Virtual (Wikipediac) Worlds?

On Slashdot they reviewed Robert Glass' book Facts and Fallacies of Software Engineering.  Fact 21 snagged by eye:  "(f)or every 25% increase in problem complexity, there is a 100% increase in solution complexity."   On BoingBoing they continued with  the controversy of Wikipedia and whether a publically vetted (open source) knowledge base can be trusted.  All this led to a few thoughts about scaling problems in future virtual worlds...

...Because of the Wikipedia controversy, Alex Halavais conducted an experiment: he deliberately introduced 13 errors into the Wikipedia only to find them corrected "within hours."  On Alex's blog, a provocative comment by Danyel Fisher pointed out that:

(What) you may have forgotten that there’s a (fairly large) number of people who read http://en.wikipedia.org/wiki/Wikipedia:Recent_Changes religiously (usually as an RSS feed)... So when you made your change, it popped up on a lot of screens—and one or another glanced at it, said “that’s wrong,” and fixed it. Indeed, check out the RC Patrol http://en.wikipedia.org/wiki/Wikipedia:RC_patrol for some idea of how this happens.

Danyel then goes on to suggest that  this approach to change is "cataclysmic" insofar as  "a great deal of interest gathers around a single page, many changes are made, and a new steady state is negotiated."  For this, he cites some stunning work (graphically speaking) out of IBM Watson Labs.

Virtual worlds...

I wondered about Robert Glass' Fact 21 and whether it might underlie the problem of virtual worlds in the future:  big places, complete places, lots of emergence... a place where scale gets in the way.     Personally, I tend to believe the most difficult scaling problems lie not in engineering and technology per se... but at the interfaces with, and the interactions within the player systems: the game world laws; the stories versus simulations; and all those cataclysmic, productive moments waiting to happen.    I can't help but think that we should somehow harness all those buzzed folk scanning RSS feeds - and  turn virtual worlds into collaborative worlds, somehow,  Wikipedia-like...   However, I cannot yet imagine what this sort of world would look like, can you?

Comments

[Cory Ondrejka]

1.

Uh, Nathan, not be pimping, but have you played Second Life? We may not have implemented it with the beauty of Jim's thesis -- oh, to have infinite time and resources -- but what you are describing is exactly what Second Life is. That has good and bad points, and we don't have the SL version of RSS figured out (yet), but we're working on it.

However, to the earlier part of your post, I'd say Glass is probably being conservative with his ratio. Having a more-or-less infinite list of "never been dones" and "can't be dones" on your to-do list does keep things interesting :-).

[Yitz Jacob]

2.

Nathan, a thought: wikipedia has one thing going for it (that coincidentally i think Second Life might have as well), namely it can be compared to the real world, or any other world for that matter, at least something reasonably objective.

The error-checking functionality of wikipedia comes from this feature. To have rss-like RCPatrols for a virtual world, you'd need to have a sort of objective framework or you would need a very simple TOS and that would be the only thing enforced by anyone and everyone.

I guess if you wanted to leave the guidelines more vague you could have a weighted democracy, like at e2 (and others) where some initial snobs 'trained' future snobs via a feedback system that weeded out anyone who couldn't understand the unspoken 'esthetic'.

The weighted democracy might drive people away though. I don't have enough experience with wikkis yet to know whether or not such a system would be necesary.

[Cory Ondrejka]

3.

Yitz> I guess if you wanted to leave the guidelines more vague you could have a weighted democracy, like at e2 (and others) where some initial snobs 'trained' future snobs via a feedback system that weeded out anyone who couldn't understand the unspoken 'esthetic'.

Oh boy, did you just describe the Well. Such as classic problem -- and one that Second Life is very conscious of -- when your early adopters decide that they are done accepting new users.

[Tobold]

4.

The average person knows something better than you. As we live in a highly complex world, and knowledge can be split into tiny sub-domains, it is not actually very difficult to be an expert on something. So nearly everybody can contribute in a meaningful way to the Wikipedia.

It is a lot less clear whether the average person would be able to imagine, code, and implement an idea for a MMORPG. At the very least such a world would be highly dependant on having very easy to use, but powerful tools.

