.)

If neither viewpoint is more correct, why not prefer the one that's more interesting?

Flatfingers> If the whole point of the code is to serve as a mediating entity, how can that code be a complete world without users (and what's in their heads)?
Richard> This would imply that a virtual world is the sum of the code plus the data plus the contents of the head of every player past and present (which, because people have contradictory views, would be inconsistent).

I'm comfortable with that implication. But then I'm comfortable with the Gödelian notion that any formal system complete enough to be able to express propositions about itself is necessarily inconsistent.

If you add the "Norrath inside our heads" to the Norrath that exists as code and data on the server to get a system that's complete, I'm not sure what's invalidated by the fact that this system is also inconsistent. But maybe I've gone down the Gödelian bunny trail one too many times....

All this being said, the notion that virtual worlds aren't systems that only exist in full when they actively serve others, but are artifacts created to serve the creator... hmm. That, for some reason, I'm not so comfortable with. It seems awfully egocentric.

[digression time again, ignore at will]

> If you get distracted by language when you're doing linguistic tasks, does that mean (fingers crossed) you can also do the "set something off" thing, too?

Yes.

Again, I first noticed this back in college. Sometimes answers to questions would come if I pushed on them, but sometimes they wouldn't. In the latter case, I found myself moving those questions to what I've since heard other people call the "back burner" to simmer subconsciously... and sure enough, in most cases there's eventually a *ding*, and out pops an answer. Maybe not the answer, but an answer, and one that seems not to have been available by pushing consciously.

After discovering computers, I (like you) adopted a computational metaphor for these two approaches: I think of them as "foreground" and "background" processes. And I can, have, and do explicitly assign questions and puzzlements to background processing while I work on something else in the foreground, because it just plain produces results.

So yes, I think we have another similarity here. But two points (not necessarily of difference):

1. Mine works while I sleep. In certain cases where I have nearly reached obsession with some difficult design issue, I've "slept on it" and awakened in the morning with a solution that actually worked. It's possible this would have happened without explicitly moving the problem to background processing... but these solutions don't seem to come to me when I don't relegate them to a background process.

2. I wonder, reading your description of multiple processes, whether I may have been shortchanging myself lo these 20-odd years. Why do I just use two processes, foreground and background? Could I have accomplished more by forking off N processes? Two have always felt "right" to me, but maybe that's just a function of not having enough CPU cycles to handle the load of > 2 processes. *grin*

Next time: How to make your pupils dilate at will!

--Flatfingers

156.

Kirk Job-Sluder>I think that when you design for human interaction, that the contents of the head of the audience become just as important as a software library.

Of course they are, but you're not designing the contents of their head. The contents of the heads of readers of novels is of critical importance to novelists, but they don't write the contents of people's heads, they just write novels. So it is with virtual worlds: you design a virtual world (or "world model" in Monkeysan's terms) with an aim to affecting the contents of people's heads, but you don't actually design those contents any more than a novelist writes them.

When studying a novel, it makes sense to take the artefact and examine it to see how it influences the reader, and to consider the author/novelist relationship as one of artistic dialogue through the device of the novel, but it's the artefact alone that is the novel. With virtual worlds, you can examine the communal relationships between players as embodied in and enabled by the software, but the players aren't part of the virtual world any more than readers are part of the novel.

If designers could design the contents of people's heads, I'd go along with you. They can't, however (and I should imagine few would want to - there's no fun in it).

Richard

157.

Me> Why is it unreasonable to conclude that what's in people's heads shouldn't be excluded from the definition of the system?
Flatfingers>Because expansive views of systems are more interesting than limiting views of systems. (More pieces = more surprising behavior.)

In that case, why not expand the system further to incorporate the views of people whose close relatives play virtual worlds, or who have political views about computer games in general, or who think they could maybe use games to teach children to be nice when they grow up? You'll get even more surprising behaviour that way!

I stop at the level of the artefact because that's all I can design. I can design it with a view to engaging people, to communicating with them, to help them have fun, whatever; I can't design the people, though. That's why I stop at the design level. If I design the players, I can't talk to them through the design - they become part of the design. As a designer, I want to say things, I don't want to pull strings.

It's interesting to consider the wider system, yes, but why should that wider system take the name "virtual world"? I see virtual worlds as places you visit, not communities you join; if you want to talk about the communities, that's good - I want to read about them - but think up your own term for it instead of extending this one..!

>I'm comfortable with the Gödelian notion that any formal system complete enough to be able to express propositions about itself is necessarily inconsistent.

But virtual worlds can't express propositions about themselves, only about parts of themselves.

>That, for some reason, I'm not so comfortable with. It seems awfully egocentric.

Well, nobody said virtual world designers had to be modest!

[Off-topic discussion coming: stop reading if you're only here for Terra Nova standard stuff]

>In the latter case, I found myself moving those questions to what I've since heard other people call the "back burner" to simmer subconsciously...

So you explicitly move them there? What does the "back burner" feel like - a place? Can you list (mentally) what you have there? Can you take things off the back burner if you don't need them any more? Can you do linguistic tasks there?

My answers to the above for how I do it are: yes; no, it's not a place, it's a memory of a mental feeling that I can pull up, but I can't "see" anything "nearby" there; yes, if I remember them, but if I forget they'll still carry on doing their stuff; yes; no.

>and sure enough, in most cases there's eventually a *ding*, and out pops an answer.

That's exactly how it feels to me - a mental *ding*!

>I think of them as "foreground" and "background" processes.

Do you get more than one foreground process? I only have the one (which is "me").

>1. Mine works while I sleep.

Mine too, but I can't set anything off when I'm asleep. Hmm, actually, I've never tried it; I wouldn't want to, though, because I have enough troubles when I start dreaming about problems - I wouldn't want to wake up and find my head full of processes all trying to figure out the answer to some question that was only meaningful in my dream world!

>these solutions don't seem to come to me when I don't relegate them to a background process.

If I make a decision whether or not to set something off and I don't set it off, then it won't run. If I don't make the decision, it's possible something could be going on anyway at a level beneath my conscious awareness, and I get an interrupt some time later. Example: I was once walking through Copenhagen and passed a jazz club. A few seconds later, I found myself playing (internally) the piece "Take 5" by Dave Brubeck. OK, jazz club, triggered memory... Except this was "Take 5" at a much faster speed than normal. I recalled that Brubeck did do that once at a famous concert - in Copenhagen. Now that's more than a triggered memory, that's something saying "hey, you just walked past a club where Dave Brubeck may have performed, go back and get the name so you can check" (I did and it wasn't). So here I was being given the answer to a question I hadn't even asked. I believe that most people do this all the time. The only difference with this and what I do is that if I think about something, I get the option of whether to set up something to find a solution or not. It sounds as if that's how it works for you, too.

> Why do I just use two processes, foreground and background? Could I have accomplished more by forking off N processes? Two have always felt "right" to me, but maybe that's just a function of not having enough CPU cycles to handle the load of > 2 processes. *grin*

Try it! See if you can have two background processes ongoing at once. It might be that you only have to change the way you visualise them or something - at least you know it's possible now.

I can't tell you what a relief it is to find someone else who can do this!

>Next time: How to make your pupils dilate at will!

I can't even flare my nostrils at will...

Richard

158.

Re Flatfingers and Richard's comments about "foreground" and "background" processes . . .

Continuing slightly offtopic . . . I generally agree with the description that it helps to think about "backgrounding" a project or task and that there is a definite "ding" upon completion (and that completion often occurs after downtime). FWIW, many of the best programmers I know talk about managing tasks this way. Also, this model would be exactly predicted by On Intelligence. I find that as I background more than 4 or 5 things that it becomes harder to remember to follow up on the "ding", ie I figure out the solution to something as I wake up in the morning but then forget about it for a day or two unless I write it down or email it to myslef.

159.

Cory>Re Flatfingers and Richard's comments about "foreground" and "background" processes . . .

I think the key aspect of foreground and background is, simply, exposure. This is not simply repetitive exposure, but varied and persistent exposure.

I've dealt with many people who, initially, are convinced they are unable to understand particular subjects like mathematics, computer science, philosophy, cognitive psychology, molecular biology, economics, etc. when they ultimately find the only reason they were unable to understand fundamental concepts in the fields was a matter of exposure. Such exposure requires diligence, and diligence requires either discipline or excitement.

Flatfinger's backburner is a method of exposure. Cognitively varying an idea, pausing, then returning. Simple repetition solidifies cognitive connections, varying it broadens the connections, and building on variation solidifies the connections it is built upon. This can have an effect of turning conscious connections into subliminal ones. This can ultimately result in a ding when the right combination of connections are made.

Cory's loss of "ding" occurs because, often, once we get that "ding" we move on. We discontinue exposure, and thus fail to solidy that combination of connections to the degree we can easily recall it. We didn't build on the "ding". This especially occurs when we are juggling multiple cognitively disparate ideas. If I continued with examining how can one communicate the "ding", what are the implications of the "ding", what are the applications of the "ding", I find I remember the "ding", but I forgot some of the applications.

This creates the effect where some people say "I don't need to write it down" on the basis that on a previous "ding" they wrote it down, but did not need to refer back to it. They then find that when they don't write it down, they wish they would have. That is because the process of determining how to communicate the "ding" helps to congitively vary and solidify it.

I would agree with aspects of the study concerning cognitive dissonance, cognitive focus and the effect of focal shift, but there must be a wide range of stimulus to understand the differences. Listening to a "Star Wars" radio play and a speech by Alan Greenspan about the U.S. economy require different cognitive requirements. Taking an order to pick up Chinese food on a cell phone is cognitively different than being confronted with the injury of your child on the cell phone. Blindly thumbing the radio knobs requires a degree of focal attention that could be considered as affecting driving focus. I've seen focus enhanced and the effects of focal shifts mitigated when aural rythm is applied to visual acuity tasks. What happens to cognitive dissonance when the hold music on your cellphone is beating to the rythm of pylons rushing past your window?

160.

Richard: When studying a novel, it makes sense to take the artefact and examine it to see how it influences the reader, and to consider the author/novelist relationship as one of artistic dialogue through the device of the novel, but it's the artefact alone that is the novel. With virtual worlds, you can examine the communal relationships between players as embodied in and enabled by the software, but the players aren't part of the virtual world any more than readers are part of the novel.

Well, at this point, we are stretching the limits of the novel as metaphor, and I'm really baffled by the use of the novel as a metaphor.

A better metaphor in my mind would be classic Dungeons and Dragons, which was a really interesting game because it was of tools for supporting emergent participant behavior. While there was a minimal set of formal published structures, most of the magic happened in the informal structures created by the participants, and in fact, most of the innovation during the early years of Dungeons and Dragons came out of participant game play: bards, barbarians, alternate settings, specialist mages, plotlines.

There are quite a few parts of Norrath which consist of emergent participant structures that are enabled by the formal design, but were not formally designed.

I stop at the level of the artefact because that's all I can design.

I think that's a pretty limiting view of design. I just read "Devil in the White City" about the Chicago World's fair. It was built by the cream of the crop of archetects at the time, and they certainly had in mind the view that they were designing not just artefacts, but human behaviors.

Eric: To reduce confusion, I work in specific examples. If you can supply me with a specific case which illustrates your point, I could better understand. I don't know what you mean when you say that.

I work with the design of socio-technical systems that help people learn. That is, I start with ideas about what kinds of participant behavior I want to encourage, and then pick the systems that encourage those behaviors. They way you phrase it, it sounds as if everyone is going at this from the perspective of "coporeal proxy! cool, what do we do with it?"

I've seen it argued many times that instant messaging is not a virtual world, and I would agree. Put a bunch of avatars (corporeal proxies) in a series of visualized rooms (spatial representation) and have them talk to each other. All of a sudden, people start to argue this is a virtual world. Functionally, it is exactly the same. What changed?

Now, I would argue that my first "virtual world" was a collaborative fiction role-play that was mediated over a threaded bulletin board system. In that case, the "model" was a shared conception of the rules of the space. For that matter, Classic Dungeons and Dragons works as a virtual world.

161.

And at least from the perspective of a participant, I never saw the distinction between the things explicitly built into the code of the system, and the things that happen online because people connect to the system as relevant. The fact that one Wizard never interacted with users, and while another Wizard acted as the "host" greeting new players within a few days, was just as much a part of the "virtual world" as the dungeon built around Beatles songs.