Another problem is that stating a piece of knowledge for an encyclopedia can be verified as being true or false. Adding something to a virtual world is more like art, there is no true or false. Whatever you create, some people will like it, and others will hate it. That makes peer review a relatively complicated process.

[Nathan Combs]

5.

Hmmmm. Sure one can recreate the objects of some objective reality with their physics and appearance in 2nd life... But is one creating as deep measure of knowledge (in the sense of, say KR defined here http://medg.lcs.mit.edu/ftp/psz/k-rep.html). For example, I don't see how one might draw any inferences against those objects as one might infer an ontology and relationships etc. from a wikipedia.

Game worlds, for reasons ya'll mentioned are also problematic.

[Jim Purbrick]

6.

The first scalability problems people tend to think about with virtual worlds are number of simultaneous users and number of mutually aware users. These are the sort of technical issues addressed by MASSIVE-3. In practice it's important to support a fairly large number of simultaneous users, but beyond several hundred, people tend to just perceive "lots".

Having made lots of simultaneous users possible, the next problem is to create lots of things for them to do. This is the software engineering scalability problem. The more you try to make a virtual world do the more complicated the virtual world becomes and making complex things simple is hard. There are also problems like how to construct a believable city in a reasonable period of time. Rome wasn't built in a day. These seem to be the scalability issues that I spend most of the time wrestling with at the moment.

Scaling up the number of people building the virtual world seems to be an attractive solution to that problem, but brings problems of its own, like: Who do you trust? How do you organise the construction effort? Who owns the world? Who's version of the world is definitive? How do you maintain quality or a vision? This is the Wikipedia problem mentioned above. Some of these are technical challenges (reconciling different versions of a virtual world is a large part of what network VR is about). Most are community and social issues. It's easy to end up with a virtual landfill site.

[ren]

7.

Sorry I’m confused.

I kind of see Cory’s point. SL is an environment where one can create objects that can have all kinds of relationships with each other. Objects can be scripted so it’s not just their appearance and physics that one can affect but lots of other things. So I’m sure one could create a taxonomy of knowledge objects, and when the RPC XML stuff kicks in could possibly RSS them.

But I’m not sure why you would – skimming the paper that you referenced then wiki type implementations seem fine as we are talking about conceptual relationships.

But when you say:

>and turn virtual worlds into collaborative worlds, somehow, Wikipedia-like... However, I cannot yet imagine what this sort of world would look like, can you?

I really am lost as to what you are getting at.

If we take game world laws, then AITID uses the combination of forums and a voting system to create and refine laws – this seems pretty wiki like. But the narrative bit?

[Nathan Combs]

8.

I kind of see Cory’s point. SL is an environment where one can create objects that can have all kinds of relationships with each other. Objects can be scripted so it’s not just their appearance and physics that one can affect but lots of other things. So I’m sure one could create a taxonomy of knowledge objects, and when the RPC XML stuff kicks in could possibly RSS them.

Sure. As with any comparison, however, there are likenesses and contrasts. For example. In the Wikapedia one can represent "Classes"/"Types" as well as instances (objects in SL). Another contrast, in the Wikapedia one can describe processes and events... Also consider the "KR negotation dynamic" suggested earlier (the cataclysmic event) - knowledge norms, metrics, and enforcement... Are any of these significant? Who knows.

SL is one "Virtual (Wikapediac) World." Are there other useful forms/possibilities?

[Andrew Burton]

9.

> Uh, Nathan, not be pimping, but have you played
> Second Life?

Hahaha. Since when is Second Life collaborative or designed with a unified purpose like a Wiki? Since when do Wikis have anything like land barons or rating gamers? Second Life is nothing like a Wiki, don't let Cory's nepotism fool you.

I'd look at There as being a potential Wikiworld, once they start selling their technology to individuals, or maybe Active Worlds right now. Both these places are designed to connect to the web (There has internal/menu-driven links, and AW's interface is built within a webbrowser), which means that you can easily tie them to wiki's or other systems of persistent storage.