162.

[back to the current main subject]

Flatfingers> ... any formal system complete enough to be able to express propositions about itself is necessarily inconsistent.
Richard> But virtual worlds can't express propositions about themselves, only about parts of themselves.

Well, they can if you hold the view that human users are required for the system to be complete. Once people can speak about themselves as part of a system, the system is guaranteed to be inconsistent.

> why not expand the system further to incorporate the views of people whose close relatives play virtual worlds, or who have political views about computer games in general, or who think they could maybe use games to teach children to be nice when they grow up?

Because direct association seems like a reasonable breakpoint for what's part of a system and what isn't. In this case, "direct association" means "actively experiences the virtual world."

If asked for a mechanistic definition of, "what is a car?" I would definitely include an engine in my answer. I might even include gasoline, since even though there's no mechanical attachment and I can take gas out of a car, it's still a direct association. But I certainly wouldn't include something like a telephone pole, even if I could see one from my car, because there's no direct relationship between my car and a telephone pole (which seems like a Good Thing to me).

A telephone pole contributes nothing to recognizing a car. But human players do contribute to recognizing a virtual world as distinct from, say, a wind-up toy. (Which reminds me of the story in Stanislaw Lem's The Cyberiad in which Trurl the constructor creates a world in a box -- at what point does that "toy" world become a "real" world?)

> As a designer, I want to say things, I don't want to pull strings.

I wonder if this might be the most pointed statement so far on this subject. If I may rephrase it: "What you think a virtual world is for defines what you think a virtual world is."

Those who think the point of creating a multiple-user world is to attract and serve multiple users (and I guess I fall into this category) will describe the system as incomplete without users; those who think the point of creating a world is simply the act of creation itself will describe the system as complete whether it has users or not.

(This reminds me of our discussion of art -- if no one ever sees it, is it still Art? Or is the intentionality of the artist all that's required for a created thing to be Art?)

Perhaps the difficulty the participants in this conversation have had agreeing with each other is more psychological than anything. To the social or utilitarian designers, a purely intellectual challenge seems sterile. (They ask, "What's it for?") But to the "pure" designers, an elegant design is a thing to be enjoyed for its own sake. (They ask, "Why does it have to be 'for' anything to have value?")

Is any synthesis of these positions possible, or even desirable?

[and now back to off-topic, which seems to have become on-topic... sort of.]

>> I [moved] questions to what I've since heard other people call the "back burner"...
> So you explicitly move them there? What does the "back burner" feel like - a place? Can you list (mentally) what you have there? Can you take things off the back burner if you don't need them any more? Can you do linguistic tasks there?

Yes, as in, "Auggh, I can't focus on this anymore (but I want to keep working on it)"; not exactly a "place," but more of sense of relative position: "here" (foreground) and "there" (background); yes, because there's only one thing at a time (that I know of) in background mode; yes, there seems to be a one- or two-week timer, after which I get disgusted with myself and consign the problem to the "I probably need to rethink my whole approach" wastebin; and no... but to answer that needs a new paragraph all its own.

Most of the things that merit background processing (for me, anyway) are design problems. That is, they aren't things like, "when should I mow the lawn?" or "what was the name of that actress?"; they're more "how can I ..." or "what's best way to ..." questions. If I were to use a visual metaphor, I'd say they were attempts to figure out how to manipulate the internal structure of a square peg of factual relationships so that the resulting structure will fit through the round hole of constraints.

"Design" to me is the process of selecting relevant concepts, then forming and reforming internal structural relationships among those concepts so that the resulting structure best satisfies a set of constraints (where "best" implies an economic assessment of utility payoff for complexity cost, per Saint-Exupéry). And I mention this only to say that it's this process that occupies most of my intellectual time these days, so those are the kinds of problems that get the lion's share of background processing time. Other problems just don't seem to benefit from that level of attention.

FWIW, I've also found that background processing can work for intellectual problems, but problems with a strong emotional component don't go anywhere in background mode. They just sort of spin in place like a hamster on a wheel. All I get from applying problem-solving techniques to those kinds of problems is exhausted.

> Do you get more than one foreground process? I only have the one (which is "me").

Nope -- for me, "conscious" == "focused," which means there's room for only one thing at a time. (This brings us back to the point about concentration being so important.)

I should mention that this "one at a time" limit may be true for me with background processing, too. I'm willing to try multiple background jobs (I've got one running right now, actually), but as Cory and Eric discuss, I wonder if this would cause me to miss hearing a *ding*. As long as there's just one background process running, I know if I get any kind of internal *ding* exactly what has just happened and what the result is (usually in a blinding flash of "Ah!").

> I can't set anything off when I'm asleep. Hmm, actually, I've never tried it; I wouldn't want to, though

Same here -- I have more than enough things to work on while I'm awake; no need to empower the Monster from the Id!

> I was being given the answer to a question I hadn't even asked. I believe that most people do this all the time.

I mentioned this conversation to my wife yesterday, and she indicated that she thought this subconscious processing effect was probably fairly common. The existence of "back burner" as a metaphor suggests that she could be right.

Deliberately making use of this effect in a conscious way, however -- I agree with you; that seems much less common.

Now we just have to wait to hear whether Raph does this, too. ;-)

--Flatfingers

P.S. Apologies to anyone annoyed by the length of this message. Of course, if you've actually read all the posts in this thread, I can't possibly do anything more painful to you....

163.

Eric>Listening to a "Star Wars" radio play and a speech by Alan Greenspan about the U.S. economy require different cognitive requirements. Taking an order to pick up Chinese food on a cell phone is cognitively different than being confronted with the injury of your child on the cell phone.

Do you have research to back that up? Because I am not aware of a any studies that have shown that the processing of a particular kind of input (aural, visual) has built-in 'content' discriminators. In fact , given current theories of cognition it's not even possible.

Sure, the examples you cite are 'cognitively different' in that result in different mental states, but how do the 'cognitive requirments' for understanding them change?

Aaron

164.

Richard: With virtual worlds, you can examine the communal relationships between players as embodied in and enabled by the software, but the players aren't part of the virtual world any more than readers are part of the novel.

Just a few counter-arguments:

Umberto Eco to my knowledge would explicitly argue against the claim that a novel is just an artefact printed on some media, and point to "open narratives" that depend on the reader develop their own interpretation of what's really going on.

Secondly, it seems odd to me that a lot of "meatspace" design is looking beyond just the ideal location of artefacts within a venue, and looking at the experience of being in the space. When my mom gushes over Lambert's as an essential part of any trip to that area, she does not mention the decoir, but their signature gimick of having servers throw fresh rolls to patrons that raise their hands. Likewise, there is a heck of a lot of design that went into McDonald's on how to minimize the amount of time people spend in the store.

The existence of some code on a server somewhere that provides a world model and methods for interaction my be the lowest possible denominator. But I think that once you are open for business, the patterns and habits of the "regulars" in the space you create become an integral part of that space.

165.

Richard>I stop at the level of the artefact because that's all I can design. I can design it with a view to engaging people, to communicating with them, to help them have fun, whatever; I can't design the people, though.

I see virtual worlds as places you visit, not communities you join; if you want to talk about the communities, that's good - I want to read about them - but think up your own term for it instead of extending this one..!

I was waiting when this would pop up. I expect this is the motivation behind much of the 'codist' stance. Basically, you feel proprietary about what virtual worlds are, and you want a definition that is unequivocal about where the locus of creation and status of a virutal world is. Specifically, you want a definition that lays all of these at your own feet. You then extend that sense of the proprietary to vocabulary itself.

The fact is, most of the codist arguments have stemmed around a combination of this notion of the proprietary coupled with bald assertion that 'this is just what a virtual world is.' This coupled with quarreling over alternate views, while failing to provide positive support for the codist position other than authority, can be frustrating. It seems to defy the whole notion of interchange.

The sense of propriety and ownership are natural feelings, of course, and if anyone is entitled to voice them, I suppose it's you. And given the implications of moving the definition of a virtual world outside that of 'the code plus some data' (see Greg's article), it's understandable that many other designers/programmers share your feelings. However, it doesn't make for strong argument, and the fact is that no matter how hard you try, you won't be able to maintain sovereignity of discourse on the subject indefinitely. Intellectual enterprises just don't work that way.

166.

Kirk Job-Sluder>Well, at this point, we are stretching the limits of the novel as metaphor, and I'm really baffled by the use of the novel as a metaphor.

The way I see it, a novel is an artefact created by an author to say something to a reader. A virtual world is an artefact created by a designer to say something to a player. That's the level at which they are similar - the level of art. I could have used painting or poetry or choreography as a metaphor if I'd wanted. Whatever the case, the creator creates "something"; I'm calling this an "artefact", but it could be more abstract than this - a one-off performance, say. This artefact says something to those who experience it. However, what drives the creator is the need to say, not the need to be heard; the artefact is the end, not the means to some other end. This is why I think it's fine to say that you can have a virtual world with no players. What the designer designed is the world; the players make of it what they will.

>There are quite a few parts of Norrath which consist of emergent participant structures that are enabled by the formal design, but were not formally designed.

It's generally very exciting for designers when they see this happening. They construct a system that allows for unforeseen emergence, but the system lays out the boundaries for that emergence and the rules by which the emergence takes place. When such emergence then takes place and everything works just as it "should", this makes designers feel good.

>I just read "Devil in the White City" about the Chicago World's fair. It was built by the cream of the crop of archetects at the time, and they certainly had in mind the view that they were designing not just artefacts, but human behaviors.

Designing to shape human behaviour is one thing; designing human behaviours is something else entirely. I want to give people freedom to be themselves; I don't want to turn them into puppets in some grandiose theatre. Designing human behaviours is not only distasteful but also immoral.

>If you can supply me with a specific case which illustrates your point, I could better understand. I don't know what you mean when you say that.

In MUD1, we had permanent death. If your character died in a fight, it was obliterated. There was no resurrection in any form. Higher-level characters were better able to defeat attackers, but had more to lose if they lost. People who picked on weaker players would win a lot of the time, but occasionally they would lose. When they lost, they lost much more than the person they were fighting would have lost. Thus, players realised that killing other players was a losing strategy, and they moved on to doing other things. They only attacked other players if they had good reason for it.

This is an example of designing to shape human behaviour. The message I was conveying was that conflict for the sake of conflict would get you nowhere. Most players took this on board; a small few decided that they got more pleasure in being an asshole than they might get from becoming a wiz (ie. winning the game), so they kept on killing anyway.

Now if I were designing human behaviour, what would this look like? I'd have to straitjacket players so they could only conform to the behaviours I had prescribed for them. They'd learn nothing about the world, other people or themselves; they'd feel like cogs in a machine, which is exactly what they would be. They would be part of the art, not experiencers of the art.

It's good that people want to study how experiencing the art builds relationships, community and culture, and I'm eager to read as much as I can about this. However, for me there is a definite line between what I design and what people do with that design when I hand it over to them. The former is my artefact; the latter can be regarded as a greater entity that my artefact enables. All I'm holding out for is the right to call my artefact a "virtual world", rather than its appropriation by its players.

>I start with ideas about what kinds of participant behavior I want to encourage, and then pick the systems that encourage those behaviors.

That's not designing behaviours, though, that's designing to encourage behaviours.

>For that matter, Classic Dungeons and Dragons works as a virtual world.

I cover this in my book. By the definition I use, D&D is almost a virtual world, but falls short because its physics isn't automated. Chatrooms aren't virtual worlds because they don't even have physics.

Richard

167.

Flatfingers>Well, they can if you hold the view that human users are required for the system to be complete.

But I'm saying that they don't have to hold any view on that subject.

>because direct association seems like a reasonable breakpoint for what's part of a system and what isn't.

In that case, I would say that the artefact itself is the direct part; the players are associated with it, but they aren't necessary components of it. Therefore, the artefact is a natural breakpoint. This isn't to say that the artefact+players isn't also a natural breakpoint, but from the point of view of defining "virtual world" I feel that the artefact itself is the better breakpoint.