Second Life doesn't tie to the web and only allows for 255 character (via email or XML-RPC) bursts at a time and has no form of automated, internal storage. So, yeah, sure, you could setup an inworld wiki system with notecards that exported data externally, but if it ever gained popularity it'd never be able to keep itself updated. Ask Cory about llWrite2Notecard() sometime.

So, yes, Second Life is great if you want to be able to build castles with someone, but in all seriousness SL is more of an Oekakiworld than it is a Wikiworld.

[Cory Ondrejka]

10.

Damn . . . dingos ate my post (OK, juggling too many Firefox panes ate my post) so let me try again . . .

Nathan, it sounds like you are saying that linked text (Wikipedia) is a more representative form than physical representations + behavior + linked text (SL). While I agree that Wikipedia is fantastic and currently has far more archived knowledge than SL, I don't agree that the subset is somehow superior to the superset. For example, architectural plans have more meaning when combined with the house that they represent than they do by themselves.

Andrew, I completely agree that SL has a long way to go WRT properly handling text (for those who aren't in on the joke, llWrite2Notecard() is a much requested feature that is unfortunately very difficult to add to our system -- but a user suggested a work around llCreateNotecard() that will be added in the near future) and integrating with the web (and the rest of the real world), however, Nathan's question was (I thought) about collaborative virtual worlds, which SL is clearly an example of. I didn't say that you should use SL (yet) to store your wiki of linked text.

Regarding your complaint that SL lacks clarity of purpose, I agree that the whole world isn't forced in one direction, since that was by design! However, at the level of 10 - 100 people (private islands or large groups building together) these are plenty of examples of clarity of vision and purpose.

You seemed to interpret "turn virtual worlds into collaborative worlds, somehow, Wikipedia-like" meant that they needed to be connected to the web. While I think that you are correct in that being a great direction for digital worlds to go, I don't agree that is a requirement.

[Wouter Ryckbosch]

11.

Quote:
Another problem is that stating a piece of knowledge for an encyclopedia can be verified as being true or false. Adding something to a virtual world is more like art, there is no true or false. Whatever you create, some people will like it, and others will hate it. That makes peer review a relatively complicated process.

I don't think that's a given, actually. At least, I don't think the line between those is as clear as you describe it. While we may consider many facts in an encyclopedia to be either "true" or "false", that doesn't mean they're objectively to be classified that way. The content of encyclopedia (and thus wikipedia) is as much a result of a common (in this case "western" and 21st century) culture as art, or indeed rules in virtual worlds can be. The first positivistic encyclopedia's (Diderot, et al) had a very distinctively different content than a current day encyclopedia.

It's the same with virtual worlds, only on a different (micro) level. People can't judge whether something is "true" or "false" in a MUD, but following my above reasoning, neither can one do so objectively in an encyclopedia. A community can, however, make choices in the positive or negative sense as a product of their culture and their training. Just as people can be taught to accept certain facts about reality, they can be taught to think in a similar way about virtual things. Of course this implies that such a world would be subject to evolution all the time - as is our real world - but that makes it all the more interesting, no?

[Andrew Burton]

12.

> ...these are plenty of examples of clarity
> of vision and purpose.

And as of last week there are plenty of examples of the truth that in SL you can't have a Wiki-like design. For a Wiki to work, it has to be open for any and all to edit. However, it was just last week that two or three island owners left their land open and editable, and what happened? Some greifer flew by and destroyed three or fours sims worth of work. Wikis have hundreds of people constantly watching them, decentralized backups (decentralized in that each wiki has its own backup), and administrative access (the owner of the wiki can access it); SL have none of these. Griefing done to a SL island can only be fixed by the grace of a Linden.

(Sure, you could argue that island owners could keep their sims "off the map," but that defeats the purpose of a collaborative, publically accessible system.)

> Nathan's question was (I thought) about
> collaborative virtual worlds, which SL is
> clearly an example of.