>A telephone pole contributes nothing to recognizing a car. But human players do contribute to recognizing a virtual world

The potential for human players does; actual human players aren't necessary. If I say "I'm building a car" then you expect that it will have a driving seat and some passenger seats, because the car will have drivers and passengers. However, it's still a car when I've built it even if no-one ever drives it. I can build a virtual world that is still a virtual world even if no-one plays it.

>"What you think a virtual world is for defines what you think a virtual world is."

I'd remove the word "for" from that (grin).

For a designer, a virtual world isn't "for" anything. It's a necessary expression of some aspect of the designer's being. Designers design because that's what they do. Their chosen medium, virtual worlds, enables them to articulate their identity in ways that other forms of expression don't, and it comes with its own set of constraints. A virtual world with no physics wouldn't be a virtual world, therefore it would be a different medium for the designer.

Players certainly play virtual worlds "for" a reason (to "have fun"), and this is something that designers take into account when they create their worlds. It gives designers one of many modalities of expression. What virtual worlds are to players is different to what they are to designers, though, in the same way that what buildings are to architects is different to what they are to inhabitants.

>Those who think the point of creating a multiple-user world is to attract and serve multiple users (and I guess I fall into this category) will describe the system as incomplete without users

I wouldn't use the word "incomplete". I might use the word "unfulfilled", however.

>those who think the point of creating a world is simply the act of creation itself will describe the system as complete whether it has users or not.

That's correct. The world is complete when the act of creating it is complete. To say that it is only complete when players are in it is to say that the players are part of the creative process. They're part of a creative process, yes, but not the creation of the virtual world. The process they're part of is the creation of the virtual world's present.

>(This reminds me of our discussion of art -- if no one ever sees it, is it still Art? Or is the intentionality of the artist all that's required for a created thing to be Art?)

This is pretty well what it boils down to, yes!

>Is any synthesis of these positions possible, or even desirable?

They're just the same question at different levels. One concerns the creation of a work of art, and the other concerns the creation of a work of design (the use of the word "designer" here to refer to the creator of a virtual world is, of course, unfortunate).

If someone creates designs for plates that will be mass produced and sold in shops, that person would be a "designer". If they were to create the same designs for plates because that's how they can best worth through issues of personal development, that person would be an "artist". It's possible to be both: you can be an artist and find your art work is suitable as design work. In fact, it's probably the case that you have to be a designer before you can become an artist. Once you are an artist, though, it's a long time before art and design are interchangeable (because to get to that stage, youhave to have said all you need to say through the medium).

Must a virtual world have players? If you're a "designer" designer, yes; if you're an "artist" designer, no; if you're a player, well it kinda does by definition..!

[Secondary topic stuff]

>not exactly a "place," but more of sense of relative position: "here" (foreground) and "there" (background)

That sounds like you could have several different "theres" then, if you put your mind to it?

>Most of the things that merit background processing (for me, anyway) are design problems.

Yes, I can see how that would be the case if you were only using one background process - you wouldn't want to "waste" it on unnecessary memory-searching or whatever.

>they were attempts to figure out how to manipulate the internal structure of a square peg of factual relationships so that the resulting structure will fit through the round hole of constraints.

So pattern-crunching, but with some "intelligence" involved rather than mindless enumeration?

>Other problems just don't seem to benefit from that level of attention.

Is it that your background process is well experienced at doing these design tasks, or that you are well experienced at them and that makes the background process be better at them? In other words, does your background process have access to skills and techniques that your foreground process doesn't, or vice versa, or do they both have access to the same techniques all the time?

>FWIW, I've also found that background processing can work for intellectual problems, but problems with a strong emotional component don't go anywhere in background mode.

My "processes" can examine emotional concepts intellectually, but I don't feel the emotions they are examining. Emotional data is available to them, but they don't change or affect my emotional state. I can reason about how I might feel if I did something generally pretty accurately, but I suspect that's because I have a good model of how my emotions work, not because I actually feel "yech, that's gross!" or whatever.

>I'm willing to try multiple background jobs (I've got one running right now, actually), but as Cory and Eric discuss, I wonder if this would cause me to miss hearing a *ding*.

Oh, I never miss the *ding* - it interrupts me every time. If I can't deal with it right then, I tell it to go away and come back later. This is "implemented" as an "oh yes, I was *ding*ed" memory; in other words, the memory of being *ding*ed brings what caused the *ding* into my consciousness.

>As long as there's just one background process running, I know if I get any kind of internal *ding* exactly what has just happened and what the result is (usually in a blinding flash of "Ah!").

My *ding*s come with unique (metaphorical) sounds, so when I get one it comes with an instantly-crystallised expression of itself. I can't hear a *ding* and not know what made the (metaphorical) noise.

>Deliberately making use of this effect in a conscious way, however -- I agree with you; that seems much less common.

It's handy, but it's not without its problems. If you don't have conscious control of it, there's always the chance you could be surprised when something pops off the back burner. If you do have conscious control of it, you know that whatever you put on the back burner is never going to pop off it. The only way you get something popping into your head unexpectedly is when it answers a question you didn't ask.

>Now we just have to wait to hear whether Raph does this, too. ;-)

Our numbers grow!

Richard

168.

Monkeysan>Basically, you feel proprietary about what virtual worlds are

I'm not sure I understand what you mean here. Are you saying that because I co-wrote the first one, I feel I have some ownership of the overall concept and thus resent it when people repurpose it?

I guess I do have some general feeling of a legacy I'd like to preserve, but this legaicy is just one thread of a much greater tapestry that is being woven all the time by players and designers. I want to see all the other threads, too, I just don't want my "idealist" one cut.

I don't feel proprietorial over the term "virtual world", though. It's just I want a term for what it is I create, each one in the past having been assigned some specific meaning when I wanted a general one. Now, ironically, I find that I want a specific meaning when you and others want an even more general one!

>and you want a definition that is unequivocal about where the locus of creation and status of a virutal world is.

I do want such a definition, yes, because I need to be able to talk about what I create. Your suggestion, "world model", doesn't work because that term is already used in virtual world design for the stratum below that of the virtual world itself.

>Specifically, you want a definition that lays all of these at your own feet.

I want a term that describes the artefact that designers design. I'm not seeking personal glory for this, and am a little alarmed if that's the impression I'm giving! I just want to be able to talk about a virtual world in the same sense that a novellist talks about a novel or a painter talks about a portrait or a landscaper talks about a garden.

>The fact is, most of the codist arguments have stemmed around a combination of this notion of the proprietary coupled with bald assertion that 'this is just what a virtual world is.'

It's more frustration than anything. I'm saying, "look, there's this thing that I create, and it has this name, 'virtual world'. Why do you want to take the name and use it for something else? It took 20 years to settle on this name, but now I'm going to have to think of some other name?!"

Can we agree that designers do design something that they regard as being complete even when player-independent? If so, can we also agree that it's very convenient for designers to have a term that describes this artefact? Can we also agree that researchers want to study the humanistic structures that this artefact enables? If so, can we agree that it's very convenient for these researchers to have a term that describes these humanistic structures?

If so, all we're arguing about is whether the designers get to use the term "virtual world" or the researchers.

>This coupled with quarreling over alternate views, while failing to provide positive support for the codist position other than authority, can be frustrating. It seems to defy the whole notion of interchange.

I've tried to describe the codist position in terms of art, in terms of artistry versus artisanship. That's the best I can do, sorry.

>the fact is that no matter how hard you try, you won't be able to maintain sovereignity of discourse on the subject indefinitely.

I don't think I ever did have sovereignty of discourse. Right from the beginning, Roy Trubshaw and I disagreed on some things. People don't talk about virtual worlds solely due to my largesse; indeed, people don't even talk about my own published papers with much thought about whether or not I might actually read what they say about my position.

I don't have sovereignty of discourse, nor do I want it. How would I ever find out anything new?

Richard

169.

Richard Bartle: Well, ok. I think a key part of this problem is that I really don't care if you use "virtual worlds" to describe an artefact. As I've mentioned earlier, I have no problem with saying that "virtual worlds" can be "much less" as long as "much less" includes some form of interactivity.

However, it seems as if you want to close the discussion off so that when we talk about "virtual worlds" we can't use that term to refer to the "much more" that results from the technical design, even when that "much more" is an explicit goal of the technical design.

Secondly, to me the distinction between designing behavior and designing to enable behavior seems like trivially picking nits. That is, we seem to be using two different phrases to describe what is pretty much the same thing.

Can we agree that designers do design something that they regard as being complete even when player-independent? If so, can we also agree that it's very convenient for designers to have a term that describes this artefact? Can we also agree that researchers want to study the humanistic structures that this artefact enables? If so, can we agree that it's very convenient for these researchers to have a term that describes these humanistic structures?

If so, all we're arguing about is whether the designers get to use the term "virtual world" or the researchers.

IME some designers do, and some designers don't. Some designers create things that are explicitly participant-dependent. That is, you seem to be insisting on planting a hard line between artefact and participant that designers have more and more frequently attacked since start of the 20th century.

Another issue is that you seem to be trying to create another dichotomy between research and design. I don't care about research for research's sake. What I care about is developing structures that help people learn and collaborate. It is my belief that looking at socio-technical systems rather than just technical systems is necessary for that goal.

The third issue is that I don't see a conflict regarding who gets to use the term. It is entirely possible to talk about "Star Wars" as a film, as a business enterprise, and as a cultural phenomena. In addition, don't participants have an opportunity to define what your "virtual world" means to them, even if it includes things like, "the place where Jo and I hang out on Friday nights?"

170.

As an example, training volunteer or paid staff to play the roles of characters in the mythology or backstory of a world certainly qualifies as a designed interaction in my mind, but one that is difficult to boil down to an "artefact."

171.

Kirk>That is, I start with ideas about what kinds of participant behavior I want to encourage, and then pick the systems that encourage those behaviors.

Kirk>The way you phrase it, it sounds as if everyone is going at this from the perspective of "coporeal proxy! cool, what do we do with it?"

It is no different. These definitions do not attempt to provide a road-map of constructing, they provide a means to describe such a construct both constructively and deconstructively.

For example, as you said, you start with an idea, then you have [kinds of participant behavior to encourage] and [systems which encourage behavior]. You are deconstructing your idea into components for consideration and implementation to construct a training class. One can also take your completed class and attempt to deconstruct it into fundamental components to evaluate and compare it to other similar classes. They may even use the effective components to construct another class. That is exactly what this framework does.

As I explained earlier, my definition of what is fundamental to virtual worlds does not describe, in that sentence, all forms of virtual worlds, thus it is only the common denominator. There are many other components all built from that idea, and all included in the framework. I'll illustrate further on your example.

Kirk>I would argue that my first "virtual world" was a collaborative fiction role-play that was mediated over a threaded bulletin board system. In that case, the "model" was a shared conception of the rules of the space.

This is an example of deconstructing your first "virtual world"...

First, there must be a SPATIAL REPRESENTATION that is considered authoritative. In this case, it is an AD&D rulebook with modules, a novel or series of novels that describe a space, and how one can interact within it.

There is the USER, which is external to the SPATIAL REPRESENTATION which desires to interact with it. The USER is able to interact with the SPATIAL REPRESENTATION through a CORPOREAL PROXY. In this case, it is the character. The CORPOREAL PROXY is considered part of the SPATIAL REPRESENTATION.

The MEDIATOR mediates the interaction between the USER and the SPATIAL REPRESENTATION relative to the experience of the CORPOREAL PROXY. That is, the MEDIATOR receives commands from the USER concerning actions of the CORPOREAL PROXY within the SPATIAL REPRESENTATION, and transmits events experienced by the CORPOREAL PROXY within the SPATIAL REPRESENTATION to the USER. In this case, the MEDIATOR can be a GameMaster (GM), or Dungeon Master (DM), SysOp, Board Mediator, or Panel. It's not quite clear from your description. In it's simplest sense, the MEDIATOR is executing the rules of the SPATIAL REPRESENTATION, and saying "yes you can do that" and "no you can't do that".

The MEDIATION MEDIUM which can also be considered the USER INTERFACE is the medium in which the USER and the MEDIATOR communicate. In this case, it is a bulletin or message board.