I apologize for saying SL was not a collaborative system. My hands were cold and I was trying to be brief. What I meant to say was that while SL is collaborative, it's not designed for any useful collaboration. People have tried to build libraries, people have tried to build museums, and none of them have ever stuck. SL is a collaborative is you want a casino or a mall, but try to do anything remotely informative like a Wiki and you fail because SL is not designed for it.

Speaking from experience, when I tried to setup a newbie club in a sim close to the welcome area back in January, Lindens couldn't quote how many rules I was breaking fast enough. However, when someone runs a club that slows down half a sim, they don't do anything. This is, to quote an SL/RL friend I was talking to about this, "the complete antithesis of a Wiki mindset."

> You seemed to interpret "turn virtual
> worlds nto collaborative worlds,
> somehow, Wikipedia-like" meant that
> they needed to be connected to the web.
> While I think that you are correct in
> that being a great direction for
> digital worlds to go, I don't agree
> that is a requirement.

I was thinking about "smart signs" during that post. Signs and buildings with little wireless webservers built in so you can read about them from your handheld as you walk by them. I kept bringing up a web connection because I was thinking about how nice it is, in There, when I buy something, to hit a website that shows all sorts of models and fashions of an item. I was thinking about how nice it'd be in SL to be able to click on a building and read about it's history; to click on a person and read about what people thought of them. I kept bringing up the web interface because an embedded browser (like in There) or a browser around the 3D bits (as in Active Worlds) is much nicer than the Notecards in SL. Right now there are two way to view text in SL: Notecards and XyText. Neither of those is as fast loading or dynamic as a web browser.

Why am I still harping on the browser when you are clearly talking about a 3D virtual world, because all great 3D things start as 2D designs. If SL, There, or whatever is ever going to be fully collaborative, it needs a system for keeping notes and discussions and ideas. So far, having an embedable web browser to connect to a real, 2D wiki seems much simpler and robust than an in-world notecard and inventory system.

Being collaborative is one thing. I can collaborate with people in "City of Heroes" to fight bad guys. Being collaborative so that your work is useful to others, like a Wiki, is something else all together. I'm all for using 3D world to create new paradigms for human collaboration and thought, much like the web and wikis have done, but I think SL is a poor choice for doing it.

[Hellinar]

13.

Yitz Jacob >Nathan, a thought: wikipedia has one thing going for it (that coincidentally i think Second Life might have as well), namely it can be compared to the real world, or any other world for that matter, at least something reasonably objective.

The error-checking functionality of wikipedia comes from this feature. To have rss-like RCPatrols for a virtual world, you'd need to have a sort of objective framework or you would need a very simple TOS and that would be the only thing enforced by anyone and everyone. <

I’d argue that having one “true” target as the end result makes the software more complex. “Lets recreate the Star Wars universe” gives you more ways to fail than creating a more novel world. As SWG amply demonstrates. At the other end of world design from implementing a well known IP, there is the “emergent” world. There the prime task is to produce a rule set that produces an “interesting” world. Interesting is a much looser and more subjective target, and hence easier to achieve. A “bug” which makes the world more interesting is automatically reframed as a feature. That’s not true in a more targeted world.

A robust example of an emergent world system is the genetic system. The base rule set is not that complex, but produces a wide variety of life forms. Its something I am trying to replicate on a simpler scale. Beside being able to produce “interesting” results, the ruleset must also allow a selection process to identify “uninteresting” results. And making those selections must decrease the probability of further uninteresting results.

In my own gardening world example, this selection can obviously be handled by selective breeding. In this case, the Players are selecting for "Beauty". Given this vague and subjective target the Players can apply their collective intelligence to finding acceptable solutions. Unlike the Wikipedia example, there is no singular "correct" solution. One group can consider a flower "colorful and dramatic", and another classify the same flower as "overly gaudy and kitsch". Neither assessment is "right".

I’d rate the Wikipedia as a poor example to follow in collaborative world building. It inherently has a “One True Path” dynamic. To my mind, a system that allows “Many Paths up the Mountain” will produce a more interesting world. Its much better suited to collaborative design. And be inherently a lot less buggy. Already in my own explorations, misplaced code that produced interesting results has become part of the basic feature set. I’d hate to see such a software development approach applied to an Air Traffic control system. But in emergent world building, errors are, to some extent, your friend.