Let's take a simple AD&D session, which has one Dungeon Master (DM) and 5 players who are playing through a module called "The Temple of Elemental Evil". The AD&D rulebooks and the module can be considered the SPATIAL REPRESENTATION. Each player has a character, and each is considered a CORPOREAL PROXY. The DM with his dice is considered the MEDIATOR. The DM's and player's voices, the table, with the pencils and graph paper are considered the MEDIATION MEDIUM.

How does this compare with a MUD? The SPATIAL REPRESENTATION, MEDIATOR, MEDIATION MEDIUM, and CORPOREAL PROXY are all contained in a program. I call that program a VIRTUAL WORLD. In AD&D, the DM, dice, pencil, paper, table, rulebooks, character sheets, and module combine to become a VIRTUAL WORLD.

From this comparison, I can say how one can look like the other to an observer. If I change the MEDIATION MEDIUM to chat channels for each separate player's communication to the DM, standardize communication between the player and the DM, and blind the players to the SPATIAL REPRESENTATION (supply rulebooks, module, character sheets only to the DM), the chat channel can effectively begin looking like a MUD from the perspective of an external observer. When a player says "look", the DM supplies what the character currently views. When a player says "stats" the DM provides that player with information about his character. When a player commits an action through his character, the DM needs to update the world response to all players in their respective perspective. This is very difficult for a human, but can be accomplished well with a computer.

This can easily explain why rules written by Gary Gygax are so readily transferrable to a program written by Richard Bartle. This can explain the parallel in response between your little brother ripping up your character sheet in AD&D and deleting your character in Everquest. The parallel between players quibbling over loot in AD&D, and complaining about automatic loot distribution in Dark Age of Camelot. Without a CORPOREAL PROXY we wouldn't ever be discussing the sociological effects of gender-bending avatars.

172.

Flatfingers>but as Cory and Eric discuss, I wonder if this would cause me to miss hearing a *ding*

Just to be clear, I did not say one would miss hearing the "ding". I said that with multi-tasking, there is greater possibility that, after the "ding", one could lose what the "ding" was about. For example, "I totally had it, but I now I lost it. I wish I wrote it down." That is, you experienced a ding, but forgot, exactly, what was in your mind when it "ding"'d. I postulated this is because you didn't continue to strengthen the connections. If you continued churning the solution you may have had a better chance remembering it. That is what I do when I don't have pen and paper after I ding. I attempt to remember it by growing new connections on it (how do I communicate it, apply it, what are the implications). I may not remember all the connections I built upon, but I have greater chance remembering exactly what it was that "ding"'d.

One application of this, in tutoring, is once the student understands, you don't stop there. You ask the student to apply the understanding in multiple problems. This has two advantages. It shows you the student really does understand, and increases the possibility of the student continuing from that point at the next session. Had you not done that, you may find upon the following session, the student re-asks "Now how did that work again?" I also use this technique to reduce start-up costs between programming complex applications or writing complex papers. I don't just stop. I use a cool-down phase to re-evaluate my position and where I'm going and write some notes on it. This keeps me from starting up again and saying "Now what was it I was trying to do here?" and spend an hour trying to get my mind back at that previous state.

173.

Richard> I want a term that describes the artefact that designers design. I'm not seeking personal glory for this, and am a little alarmed if that's the impression I'm giving!

Do not be alarmed! I wasn't implying personal glory when I said "at your own feet". At that point I just meant you as a designer. I really didn't intend any of what I wrote as an attack, btw. :)

Richard> I just want to be able to talk about a virtual world in the same sense that a novellist talks about a novel or a painter talks about a portrait or a landscaper talks about a garden.

I completely understand. The problem is that what you desgin is unlike any of these things in a pretty fundamental way. So it may not make sense to talk about virtual worlds in the same way. We may need to come up with new ways of talking about virtual worlds.

It's also important to note that even if, on some accounts, you do not actually 'create' a given virtual world, I think there is *no* doubt that you (the devs) have designed a given virtual world. SL as a virtual world might be said to be *created* both by players and designers; however, the virtual world was certainly 'desgined' by the designers, even if part of that design involves allowing others to build on to it (which involves design on their part). This seems very close to how landscape architects talk about their work, fwiw, so I don't think you lose much in the move from 'create' to 'design'.


Richard> All we're arguing about is whether the designers get to use the term "virtual world" or the researchers

I don't think anyone is arguing that designers shouldn't get to use the term 'virtual world' to describe what they *design*. What's at issue is whether, in addition, you actually *create* (bring into existence yourself) virtual worlds.

In addition, it may turn out that virtual worlds are not enough like real places to make your use of 'virtual worlds' problematic from an ontological point of view. In that case, researchers across disciplines can just accept that the term 'virtual world' is simply a misleading metaphor, that can be used quite freely and liberally.

In my view this is the big question and the motivation for my pushing on your definition of virutal world. It seems to me that if virtual worlds are in some very real way 'places' or 'worlds', then we should use the term virtual world to describe those places and worlds. And these new 'places' or 'worlds' may not consist soley of the code your teams create or even its execution in hardware. Who knows? Either way though, devs still stand as the designers, as the artists, as the scientists who enabled the birth of their virtual world. It just might be that you don't 'create' them.


On another note:

Kirk, I find your claim that D&D campaigns might count as virtual worlds really interesting. I certainly found myself thinking the same thing when we wrote Smartbomb. It is at least plausible. Yet such compaigns have no 'code + data', no 'server' and 'client' in any normal sense. D&D campains have features that can be seen as analgous, but that's not the same thing. Thanks for bringing that up.

174.

Eric: Perhaps it is just me, but I find it frustrating to be told what I really "fundamentally mean" when I talk about how I try to design, or how I engage in online spaces.

I don't think this conversation is really about models, mediators, or representations. I think we are all in agreement that these things are essential. I think the two key issues of debate are the following:

1: Are there aspects of designing interactive spaces that go beyond just the creation of technological artefacts?

2: Do the actions of participants become a critical part of the identity of that space?

I think the answer to both of these questions is "yes." This has less to do with defining the minimum possible virtual world, and everything to do with defining, "What is Norrath/LambaMoo/The Forgotten Realms/Morrowind."

175.

Monkeysan>I am not aware of a any studies that have shown that the processing of a particular kind of input (aural, visual) has built-in 'content' discriminators.

I don't know what you are saying, so I can't argue with it. I can only consider that you didn't quite understand what I was saying. Let me simplify:

Although stimulus may be applied in a common medium, it may not require the same requirements of focus.

Here is a simple illustration of what I'm talking about:

Given the two visual stimuli:

Stimulus #1 = [FQRPFHXTIW]
Stimulus #2 = [REPETITIVE]

Each stimulus has 10 letters. I flash each of them at you for 0.5 seconds. Ask the subjects to duplicate on paper what they saw.

How many got Stimulus #1 correct? How many got stimulus #2 correct? Is the number different? Why?

Repeat the experiment, but apply stress. For example, an obnoxious arythmic sound, like the squawking of excited birds or dogs barking. Or simply add a time limit that they only have 3 seconds to write it down. Do the numbers of correct and incorrect duplication change?

176.

Or to put it another way, is there something to be said for the fact that "placeness" for a participant sometimes involves something expressed in a classic theme song: "Sometimes you want to go, where everybody knows your name, and they're always glad you came."

177.

Eric the problem with your 'deconstruction' of D&D is that in every campaign I've ever been involved with, exactly NONE of the players thought that the rules provided what you call the SPATIAL REPRESENTATION. The spatial representation was a collaborative product of the DM and the players. Even in the (very rare) instance where we based a campain on a prepublished or prewritten module, we did not consider those rules as the rules of SPATIAL REPRESENTATION. In fact, we didn't consider even consider them rules.

And believe you me, there was no sense in which the players saw themselves as USERS. They saw themselves every bit as much involved in creating and coordinating the details of the 'virtual world' as the DM.

Also I think CORPOREAL PROXY is very misleading. For one thing, CORPOREAL implies that the player must be embodied in the virtual world. There are military virtual worlds where some users are instantiated only by a location in space. That is, they have a point of view and that's it. Second, the word PROXY implies AGENCY on the part of whoever/whatever is acting as the proxy. Avatars, character sheets, and data do not have agency.

All that said, I think that you have presented an elegant model of the information flow in a virtual world. I just don't think it's a model of virtual worlds. That is, it doesn't tells us what a virtual world is or provide a definition of one.

178.

Kirk>I don't think this conversation is really about models, mediators, or representations. I think we are all in agreement that these things are essential.

When I read this thread, many of the arguments concern exactly that. Questions like "What is a world representation" and "what is essential to be called a virtual world" and "What is the difference between graphics and text interfaces to virtual worlds?". There were multiple posts just trying to describe what a "server" is. Now that I answer what is essential and specifically define my words, that's not what this conversation is about. It's difficult to talk about something you cannot define.

Kirk>I think the answer to both of these questions is "yes." This has less to do with defining the minimum possible virtual world, and everything to do with defining, "What is Norrath/LambaMoo/The Forgotten Realms/Morrowind."

I would answer both questions yes, too, if I knew what you meant by "space", "artefact", "actions", "participants", "critical", and "identity". As I've said before, the framework is not just about the minimum possible world. It is about all possible worlds and can facilitate defining Norrath, LambdaMOO, Forgotten Realms, and Morrowind in explicit terms to compare them, determine what is different, and discuss the effects of their differences. Isn't that what this conversation is about?

I've provided experiments, observances, explicitly defined terms, concrete cases, and illustrated a process of deconstruction and comparison. Why do opposing arguments seem to have such difficulty in presenting the same?

Kirk>"Sometimes you want to go, where everybody knows your name, and they're always glad you came."

That is a potential effect of a virtual world. If this was, or was essential to, a virtual world, we would be unable to discuss other constructs that do not exhibit close-knit communities regardless of their potential to. Further, I cannot guarantee close-knit communities in a virtual world, I can only attempt to facilitate their emergence. Even further, what happens when you transplant a close-knit community into a virtual world? Clearly, it was not the virtual world that facilitated it. It may only have provided the mechanism to maintain it. Because of the characteristic weakness of self-containment of virtual worlds, though, even if the virtual world failed to have the components necessary to sustain a community, such failures can be circumnavigated by communication external to the virtual world. The framework I work in discusses these specifically as matters of SOCIALIZATION and CONTAINMENT which can be affected by matters of ACCESS and SPATIAL REPRESENTATION.

179.

Monkeysan>every campaign I've ever been involved with, exactly NONE of the players thought that the rules provided what you call the SPATIAL REPRESENTATION.

You have a clear misunderstanding of the term. Please, define it in the terms you understand it so that I can address it.

The rulebook says what you cannot do, what you can do, and how you can do it. The campaign module provides maps which illustrate to get from this area, you must traverse this area. These are the creatures you may encounter. The rulebook says how you create a character and what that character can do. All this is the SPATIAL REPRESENTATION. This may be similar to what Bartle is talking about when he says the world and it's physics. This provides a structure for what everything is, how everything works, what it looks like, and how it interacts.

I did not come to this conclusion by taking a survey of all people who play AD&D and never suggested I did, as you argued. I came to this conclusion by deconstructing known virtual worlds, including AD&D. It is used as a model of description and comparison where each term has an explicit definition.

Monkeysan> there was no sense in which the players saw themselves as USERS.

The direct analog would be players. Thus the difference between the players guide and the DM's guide. I never said the players saw themselves as USERS. I identified them as such. Their direct analog is a USER of a MUD, which I consider a virtual world. Thus the term.

Monkeysan>They saw themselves every bit as much involved in creating and coordinating the details of the 'virtual world' as the DM

I can see it now:

DM: You see 4 goblins.
Player1: I kill them. They are all dead. I open the treasure chest and find 2000gp and a Bag of Holding.
DM: Um, no you did not.
Player1: I am every bit involved in creating and coordinating the virtual world as you.
DM: Says who??
Player1: Says Monkeysan.
DM: You are only as involved as we agree to.
Player1: Nuh uh. Every bit as involved as you.
DM: We need to talk.

I never said the DM and the players are not involved in creating and coordinating the details of the virtual world. Clearly, in the case of AD&D, they must come to a consensus on what comprises the spatial representation, that is, they must decide which rules will be followed at some point. I only described how the roles are different. In AD&D I can argue with the mediator (DM) and take my lead figurines and leave if I'm not enjoying myself. Normally, if the computer is a mediator, there is little room for argument.

Monkeysan>CORPOREAL implies that the player must be embodied in the virtual world.

Again, you focus only on the real world. I am discussing a virtual world. CORPOREAL can be considered a tangible agent of interaction, and thusly, a tangible agent of interaction within the spatial representation. In order to interact with the SPATIAL REPRESENTATION, the user must act through something which exists within the SPATIAL REPRESENTATION.

Monkeysan>some users are instantiated only by a location in space. That is, they have a point of view and that's it.

That point of view is a tangible location in the virtual space that can be moved by the user. This construct is what creates the viewing frustum. I made no mention of the dimensionality of corporealization, or whether such corporealization can be viewed by other users, or by the user themselves. It is still a tangible entity recognized within the SPATIAL REPRESENTATION.

Monkeysan>the word PROXY implies AGENCY on the part of whoever/whatever is acting as the proxy. Avatars, character sheets, and data do not have agency.

When a player wants to swing a sword at a goblin, they must say to the mediator their avatar/character will do so. The avatar is authorized to act on behalf of the user. That is, the avatar has the capability of swinging the sword on the goblin, and thus has agency the user does not. The avatar becomes a corporeal substitute for the user. That is the very definition of a proxy. Such sword swinging requires corporealization, and thus corporeal proxy. Consider further common use of the term Internet Proxy and Proxy server and what they do.

Monkeysan>it doesn't tells us what a virtual world is or provide a definition of one.

That's basically what I did. Look back through the posts for the definition. You may have missed it.

180.

Eric: Well, I don't think the specific examples are helping much. For example, while you show that your framework can describe D&D well, it does not really make the case that your framework is the only possible one, reasonable one, or even the best one. They are not convincing to me that the notion of looking at a virtual world as a designed artefact, is necessarily better than looking at it as a behavior setting.

Your framework is good. But I think it's missing an important piece:

The MEDIATOR mediates the INTERACTION between the USER and the SPATIAL REPRESENTATION relative to the experience of the CORPOREAL PROXY. (emphasis added)

To me, it's as plain as the rather large nose on my face that everything that is designed is designed with a specific INTERACTION in mind. A hammer is designed for the INTERACTION of pounding things. A cup is designed for the INTERACTION of holding things. D&D, Morrowind, NWN, and Everquest are designed for the INTERACTION of fantasy roleplay. Adult fetish muds are designed for a different kind of fantasy role play. Doom, Quake, Marathon and Halo are designed for the INTERACTION of competition through virtual combat. A cafe is designed for the INTERACTION of selling and buying beverages and food.

What is the minimum I need to know to understand what Eric's Virtual Room (EVR) is all about? Ok, we have a MEDIATOR, a USER, a SPACIAL REPRESENTATION and a COPORIAL PROXY. I'm going to think, "what's the point." Now if you say that I can make the bed, sort the sock drawer, and read the books, now I know what you are talking about.

When you tell me what kinds of INTERACTIONs your system supports, then I can start getting a vision of what is involved.

It seems we are talking a bit at cross purposes here. You consider "essential" as the lowest common denominator over all spaces. I consider "essential" as the least you need to say in order to provide a meaningful description of your space.

To me, saying that we should consider user behaviors as not essential to "virtual worlds" is rather like saying that pounding is not essential to hammers.

181.

And of course, all this is ignoring the facts that what a user sees as the quintessential aspects of a "virtual world" may be something like, "where I go to meet my friends on Friday." I don't see why we can't use a multiplicity of frameworks and views of what we are talking about, or why the person who designs the artefact gets primacy over the people who use the artefact.

182.

Eric>The rulebook says what you cannot do, what you can do, and how you can do it. The campaign module provides maps which illustrate to get from this area, you must traverse this area. These are the creatures you may encounter. The rulebook says how you create a character and what that character can do. All this is the SPATIAL REPRESENTATION.

It does indeed provide rules to that effect. However, they are not exhaustive. There are many things in the campaign I can and cannot do that the ruleset has no governance over. To that extent your model doesn't fit the data.

Eric>The campaign module provides maps which illustrate to get from this area, you must traverse this area. These are the creatures you may encounter. The rulebook says how you create a character and what that character can do.

So, if I didn't play with a campaign module, it wasn't a virtual world? What if we didn't use maps at all? What is spatial then about the D&D ruleset?

And again, the rulebook provides guidelines, but the creators of D&D acknowledge themselves that they did not intend those to be a full account of the rules that govern a particular campaign.

Eric>The direct analog would be players. Thus the difference between the players guide and the DM's guide. I never said the players saw themselves as USERS. I identified them as such. Their direct analog is a USER of a MUD, which I consider a virtual world.

But your analogy doesn't hold because as defined above, you don't seem to allow that the USER can collaboratively involved in both determining and 'executing' the SPATIAL REPRESENTATION.

Eric>

Player1: I am every bit involved in creating and coordinating the virtual world as you.
DM: Says who??
Player1: Says Monkeysan.
DM: You are only as involved as we agree to.
Player1: Nuh uh. Every bit as involved as you.

You're presenting a false dilemma. The last exchange between DM and Player1 are not mutually exclusive.

Eric first said> In this case, the MEDIATOR can be a GameMaster (GM), or Dungeon Master (DM). In it's simplest sense, the MEDIATOR is executing the rules of the SPATIAL REPRESENTATION, and saying "yes you can do that" and "no you can't do that".

Eric later said>I never said the players are not involved in creating and coordinating the details of the virtual world. Clearly, in the case of AD&D, they must come to a consensus on what comprises the spatial representation, that is, they must decide which rules will be followed at some point.

Sounds like a contradiction to me. At one point you say that the DM is the MEDIATOR because "he is executing the rules of the SPATIAL REPRESENTATION," which you define as the ruleset for D&D. Then you say that the players are involved in deterimining which rules will be followed. Sounds like they have a role in execution, as in the case where the DM miscalculates a saving throw and the USERS correct the mistake, for example. And that sounds a whole lot like the players function as MEDIATORS too.

Eric>Again, you focus only on the real world.

No, I am focusing on the notion of what it is to be CORPOREAL, which I take in this context to mean embodied. It's hard to see how it could mean much else.

Eric>CORPOREAL can be considered a tangible agent of interaction, and thusly, a tangible agent of interaction within the spatial representation.

Well, first, there is nothing 'tangible' at all about virtual worlds.

Second, there is NOTHING about being corporeal, in any sense of the word, that implies agency.

Eric>In order to interact with the SPATIAL REPRESENTATION, the user must act through something which exists within the SPATIAL REPRESENTATION.

Not really. I can be assigned causal powers in a spatial representation without acting through anything that 'exists' in the spatial representation. When I manipulate a molecular model from the command line, for example, I am interacting with a SPATIAL REPRESENTATION without acting by means of anything that 'exists' within that representation. It seems equally possible to build a virtual world where you only interacted with the SPATIAL REPRESENATION by a means that has NO representation or instantiation of myself as any kind of object or agent that exists in the SPATIAL REPRESENTATION.

Eric>I made no mention of the dimensionality of corporealization, or whether such corporealization can be viewed by other users, or by the user themselves.

Then in what possible sense is it 'corporeal'? Points of view and points in space are neither tangible nor corporeal.

Eric> It is still a tangible entity recognized within the SPATIAL REPRESENTATION.

As I said above, it's possible to have causal powers in a SPATIAL REPRESENTATION without being tangible or even recognized as an entity within that SPATIAL REPRESENTATION.

I think you need to define tangible, because any common use of the term makes no sense in the context of a virtual world.

Eric>The avatar is authorized to act on behalf of the user. That is, the avatar has the capability of swinging the sword on the goblin, and thus has agency the user does not. The avatar becomes a corporeal substitute for the user. That is the very definition of a proxy.

Actually, it's not the case that the avatar has agency. Inanimate objects do not have agency. That is, unless you are prepared to provide a radical new theory of action and agency. Since the avatar doesn't have agency, it's not a proxy.

Eric>That's basically what I did. Look back through the posts for the definition. You may have missed it.

I wish I had missed it. You've provided a functional taxonomy for the flow of information in a virtual world, but you have not provided a description of what a virutal world actually is, as in what it is to *be* a virtual world, unless you consider virtual worlds to be purely functional constructs. But in that case, many of your gripes with other views of virtual worlds are inconsistent with your own definition of a virtual world.

Eric>I've provided experiments, observances, explicitly defined terms, concrete cases, and illustrated a process of deconstruction and comparison. Why do opposing arguments seem to have such difficulty in presenting the same?

Eric, I really don't think this claim is justified. Every one of the things you pine for has been presented by the arguments I infer you are talking about.

The first time you made this kind of claim, it was addressed respectfully and taken seriously. In subsequent posts I, for example, summarized my previous arguments again, showing how they did in fact present the kind of argumentation you claimed was missing. You also repeatedly accuse me of ‘misunderstanding the point completely’ when I point out that the words you are using don’t have the connotations you are ascribing to them. You then use the excuse that “I don’t understand what you mean” in order to dismiss arguments, implying that I am somehow being obtuse or unnecessarily esoteric.

This time I can only /boggle. As someone who has dedicated their career to cognitive science and the philosophy of science, compuation, mind and cultural transmission, I find it almost insulting to come upon the same accusation again. Especially when your 'explicitly' defined terms are incompatible with already defined and accepted uses of such terms in precisely these kinds of philosophical contexts.

Oft-times on this list, people are intimidated to stand down on a subject because "they don't make the games and therefore (allegedly) don't know what their talking about."

However, I have never seen posters told to stand down on a topic because they weren't sufficiently grounded in a discipline that takes the topic as one of its primary objects of study and which has been studying the issues surrounding said topic for decades, centuries or even millennia before digital worlds ever existed.

I'm fine with that. This isn't my blog after all, and I'm content to play by the rules and epistemic hierarchy established by Terra Nova. Besides I'm no fan of appeal to authority. It's just not compelling.

But I don't appreciate the repeated implication that somehow I am not expressing competent arguments in a manner or method consistent with intellectual debate on this subject.

No hard feelings, but that's how I see it.

Aaron

183.

I believe we are all part of an experiment designed to produce the longest thread in the history of Terra Nova.

To continue to do my part, therefore:

Richard> Designing to shape human behaviour is one thing; designing human behaviours is something else entirely. ... Designing human behaviours is not only distasteful but also immoral.

This wasn't addressed to me, but I'd like to comment on it because it's an important distinction that's worth more exposure.

Doesn't the word "rules" mean essentially the same thing as "designing human behaviors?" A rule is a limit on behavior, or the requirement of a specific behavior, that is "designed" to achieve some purpose -- so isn't "rules" a satisfactory way to describe the means by which we design human behaviors? ("Laws" would be another synonym for this mechanism when it's expressed in a political context.) If not, what's the practical difference? Is there a way to design human behavior without using rules?

If "rules" does capture the essence of the idea, doesn't any virtual world that can be created today or in the foreseeable future have to have some rules? Some rules are implicit in the finiteness of any virtual world whose code can be written. Maybe some day we'll be able to create a virtual world where if you can imagine it, you can do it... but until then, I'd say that if it's code, it's got rules. (Cf. the Code Is Law discussions.)

To follow the proposition, then, doesn't this mean that every virtual world ever developed to be played by people is, to some degree, immorally attempting to design human behaviors?

Would it be acceptable to back off slightly from this position to say merely that, whenever possible, shaping human behavior is to be preferred over designing human behavior?

I could get behind that statement completely. It recognizes the occasional need to set some limits on behavior (i.e., it allows rules), but that formulation also recognizes that it's often more effective to design metarules that allow desirable behaviors to emerge naturally.

(Some might question whether there's a meaningful ethical distinction between these two philosophies, since in both cases we're intentionally designing to get players to do something we want them to do or not do something we don't want them to do. But I'm inclined to see an important difference between "make" and "encourage.")

To indulge in a little design theory, this idea of designing to shape human behavior: would I be right in thinking that a good mechanism for accomplishing this would be "significant actions have significant consequences"? In other words, let players do whatever the hell they want... but if an action matters to the world, there will be some kind of consequence keyed to that action, and the degree of effect of the consequence will somehow be roughly proportional to the degree to which the action can affect the world.