[Cory Ondrejka]

14.

Andrew> SL have none of these. Griefing done to a SL island can only be fixed by the grace of a Linden.

Agreed, that ease of undo in a physical universe is a very different proposition than in a text repository :-). That being said, land and object owners have a fairly rich set of permission choices. Also agree that content in SL is, in general, more ephemeral, although this is a function both of rapid evolution/selection pressure and the fact that people have to pay to support their creations. However, there are many examples of educational resources in world -- generally education related to SL, of course -- and that knowledge has definitely been growing and increasing over time. Of course, where appropriate, it has migrated into Wikis (such as http://badgeometry.com/wiki/HomePage).

[Andrew Burton]

15.

> Agreed, that ease of undo in a physical universe is
> a very different proposition than in a text
> repository :-).

SL is not physical, though. Every prim is simply a representation of an XML file. That being said, the ease of undo in SL should be as simple as a test repository.

> Of course, where appropriate, it has migrated into
> Wikis (such as http://badgeometry.com/wiki/HomePage).

I find it funny that your point to an off-site, user-owned codebank as opposed to the SL forums. (Should I mention that when SL deleted a years worth of code discussions, they didn't have a backup?) This makes my point exactly: useful, collaborative software needs to be owned by private people who want to use it for a purpose, not a company who wants IP rights on a platform. Second Life is an excellent sign, pointing in the right direction, but it's not the end of the journey.

[Cory Ondrejka]

16.

Andrew> That being said, the ease of undo in SL should be as simple as a test repository.

Sorry for continuing to take this so far OT, but I'm afraid that isn't the case, Andrew. If you wanted to undo the entire world, perhaps, but since local areas are entangled with other areas, it becomes a rather difficult problem. Simplest examples: you move your unique painting to your friends house and then rollback your home. Now you have two paintings. Or, you and your neighbor have worked together to build a physical structure. You rollback your part, causing your neighbor's portion to collapse.

Andrew> I find it funny that your point to an off-site, user-owned codebank as opposed to the SL forums.

I'm sorry that SL wasn't the right experience for you, but I think the fact that LSL expertise resides in outside forums is a sign of strength in very similar ways to the fact that our currency is traded on GOM or IGE. What could be cooler than attracting residents with this level of talent?

[Brask Mumei]

17.

One thing wikis have going for them is the ease of editting. The textformatting rules are as unobtrusive as possible - reading the raw text is as easy as reading the marked up text.

This has two powerful results:
1) It is easy for people to write content. The barrier to entry is as low as possible.
2) It is easy for people to correct content. One can easily undo anothers mistake. One can easily get a diff, as the raw code diff is *meaningfull*. (Compare and contrast with doing a binary diff of two .doc files, for example :>)

I really think the Virtual World equivalent of a Wiki would be the CVS world that was discussed earlier. A world where any part could be rolled back by any user, or rolled forward. Or editted.

One major difference between a VW and a Wiki would be the amount of griefing. Grief players often are the end result of a normal player who decides to play a different meta game as the raw game wasn't fun. A wiki with no base game is less likely to result in that mindset. One does not get bored reading the dictionary. One merely stops reading it :> It seems much harder for people to stop playing a "game" or "world", and thus idle hands become the devil's tools.

- Brask Mumei

[Cory Ondrejka]

18.

Brask> and thus idle hands become the devil's tools

Ah, lots of recompiles and long database queries today . . . anywho, that's a really good point.

[Andrew Burton]

19.

> Sorry for continuing to take this so far OT, but I'm
> afraid that isn't the case, Andrew. If you wanted to
> undo the entire world, perhaps, but since local
> areas are entangled with other areas, it becomes a
> rather difficult problem.

How is this OT; we're discussing the pros and cons or collaborative system, no? Anyway, without going into the details of handling lots of inter-linked text, I didn't mean to imply that it was a snap. However, it isn't physical. You don't need a real bulldozer to fix land and builds in SL. Somewhere there's a MySQL database, each prim (painting, house, etc.) has an individual key, and they are essentially text; with the proper tools (Perl, Python, etc.) and logs you can write algoriths to manage these.