Is that shaping? Or is it just another way of imposing rules?

Finally, I'd like to suggest an idea from political theory I've been mulling over for some years that might have some application here: laws of the form "you may not do X" are to be preferred over laws of the form "you must do Y."

It's a question of economic virtue. If I force you to expend your resources in some way by by requiring you to perform a particular action, then you have no choice in the best way to expend the resources needed for Y. If I take this approach in most things, then my world is probably going to be economically inefficient (unless I'm an intelligent, well-informed, and benevolent sovereign... which isn't likely, though it may be more likely in game design).

If on the other hand my rules tend to be restrictive -- if they take away from you one particular mechanism for action -- then, humans being the astonishingly creative types they are, you're likely to still be able to find a way to get what you want; you'll just do something other than X to get it. Because I haven't restricted Z and Q and B, you're free to innovate other (and possibly more efficient) means for achieving your goal. As long as I don't go nuts and restrict virtually everything, the economic efficiency of my world should be reasonably good.

Is there a lesson here for virtual world design?

> This isn't to say that the artefact+players isn't also a natural breakpoint, but from the point of view of defining "virtual world" I feel that the artefact itself is the better breakpoint.

I believe I understand. There are actually several breakpoints we could describe, each with its own consequences. Please let me know if the following correctly matches up with your definition (and anyone reading, feel free to suggest appropriate terminology for the question marks!).

Code = ? ("the construct", perhaps)

Hardware + code + system data = virtual world (artifact*)

Hardware + code + system data + user data = ??

Hardware + code + system data + user data + user beliefs = ???

I suspect that the problem is the word "world." Most people (I think) are going to react negatively (as some of us in this thread have) to being told the equivalent of, "even if you play in the world, you're not part of the world." I sincerely believe that most people are going to have severe trouble accepting the proposition that their acts of subcreation (starting a formal guild, organizing a weekly raid, writing and sharing fan fiction, etc.) don't exist in the "world," that a "virtual world" excludes them.

So while there's nothing inherently wrong with your specific definition of what "virtual world" means, in a practical sense it may make communication on the subject more difficult because it seems to be so different from the more obvious definition (a "world" includes the people interacting with/in that world and what they do).

> Must a virtual world have players? If you're a "designer" designer, yes; if you're an "artist" designer, no; if you're a player, well it kinda does by definition..!

Isn't Second Life a world where the players are the designers, where in many cases play == design?

Would Second Life be a "virtual world" without its player/designers?

(I don't disagree with your categorization; I'm just tapping the boundaries to see how much play there is in them. *g*)

[back to the secondary topic]

> Is it that your background process is well experienced at doing these design tasks, or that you are well experienced at them and that makes the background process be better at them? In other words, does your background process have access to skills and techniques that your foreground process doesn't, or vice versa, or do they both have access to the same techniques all the time?

I've wondered about this. I'll try to answer, but we're approaching the limits of introspection -- eventually I'm just guessing at how and why I think the way I do.

The simplest way I can describe the function/value of background processing for me is that it's more creative. When I'm try to generate a good solution to a design requirement, my conscious side (the "foreground" mode) tries the common, brute force, experiential approaches. I apply the solution processes that have worked in the past. I like to think I can do this in a fairly creative way, but ultimately it's more about applying known tools.

It's when working on the problem with those tools for a day or so generates no solution even close to being effective that I'll turn the problem over to background processing. In that mode, what I feel like I'm doing is getting rid of the conscious constraints; I'm opening up the search to consider possibilities that I wouldn't consider in a conscious evaluation. The background mode seems more creative because it tests possibilities from everything I know, rather than just from the things my conscious mind says are relevant.

Maybe a good way to describe foreground vs. background is "bounded search" vs. "unbounded search," respectively. The bounded search is efficient but may fail to find a solution outside its predicted range; the unbounded search can test more possibilities but isn't guaranteed to terminate on its own in any reasonable amount of time.

Something to mention here is that I'm in no danger of thinking of the background processing as being somehow outside of myself, or someone else's voice inside my head -- it's all me. But when the *ding* happens as an odd but good-looking solution to a hard problem seems to just appear from out of nowhere... I can see how something like that could be considered "divine inspiration." Have others in history done this same kind of thinking, but described it using some other non-computer-based metaphors? (I'm not trying to compare myself to anyone here; it's just an interesting thought.)

> I can't hear a *ding* and not know what made the (metaphorical) noise.

Exactly. The *ding* isn't a signal that's separate from the answer; it's just how we're describing the feeling of suddenly knowing the answer. (And it is probably related to the "Eureka!" effect, or what Martin Gardner described as the "Aha!" moment.)

Eric R.> Just to be clear, I did not say one would miss hearing the "ding". I said that with multi-tasking, there is greater possibility that, after the "ding", one could lose what the "ding" was about.

Distinction noted -- thanks for the correction.

The reason I misunderstood is probably given just above: for me, there's no real separation between the *ding* and the answer; they're unitary. So to miss hearing a *ding* means the same thing as saying that I've missed realizing an answer. In a multi-tasking model (i.e., there's more than one background process), saying that I'm concerned I might miss hearing a *ding* is just another way of saying that handling a background interrupt might be non-interruptable, causing subsequent background interrupts to be discarded.

(How did people describe this kind of thing before computational metaphors, anyway? Divine inspiration? Demonic possession? Alien tampering? All of the above?)

Whew. Enough for this round of the experiment!

--Bart (AKA Flatfingers)

Note: I've stuck with using "Flatfingers" here because it's an old nickname that's unique enough to help me identify my stuff when I'm searching for old messages. But recent discussions have suggested that being a good citizen here means using real names when possible. So, since the privilege of contributing here matters to me, call me Bart. Unless the other Bart is posting in the same thread, in which case resolve the collision as you see fit. ;-)

* [When I use the word "artifact" I'm not correcting the English spelling of "artefact"! I'm just using the American English equivalent since that's the standard where I live.]

184.

Kirk Job-Sluder>you seem to be insisting on planting a hard line between artefact and participant that designers have more and more frequently attacked since start of the 20th century.

There is a line there, but it's not important whether it's hard or not. All I want is to be able to call what's on the designer side of the line "virtual worlds". I'm not saying that what's on the other side is worthless or unimportant - I just don't want it taking the name so we have to think up yet another one for it.

>Another issue is that you seem to be trying to create another dichotomy between research and design.

I wasn't too happy with the word "researchers". when I originally wrote it, I had "designers" and "other people", but that seemed too vague. I changed it to "researchers", as that's what we seem to have here. It's not a good term, though (eg. I'd classify myself as a researcher..!). Maybe I should just have stuck with "other people".

>The third issue is that I don't see a conflict regarding who gets to use the term.

In general there isn't, but when you get to the level of detail of discussing whether (virtual) worlds that are graphical are fundamentally different from ones that are textual - which is what started this thread - then you do need to have this level of fidelity. Personally, I don't think the response to Cory's statement changes depending on which version you use (it's "no, they aren't fundamentally different" in both cases), however some people do seem to think it makes a difference.

>In addition, don't participants have an opportunity to define what your "virtual world" means to them

They do indeed, but here we have to be more precise than that. In the UK, most people call vacuum cleaners "hoovers", as if there were some verb, "to hoove", and vacuum cleaners did it. Even Dyson machines are called hoovers. However, to people who sell, design and manufacture them, they're all vacuum cleaners. A UK specialist blog on vacuum cleaners would have to call them vaccum cleaners, even though the majority of users call them hoovers.

Richard

185.

Kirk Job-Sluder>you seem to be insisting on planting a hard line between artefact and participant that designers have more and more frequently attacked since start of the 20th century.

There is a line there, but it's not important whether it's hard or not. All I want is to be able to call what's on the designer side of the line "virtual worlds". I'm not saying that what's on the other side is worthless or unimportant - I just don't want it taking the name so we have to think up yet another one for it.

>Another issue is that you seem to be trying to create another dichotomy between research and design.

I wasn't too happy with the word "researchers". when I originally wrote it, I had "designers" and "other people", but that seemed too vague. I changed it to "researchers", as that's what we seem to have here. It's not a good term, though (eg. I'd classify myself as a researcher..!). Maybe I should just have stuck with "other people".

>The third issue is that I don't see a conflict regarding who gets to use the term.

In general there isn't, but when you get to the level of detail of discussing whether (virtual) worlds that are graphical are fundamentally different from ones that are textual - which is what started this thread - then you do need to have this level of fidelity. Personally, I don't think the response to Cory's statement changes depending on which version you use (it's "no, they aren't fundamentally different" in both cases), however some people do seem to think it makes a difference.

>In addition, don't participants have an opportunity to define what your "virtual world" means to them

They do indeed, but here we have to be more precise than that. In the UK, most people call vacuum cleaners "hoovers", as if there were some verb, "to hoove", and vacuum cleaners did it. Even Dyson machines are called hoovers. However, to people who sell, design and manufacture them, they're all vacuum cleaners. A UK specialist blog on vacuum cleaners would have to call them vaccum cleaners, even though the majority of users call them hoovers.

Richard

186.

Sorry for the double post there - typepad is screwing up and throwing Javascript errors at me.

Richard

187.

Bartfingers>Doesn't the word "rules" mean essentially the same thing as "designing human behaviors?"

Rules are limits on human behaviour, and designing the rules is therefore designing the limits. There's a paradox at the heart of games, though, in that these limits serve to free up players to do things they couldn't do without the limits (while everyone plays by the rules). Too few rules to a game or too many rules makes the game unfun; just enough rules makes it fun.

I'd therefore say that designing rules isn't designing behaviour, it's designing for behaviour. If you design the behaviour, that's too many rules; as a designer, you want to design to enable players to behave in ways they couldn't otherwise, but you can't straitjacket them by actually designing those behaviours too. That would stop it from being a game and make it something else (a play?).

>Is there a way to design human behavior without using rules?

I think maybe I'm using a different understanding of "designing human behaviour" than you. To me, designing human behaviour is an abomination - it's like turning people into robots. How would you go about designing a behaviour? Write down all the ways you want the person to respond to different stimuli? Argh, no, I don't want that! I want to free them to behave in ways they couldn't behave otherwise, I don't want to dictate how they should behave.

>Would it be acceptable to back off slightly from this position to say merely that, whenever possible, shaping human behavior is to be preferred over designing human behavior?

I'd be much happier with that, yes. As a designer, I want to shape behaviours not through imposing them, but through argument through artistic expression. In my earlier MUD1 example about the relationship between permanent death and player killing, I was describing how I attempted to shape behaviour through the design of the game, not by stopping people from attacking each other (which would have been very easy - just don't implement it!) but by enabling them to figure out for themselves that it wasn't going to work. I was essentially embedding a political point of view in the game mechanics. Whether people took it on board was up to them, but at least they had to think about it. I felt I was trying to shape their outlook, not their behaviour, but since the latter follows from the former I suppose we can say that indirectly I was indeed trying to shape their behaviour.

>I'm inclined to see an important difference between "make" and "encourage."

Me too. It's better that people accept a law because they understand and appreciate the rationale behind the law's introduction, rather than accept it because they'll be beaten with rods if they don't.

>would I be right in thinking that a good mechanism for accomplishing this would be "significant actions have significant consequences"? In other words, let players do whatever the hell they want... but if an action matters to the world, there will be some kind of consequence keyed to that action, and the degree of effect of the consequence will somehow be roughly proportional to the degree to which the action can affect the world.

That would be an ideal, although of course it doesn't have to be that way. A designer might want to experiment with some modified "zero tolerance" strategy that comes down hard on you for minor crimes, but if you carry on regardless then you're more able to avoid capture or detection, until eventually you're pretty well untouchable and can shoot people with impunity even though other people are going to the electric chair for spray-painting derelict walls.

>Is that shaping? Or is it just another way of imposing rules?

That would be shaping.

>Finally, I'd like to suggest an idea from political theory I've been mulling over for some years that might have some application here: laws of the form "you may not do X" are to be preferred over laws of the form "you must do Y."

Ah, the Anglo-Saxon tradition versus the Napoleonic code! Everything is legal unless we say it isn't versus everything is illegal unless we say it is.