If it's hard and if SL wants to be a Wikiworld, they need to develop these tools so users can do these repairs. Version control is a serious concern for Wikiworlds.

> What could be cooler than attracting residents with
> this level of talent?

I'll find GOM and IGE cool when you show me a Wiki (or chatroom, or OpenSource project) where the ability to buy into a project made it any better. Maybe I'm misunderstanding you, but I don't see GOM or IGE making any SL players into millionaires. However, I do see it allowing land barons to make the entry barrier for newbies steeper.

IGE and GOM, in my mind, are dangerous. They're essentially the same as a rating system on the Wikipedia where an eighty year old librarian would have editing authority over a twenty-something programmer because of her degree. She may know more about proper writing, but I'd rather read an entry on programming by a hacker than a librarian. IGE and GOM are more inclined to give people with money an advantage in SL than they are make money for creative people... That's what I've seen anyway.

Which is to say: what creativew talents have been attracted by GOM or IGE that SL wouldn't have attracted by itself. (Now we are off-topic.)

But really, that's another reason why I don't think SL is the proper tool for a Wikiworld of any kind. It's more inclined to be a virtual world, with an economy, with organizations, etc. than it is inclined to be an objective tool. (Is that closer to the topic?)

[Nathan Combs]

20.

One merely stops reading it :> It seems much harder for people to stop playing a "game" or "world", and thus idle hands become the devil's tools.

:) What if that were the game... those who seek to subvert the logical underpinnngs of the world... and those who wish to defend it... If one were to just flip Danyel's "cataclysmic" perspective - it would be a titanic evolutionary struggle (punctuated).

[Brask Mumei]

21.

Nathan> "What if that were the game... those who seek to subvert the logical underpinnngs of the world... and those who wish to defend it... "

It is always easier to burn bridges than build them. Thus, I would guess such a system would devolve into a bridge burning wasteland as everyone signs up to the "subvert" faction.

The big problem is that often the defenders have another goal: to *use* the logical underpinnings of the world. Whilst defending it, they can't use it, so are frustrated.

If both factions are the entire point of the game (there is no *using* the world), then there really is no subvert vs defend faction. It is just two (or many) subvert factions with different end goals in mind. The standard Wiki answer is to split the factions into seperate WikiWords. The PVP answer would be to turn the server into http://www.corewars.org/.

- Brask Mumei

[Hugh Perkins]

22.

> Personally, I tend to believe the most difficult scaling problems lie not in the domain of engineering and technology per se... but at the interfaces with, and the interactions within the player systems: the game world laws.

I tend to agree with this. Metaverse or not, Linden Labs can only afford to support one single core vision for the SecondLife GameLogic.

Solutions? Better men than I have tried to answer that. That said, perhaps making the gamelogic OpenSource so that anyone can modify it (something like a Wiki I guess?) could be a part of a solution. At any rate, that's what I'm betting on currently; obligatory plug: http://metaverse.sourceforge.net

Hugh

[ren]

23.

As we went off on an SL tangent – is not A Tail in the Desert the best example of this?

As I said the legal system is created by players then, unless there is a code veto, put in by the devs. Each current law has a pretty good history of being discussed pre and post implementation, now if the game world were linked with some kind of text resource this would be a neat wiki-like world, its gota be pretty wiki’ish right now, hasn’t it?

Or am I still barking up the wrong palm tree?

ATITD 2.0 Beta is running btw: see ATITD2

[Tobold]

24.

Tale, ren, tale, not tail. :)

A Tale in the Desert is collaborative, but the main collaborative effort is not to create something new. The civilization just follows the skill tree, which is given. The law system of ATITD is actually changing the game itself, so it would count, but the major activity in ATITD is chasing after preset goals, not much different from chasing after levels and skills in other games.

[Andrew Burton]

25.

ATID is also finite, which could disqualify it as a Wikiworld. I've not played ATID, but it seems that a Wikiworld (a collaborative, 3D system to provide some ongooing use) would have to be completely non-linear.

[Nathan Combs]

26.

A quick attempt at some sort of summary.

Hinted at in a number of places in this dicussion, are what appears to me the "classes" of wiki-worlds. From the strongest to weakest:

1.) the "wiki is the game" e.g. hypothetical game where there are "those who seek to subvert the logical underpinnngs of the world... and those who wish to defend it" (above)

2.) the wiki is the specification of the game world - the knowledge-base, where specification implies implementation of the game world.

3.) the wiki is a description of the game world.

Def: "wiki" = the "full expressive potential" of a wikipedia.

From the discussion above, SL might then be an example of a weak form of (2.): its "wiki representation" is not as expressive as that of the wikipedia, for example.

ATID might then be a weak for of (3.) insofar as best I know (correct me) the discussions don't necessarily imply the implementation (though developers may try, they interpret).

[ren]

27.

Tobold > Tale, ren, tale, not tail. :)

/hits head on keyboard (again)

[higbee]

28.

Not to open a can of worms back up, but I wish to argue some of Andrew's points about SL specifically. I do own my own sim. 2 in fact. My wife owns one of them. We were struck by the griefers that rendered our sim to water. Yes we did have to requisition Linden help in this fashion, however the response was infintely more timely than in the past. Additionally many of the USER ADMINISTRATION tools, or Linden "abilities" as it were, have been given to us as recently as yesterday. Managing our sims to the fullest. Now I will happily agree that SL has a long way to go before it has the openness or interactivity of a true Wiki, but I think each update following will see more and more administration of our sims until we truly own and administrate them.

With GOM and IGE, though they do encourage a "devilish" type of player, without them I would have a lot more trouble paying for my islands, so they do serve a very real and important purpose.

I think you will see the private islands become more Wiki-like at a faster pace because they are inherently more collaborative. There will always be some sort of stranglehold on the mainland, but without this there would be utter chaos as described in the theories above. HOWEVER, and most importantly, it is a very simple teleportation gateway to get to your favorite wiki-like place.

I think however there is one extremely important note that no one has touched on: The Wikipedia is for geeks like you and me. Period. You will not find John Q Public hanging out on the Wikipedia. Cmon. You will however find John Q Thereian hanging out in a PVP private island playing some game that is not available anywhere else.

It's evolution we are talking about. The internet's first couple years sure weren't very wiki-friendly. But look now. SL's evolution is ten times that speed, and if you don't agree you haven't spent enough time there lol

[Andrew Burton]

29.

I'm going to agree and say all of Higbee's points are technically true. Without getting very, VERY anal on my part he's correct. There is one point I want to contend, however:

> The Wikipedia is for geeks like you and me. Period.
> You will not find John Q Public hanging out on the
> Wikipedia. Cmon.

The same thing that was said about the personal computer twenty years ago, video games fifteen years ago, the web ten years ago, and cellphones as little as five years ago is being said about blogs, wikis, and virtual worlds now. Yes, wikis are just for geeks now, but they are being used by a LOT of geeks.

John Q. Thereian playing a SL game is a far cry from John Q. Public even knowing what SL or There is. I'll kind of give you an example that just came to mind; Nate, here's your picture of a Wikiworld: the army's implementation of There.

It's collaborative, it's dedicated to a purpose, it's useful, and it's 3D. It's also being used by John Q. Enlistee, who may have never played an online game before in his or her life. Does that fit the bill?

What was I saying? Who cares. I answered your question Nate: a Wikiworld would look like what the Army is doing, which looks a lot like There.

[higbee]

30.

Good point. touche =) However I do believe that SL is moving in this direction. I do not necessarily believe that it is the end all Metaverse for which we are all searching, but at the rate that it is evolving I think it is extremely possible... Usefulness is garnered by the user. If it is truly to be a World, then it must encompass many different focal points. If a wikiworld has only 1 purpose, then it limits the usefulness because there are less users. In this scenario there would need to be a WikiUniverse, which would be a gateway to them all. Is this to be your web browser? Then we are all still crippled by a program.