>Is there a lesson here for virtual world design?

Yes, there is (I even mention it in my book, when discussing Raph's "Declaration of the Rights of Avatars", which takes the US Bill of Rights as its model). If I were French, of course, I may have a different opinion...

>Code = ? ("the construct", perhaps)
I'd include the basic data models in this, and call it either "the code" or "the software".

>Hardware + code + system data = virtual world
This is what I generally mean by "a virtual world", yes.

>Hardware + code + system data + user data = ??
I'd call a specific, running instance of the software on the hardware a shard or an instantiation (and I reluctantly go with "server", too). This would also be "the virtual world" if I were talking about a virtual world in particular.

>Hardware + code + system data + user data + user beliefs = ???
Yes, I can see why people might want to have a term to describe this! I can't think of one offhand without using some "virtual world plus ..." format.

>most people are going to have severe trouble accepting the proposition that their acts of subcreation (starting a formal guild, organizing a weekly raid, writing and sharing fan fiction, etc.) don't exist in the "world,"

They do exist in the world, but they aren't the world.

>Isn't Second Life a world where the players are the designers, where in many cases play == design?

No, the players of SL aren't the designers. They are designers, but of world content, not of the world. They don't get to change the physics. They design within the parameters determined by the designers.

>[back to the secondary topic]
Usual off-topic warnings apply.

>The simplest way I can describe the function/value of background processing for me is that it's more creative.

More? It's the other way round for me.

I find my foreground process is more creative because it can draw on more resources (specifically, linguistic resources my other processes can't use). It can, however, get "blocked" easily when there's some heavy linguistic meaning associated with something that's just derailing my attempts to think of it some other way. Also, it gets interrupted when anything happens, so extended periods of concentration mean either a peaceful setting or that I have to start switching my senses off...

>In that mode, what I feel like I'm doing is getting rid of the conscious constraints; I'm opening up the search to consider possibilities that I wouldn't consider in a conscious evaluation.

Hmm, I see what you're getting at. I probably would consider them at a conscious level, it's just there are so many of them that I wouldn't actually want to - it would be too boring. Anything mechanistic that's going to take me ages gets a process set off to do it. I actually enjoy the trying different methods stuff, though, which is why I like to do it at the conscious level. My other processes can do it, but I don't let them loose on that kind of thing very often.

It sounds as if you're saying that your background process is more imaginative than your foreground process? That's a pretty good deal. With me, I know that whatever my background processes do, my foreground process could also do (and more). I don't have anything in reserve like you have.

>Something to mention here is that I'm in no danger of thinking of the background processing as being somehow outside of myself, or someone else's voice inside my head -- it's all me.

Yes, same here. I regard the background processes as just another part of me, like my arm or my eyes, which I can control. The "me" of my personal identity, though, is the foreground process - the one that can use language.

>But when the *ding* happens as an odd but good-looking solution to a hard problem seems to just appear from out of nowhere...

I have multiple processes, but I can tell which one of them has produced the *ding*. Accessing how it did it isn't always easy, though. How did I remember the name of some student from 1988? I know I did, but as for how...

>Have others in history done this same kind of thinking, but described it using some other non-computer-based metaphors?

It's possible, yes. I have a feeling it's a hardware thing rather than a software thing, though. Most of the religious mystics of the past seem to have had software (mind) issues, whereas I suspect with me it's a hardware (brain) thing. It's more likely to be left/right brain related than some form of self-referential mental reflection thing.

>The *ding* isn't a signal that's separate from the answer; it's just how we're describing the feeling of suddenly knowing the answer.

Yes, that describes it for me precisely.

>(How did people describe this kind of thing before computational metaphors, anyway? Divine inspiration? Demonic possession? Alien tampering? All of the above?)

"Voices inside their head"? Those Greek Gods were always talking to heroes...

>So, since the privilege of contributing here matters to me, call me Bart.

Hi, Bart!

>When I use the word "artifact" I'm not correcting the English spelling of "artefact"!

That's OK, I'll forgive you - we use both spellings here (grin).

Richard

188.

Richard: There is a line there, but it's not important whether it's hard or not. All I want is to be able to call what's on the designer side of the line "virtual worlds". I'm not saying that what's on the other side is worthless or unimportant - I just don't want it taking the name so we have to think up yet another one for it.

I guess I still don't see the conflict here. I think that if you were to ask different people from different fields, "what is a shopping mall?" you would get a variety of different definitions. Some might focus on archetecture and partitioning, some might focus on the economics of leasing spaces, others might focus on shopping malls as a site of interaction for teen communities. And yet, I don't see the the outpouring of angst from archetects about the way in which sociologists view shopping malls that I do here regarding "virtual worlds."

In what way does defining a virtual world as a collection of behavior settings or an ecology mediated by a spacial metaphor prevent you from talking about a virtual world as a designed artefact?

In general there isn't, but when you get to the level of detail of discussing whether (virtual) worlds that are graphical are fundamentally different from ones that are textual - which is what started this thread - then you do need to have this level of fidelity. Personally, I don't think the response to Cory's statement changes depending on which version you use (it's "no, they aren't fundamentally different" in both cases), however some people do seem to think it makes a difference.

I think that saying we need to approach this question all from the same framework is incredibly limiting. Different frameworks provide different forms of analysis and different insights. The "designed artefact" view opens up a rich base of tools designed to look at cinema and novels. The "behavior spaces" view opens up tools designed to look at offices, restaurants and shopping areas. The "ecology" view opens up tools designed to look at the flow of different types of resources through systems.

189.

Kirk>it does not really make the case that your framework is the only possible one, reasonable one, or even the best one.

Never said that. I'd be happy to entertain thoughts on others, which I do with Bartle's and Koster's and the many other academic papers on the subject. I only said it is -a- framework.

Kirk>Everything that is designed is designed with a specific INTERACTION in mind.

I never said a virtual world is simply interaction. In my definition there are other traits. Not all things which interact are virtual worlds, as you noted you plainly have understood.

Kirk>Now if you say that I can make the bed, sort the sock drawer, and read the books, now I know what you are talking about.

Thus the definition -and- the cases which illustrate it. I find that distinction important too.

Kirk>I consider "essential" as the least you need to say in order to provide a meaningful description of your space.

Exactly. Take an object which you plainly consider a virtual world. Identify its components. Take it apart piece by piece. At what piece is it no longer a virtual world? Do you find that what is left is a common denominator among other constructs you consider a virtual world?

The term "least meaningful description of your space" though is highly subjective and that is where I see much of the confusion. For example, I can show equivalences between a MUD and Everquest II by placing them in a context which can define both as the same. This is a comparison tool which maps similar components together. Now, Everquest II might be much much more than the MUD. When you look at the definition which compares the two, it does not serve to define how Everquest II is different, or show that Everquest II is, perhaps, much much more. That is not to say the framework is unable to do so, as it can. For a simple illustration, the property of ACCESS governs how many users can access the virtual world among other things, in this case, simultaneously. The MUD may only allow 25 users to access simultaneously, where a world server on Everquest II may allow for 5000. This difference can have clear effects on community. Another example, the MUD may not have coded guild structures to facilitate internal structures of membership and organization, where Everquest II does. These are rules which affect player relationships which I place under the property of SOCIALIZATION. An observer may find that USERS of both MUD and Everquest II assemble in guilds regardless if they are coded by the virtual world or not. The difference is that Everquest II actively facilitates it, and, depending on the particular implementation, can have other effects.

Kirk>saying that we should consider user behaviors as not essential to "virtual worlds" is rather like saying that pounding is not essential to hammers.

I never said that user behavior cannot be an essential part of virtual worlds. On the contrary, the definition I provided implies USER in the term CORPOREAL PROXY, and described it as a proxy for the USER. I've only asserted that (1) the USER need not be human, and (2) the USER need not be connected to identify the construct as a virtual world.

To use your analogy, we call the object a hammer because of its structure and what it is made to do. There are many uses for a hammer, but we continue to call it a hammer. If a robot uses a hammer, it's still a hammer.

One more analogy which I hope can integrate our two arguments. I would completely agree that there is a difference between a virtual world without users connected, and a virtual world with users connected. Each user changes the experience. In game theory I can call this the property of added value. Each player changes the game with their presence, and there are constructs to determine what that change is. I can assert that a game is components such as RULES, PLAYERS, SCOPE, TACTICS, etc. and discuss characteristics of added value with each additional player. The framework I describe with virtual worlds discusses the same.

190.

Just a bit of trivia for all:

I saved a copy of the thread in Word and it's 162 pages with standard formatting.


Aaron

191.

Kirk>I think that saying we need to approach this question all from the same framework is incredibly limiting. Different frameworks provide different forms of analysis and different insights. The "designed artefact" view opens up a rich base of tools designed to look at cinema and novels. The "behavior spaces" view opens up tools designed to look at offices, restaurants and shopping areas. The "ecology" view opens up tools designed to look at the flow of different types of resources through systems.

I agree. Using the same framework leads to impoverished analyses and prevents some useful insights from being discovered.

My question to you, though, is whether it's effective to redefine terms like 'virtual world', 'world model', 'community', 'place', 'world', 'spatial metaphor', etc. for each framework. It seems to me that without a shared vocabulary it's hard for the differing frameworks to have common points of contact. Points that help us know whether we are talking about the same object from different vantages or whether we're talking about completely different things altogether. Without a shared vocabulary I don't see how we are able to accurately apply insights from one framework to theories in another framework, let alone communicate these theories to one another.

192.

Aaron>There are many things in the campaign I can and cannot do that the ruleset has no governance over.

But it must be agreed that you can do them. This shared agreement is part of the SPATIAL REPRESENTATION. I never said all rules need be contained in the rulebook. I said the rulebook is a part of the SPATIAL representation.

Aaron>So, if I didn't play with a campaign module, it wasn't a virtual world? What if we didn't use maps at all? What is spatial then about the D&D ruleset?

It was provided as an example. If one is playing with a module, it is considered part of the spatial representation. If one is not, the non-existence of one need not collapse the spatial representation. For example, if all you did was roll dice to determine an outcome of battling characters with no relation to such actions in a space, I would not consider that a virtual world, thus the term SPATIAL. We're back to the random number generator that Kirk provided as an example. I do not consider the act of comparing numbers generated by an RNG to be a virtual world.

Aaron>The last exchange between DM and Player1 are not mutually exclusive.

The first exchange is the point, and I see you left it out. The DM and the player have a separation of roles. That is the very reason for the term Dungeon Master. I would not disagree that such roles can intersect. If you really wanted to split hairs, for example, a player may be extended the courtesy of rolling the dice to determine a hit by their character. It is an action of trusted mediation, and that is why I considered the dice, and rolling them, part of mediation. The player cannot explicitly choose the outcome of the dice in this circumstance.

Aaron>you don't seem to allow that the USER can collaboratively involved in both determining and 'executing' the SPATIAL REPRESENTATION.

On the contrary, I do allow for it. I never said the SPATIAL REPRESENTATION could not change, and I never said that the USER could not be involved in modifying it.

Aaron> the case where the DM miscalculates a saving throw and the USERS correct the mistake, for example.

Or the case where players in DAoC complained that the PRNG was not balanced. The users expressed an issue with mediation. It is important for the users to trust mediation. The mediator must still recognize the miscalculation. What happens when the users don't trust mediation. It probably has the same result in both places.

Eric>Again, you focus only on the real world.

Aaron>I am focusing on the notion of what it is to be CORPOREAL, which I take in this context to mean embodied

Aaron>there is nothing 'tangible' at all about virtual worlds.

Relative to the real world, there is nothing tangible about virtual worlds. Relative to a virtual world, there is. As I've explained before there are fundamental differences between a virtual world and the real world which can really offset our normal understanding of philosophical constructs. What might be true in virtual worlds may not be true in the real one.

As an example of a simple analog between our bodies in the real world is having a body to interact through in the virtual world. Our real bodies are corporeal relative to the real world, and the virtual body is corporeal relative to the virtual world.

Aaron>Second, there is NOTHING about being corporeal, in any sense of the word, that implies agency.

Thus I follow it with the term proxy.

Aaron>you need to define tangible, because any common use of the term makes no sense in the context of a virtual world.

tangible (adj): Possible to be treated as fact; real or concrete:

I am not talking about something -you- can touch. I'm talking, again, in terms of the virtual world, which is something a program can touch. I can say there exists a frustum viewing plane with lower left corner (x1,y1,z1) and upper right corner (x2,y2,z2) with normal vector V. This is a fact. To the program, this exists, can be manipulated, and is internal to the spatial representation. The user is able to move this construct around.

Aaron>many of your gripes with other views of virtual worlds are inconsistent with your own definition of a virtual world.

Your view of what you consider a virtual world, as I understand you, is also what I would consider to be a virtual world. But it seems your contention is that what I view as a virtual world is not. Many of my arguments are only in response to your criticism of how I view a virtual world to be. I believe I am only examining the construct at a more fundamental level. That is to say, the framework I operate in can explain the difference between the CNN web site and Everquest, or an IM channel and a MUD, but it can also discuss the commonalities and differences of Morrowind and Everquest or DAoC and SL. That is why, before, I attempted to communicate that we actually agree in a general sense on what virtual worlds are, it's just how we look at them and how we draw lines in the grey areas.

Aaron>incompatible with already defined and accepted uses of such terms in precisely these kinds of philosophical contexts.

What is the precise philosophical context for defining constructs outside of our material plane? How do I describe something that exists outside the Universe? If I stood outside the Universe, and found that I was the dream of an alien slug in some alternate reality, how can I still maintain that my body back in the Universe is corporeal? In physics, are the singularities of black holes considered part of the material plane? If my consciousness leaves the Universe, does it still exist? Can it come back? If it does, where did it go?

When you operate relative a virtual world, one is faced with these difficult constructs. It doesn't make for simple discourse. When I discuss such things with others, I attempt to look beyond the words to understand the point, and thus I use as many examples as I can.

Aaron>repeated implication that somehow I am not expressing competent arguments in a manner or method consistent with intellectual debate on this subject.

There are times where I clearly do not understand your argument. This is not a matter of any strategy or suggestive statement. I can understand that is frustrating. But I do still make an effort to address it and explain myself using different analogies or examples as a framework for response to examine common terminology. In the same sense, there are times where I understand my terminology may not be clear either, and I try to re-phrase and illustrate rather than repeat. For example, the whole real vs. virtual can become rather mind-bending at times and it is easy for arguments to become tangential (like discussing the Heisenberg Uncertainty Principle or Schrodinger's Cat).

193.

I agree that the fact of virtuality makes many things difficult to talk about.

Eric>How do I describe something that exists outside the Universe?

You can't. Nothing exists outside of a universe. Existence has no meaning outside of a world (universe). Part of the problem is there is no 'outside' beyond a given world. There is no spatiotemporal reference at all, since there is no space and time, for example. If there were, it would be part of some universe. Existence has to be defined with some world whether possible or actual. These are the same kind of problems that Locke and Leibniz argued over at the end of the 17th Century. For an excellent modern discussion of these kinds of difficulties I refer you to David Lewis' discussion of possible worlds and modality (basically necessity/contingency, etc.), which is laid out in "On the Plurality of Worlds."

Eric>I believe I am only examining the construct at a more fundamental level. That is to say, the framework I operate in can explain the difference between the CNN web site and Everquest, or an IM channel and a MUD, but it can also discuss the commonalities and differences of Morrowind and Everquest or DAoC and SL.

I agree. The model you have been using in your analysis of D&D for example is effective. My only point is that it is primarily a functional description of how virtual worlds work. By functional, I mean that you are describing virtual worlds as a process, wherein the various components you site are not defined in terms of their essential properties but in terms how they function. That's a perfectly acceptable way to do go about things. Perhaps that is precisely what you intended, in which case I understand you much better. What you didn't provide, and it's a toughy, is what physical states/properties/relationships your functional world supervenes on.

Eric>That is why, before, I attempted to communicate that we actually agree in a general sense on what virtual worlds are, it's just how we look at them and how we draw lines in the grey areas.

I think this is probably true. But rather than disagreeing about grey areas, I got the impression that you were expecting your functional theory of virtual worlds to have explanatory power in those grey areas that can't have.

Much of the debate here has had these same problems, in many cases, myself included. Part of this is for brevity and part of this is that I think people are hitting on philosophical problems that can't be hashed out in the blog form.

Eric>What is the precise philosophical context for defining constructs outside of our material plane?

This is a very good question.

Plato had a whole system for talking about constructs outside our 'material plane'. In fact, he didn't even consider our material plane real.

Descartes wrote the meditations precisely to explain that the mind and the body existed on different planes.

Current philosophers of science argue about the very same kinds of questions. For example, are the physical laws of our universe themselves physical? Many many of them would say no.

But to get back on topic, the area of philosophy that has the most relevance to the kinds of discussion we are having here is contemporary theories of consciousness. There are functionalist views, dualist views, epiphenomenal views, emergent views, eliminativist views, etc.

The reason I say it's relevant here is that philosophers and cognitive scientists have largely come to believe some kind of physicalism. Namely that the mental supervenes on the physical. This simply means that all the mental facts are determined by the physical facts.

This yields up a pretty tough problem, called the 'hard problem' of consciousness. The hard problem is bridging the explanatory gap between the physical descriptions of our brains and our phenomenal experience of ourselves.

I think that much of what's being discussed here is completely analogous. How does a set of code and data which are essentially descriptions constitute a virtual world?

Richard's view is essentially eliminativist: He thinks that there's the code and the data and that's it. Once you have those, you have a (minimal) virtual world. Everything else is illusory.

Raph's view seems mostly functional: He seems to think that, yes you need what Richard requires, but you also need to have that code and data running before you can instantiate a virtual world. The description provided by the code and data alone is insufficient. Nevertheless, he is quite clear that all that takes place on 'the server' (that part of the discussion, though, is really a different topic)

Kirk seems to be leaning more toward a kind of behaviorist view: Virtual worlds are just sets of behaviors. Yes, there are structural/organization requirements that the behaviors supervene on, but the virtual world itself is primarily the behaviors.

The problem with this discussion that discussions of theories of consciousness don't have is that theorists of consciousness can agree to some reasonable degree what consciousness is. After all we assume we are all conscious, and despite differences of perception, we feel pretty sure that while we differ in token we are similar in type. As famously put by Thomas Nagel, consciousness is the 'what it's like to be' quality. (Nagel's 1974 paper "What It's Like To Be A Bat" is probably one of the most read papers on the subject bar none.)

Our discussion of virtual worlds has currently lacked that. That's part of the problem and why we are often talking past each other, imo.

For my own part, I'm convinced that there is an analog that can capture virtual worlds--namely, the 'what it's like to actually be somewhere that isn't there' quality. Note the importance of the word 'actually'. I use that because I'm not talking about the transport that books, films and television offer. It also puts a big responsibility on me to say in what way virtual worlds are Real places. It is to that end that I argued with Raph and Richard over the importance of an observer, human or otherwise.

THE TAKE HOME MESSAGE ACCORDING TO MY THEORY IS THAT WHILE THE VIRTUAL WORLD CONSTRUCT IS WHOLLY A DESIGN ARTIFACT OF THE DEVs, THE VIRTUAL WORLD ITSELF MAY REQUIRE AN OBSERVER BEFORE IT "BLOOMS" (EMERGES, IS CREATED, BECOMES, EXISTS) INTO A "REAL" PLACE, TO THE EXTENT THAT VIRUTAL WORLDS ARE "PLACES" AT ALL, RATHER THAN COMPLEX ILLUSIONS.

I don't know if that answers any of your questions or simply obfuscates them further.

NOTE: If I mischaracterized anyone's position please don't jump down my throat. I provide those caricatures solely to help illustrate what I see as an analogy between some of the philosophical problems of consciousness/mind and those of virtual worlds. Further, please don't take my summaries of various philosophical positions too seriously. They are grossly abbreviated, since this thread and my post is clocking record amounts of space.

BTW: A fantastic overview of these problems is given in popular terms by Rita Carter in Exploring Consciousness.

194.

Aaron: Kirk seems to be leaning more toward a kind of behaviorist view: Virtual worlds are just sets of behaviors. Yes, there are structural/organization requirements that the behaviors supervene on, but the virtual world itself is primarily the behaviors.

Well, I think this is one of the primary points of difference. I'm approaching Virtual Worlds from a CMC/Social Networks perspective, and not from a Media Design perspective.

195.

Yeah, I think that view has a lot to recommend it. I hope I didn't butcher your view too horribly.

196.

Aaron: Yeah, I think that view has a lot to recommend it. I hope I didn't butcher your view too horribly.

Oh, you didn't. I just felt the need to clarify because I don't think we will ever have much of a consensus between such radically different starting points.

However, I don't think it's that big of a deal, in Education a School can be:
A building,
An educational system,
A community,
A labor system,
and a technical system.

In many cases these are fuzzy boundaries. My sister worked at a school that was housed in a former warehouse that was never intended to be a school. All the fixtures of a school were just moved in and bolted on.

197.

True that, Kirk.

However, though we may come from different starting points, I don't think we end up at as radically different points as it might seem. To be honest my primary intellectual interest in digital worlds is how they operate both as exemplars and laboratories of cultural transmission and cultural evolution, a project that definitely overlaps with social network theory. I don't know about CMC, cuz I'm not sure what that stands for. I'll guess and say computer-mediated communities, but that's just a stab.

Anyway, my theory of digital worlds is primarily designed to provide a philosophical framework that allows me to make at least a plausible argument that digital worlds, though virtual, are in many ways very real places with very real cultures. And I mean real with a capital 'R'. I want to be able to show that digital worlds are more than 21st century pool halls.

Aaron

198.

Cory> Physically simulated 3D worlds are fundamentally different from text worlds.<

“Fundamental” is more than I would care to speak to. But I do see a significant functional difference between (model based) 3D worlds and (description based) text worlds. That is, the possibility for emergent behaviour. Much of the emergent behaviour I am interested in relies on the complex interactions between objects in the four dimensions of space and time. I don’t see how you could do this in a script driven text based system.

Here is a comparative example from an area in which I very familiar, growing virtual flowers. The current popular algorithms are all descriptive in nature. Put a brown stalk here, long red leaves here, and blue flowers on the top. This could be easily implemented in either a 3D world or text based script generated world.

But when you move to a more biological like system, the shape, color, size etc become emergent properties of a gene based seed. Each cell grows and splits according to its immediate environment, and the changes that engenders in gene expression. Even in the relatively simple model I am using, given a random gene string, there is no way to predict the final appearance of the flower short of growing the seed. You have to run the 3D model to see what behaviour emerges. Some simulation of space and time is essential to providing a view of the world. I don’t see how you would generate that level of emergent behavior in a pure text based world. I see emergent behaviour as a key to providing consistent but unpredictable worlds at a reasonable cost, and some of the most interesting emergent behaviours require an underlying 3D model. So 3D worlds have the edge there.

199.

My two cents on defining “Virtual World”.

First of all, with due respect to Richard, I would like to see the term “Virtual World” drop out of usage. “Virtual” has too many overtones of “not Real” and hence “not significant”. I regard the computer generated worlds I inhabit, and am attempting to design as decidedly real, if clearly computer rendered. Hence

Hardware + code + system data = Computer Generated World.

This is the artifact world designers build. This computer based definition gets rid of the D&D world. And even more so of Barsoom, or Tolkien’s Middle Earth, which I would call imaginary worlds.

Hardware + code + system data + user data + user history + user beliefs = Online World

Here the emphasis switches to online community that inhabits the world. The significant shift is that people are now online together in the built world.

It’s a bit like the word “house”. House builders build houses, and when they say “house” they mean the artifact they built. But when friends stop me on the street and say “how is your house doing?”, they mean more than just the structure, but also the community inside it. It would be good to have separate words for those wider and narrower definitions.

200.

Kirk Job-Sluder>In what way does defining a virtual world as a collection of behavior settings or an ecology mediated by a spacial metaphor prevent you from talking about a virtual world as a designed artefact?

When it comes to refuting the suggestion that "Physically simulated 3D worlds are fundamentally different from text worlds", which is how we got into this discussion.

I'm also unhappy with the idea that a virtual world isn't a virtual world unless it has players.

Richard

The comments to this entry are closed.