Let me throw this out: PHP for example is an Opensource language that is extremely expansive. However we all must bow to the version changes before we implement our own changes. Does this not echo what happens in a wiki? If that is true, then if the response to the user's desire is fast enough, any program could be bastardized into being called open-source. I think everyone would agree this is silly. However, with this comes an important point discussed earlier. If there isn't a well-spring from which it all comes, we are more likely to divide further because we all have different styles and likenesses. Having some structure is unfortunate but desperately important; i.e. a small TOS, or rules, etc.

This is what occurs with "There" now; it's just staffed by several personnel 24 hours a day doing dynamic scripting and hardwiring. There is only so much open-source that can be created with any program. Consequently, I do not see how this makes "There" more of a Wikiworld than SecondLife.

SL provides a platform for creating internet 3D building technology(linden-side) and content(client-side) that is being tested on a near-Wiki scale (due to speed of evolution and response of creators). The internal cry is so great for more flexibility it is only a matter of time I believe. I will happily put in 2-5 years of concealed testing to be on top end of users with a vast inventory of content when it becomes open-source =)We shall see heh

*Sorry... don't post often so I say a lot when I do lol

[Andrew Burton]

31.

> If a wikiworld has only 1 purpose, then it limits
> the usefulness because there are less users.

By this logic, a web page about dogs is worthless because it doesn't cater the interests of cat owners. And you're logic would be right if not for the hyperlink. However, because of the link, the dog page can be linked to a page about pets in general, which could link to a cats page.

> In this scenario there would need to be a WikiUniverse,
> which would be a gateway to them all. Is this to be
> your web browser? Then we are all still crippled by a
> program.

We already have a WikiUniverse crippled by the browser, it's called the World Wide Web. This may be a bit tangental, but this is why we need some company (Linden Lab?) to sell shrink wrapped VW software. Then you can build this 3D, Wiki universe; you're not limited to one over-generalized world owned by one company.

IMVU is exciting because it's a virtual world that actually builds atop existing software, and doesn't try to replace it. I'd wager to bet that IMVU, unlike There or SL, will have embedded hyperlinks; it'll have to if it wants to compete against IM's. (Apologies if this paragraph is a bit off-topic, but IMVU exemplifies the reason I was praising There and AW originally: it doesn't replace existing tech, it builds atop them.)

> However we all must bow to the version changes
> before we implement our own changes.

No, you don't, because PHP speaks via the HTTP protocol. Now if HTTP changed, it'd be a different story. However, I can change anything I want to my installation of PHP, and as long as it both works on my computer and speaks through HTTP, I can ignore any new versions.

> Having some structure is unfortunate but
> desperately important; i.e. a small TOS, or
> rules, etc.

Don't mistake structure and protocol. The internet has many, many protocols; Second Life alone uses at least two: TCP and UDP. However, the internet as a whole has no Terms of Service. You can't be banned from the internet. (This is off topic, sorry.)

> Consequently, I do not see how this makes
> "There" more of a Wikiworld than SecondLife.

The Army uses There's technology as a collaborative system to instruct people, answer questions, and record data for future use. Second Life has malls that people work together to build. What about that makes Second Life more of a Wikiworld?

> SL provides a platform for creating internet
> 3D building technology(linden-side) and
> content(client-side) that is being tested on
> a near-Wiki scale (due to speed of evolution
> and response of creators).

In all fairness, and I'll take responsibility if I'm the cause of this misconception, comparing Second Life's development to that of the Wiki's is another case of apples and oranges. The Wiki, Kwiki, cliki, and others are all different programs that use the same model, are of the same genre. Second Life is one program, it's not its own genre. That said, I'd wager even more money that more people use Wikis than they do Second Life.

For the record, I rebutted Cory's first post because I've never seen people do anything remotely Wiki-like in Second Life. On the other hand, I have seen the U. S. Army create a collaborative education system in there. Cory said SL was wiki-like, I said it wasn't. However, can we please get back to the main post: what would a Wikiworld be like and what kind of technology would it require? That's the exciting stuff.

(If you want to argue Second Life with me, IM "Jarod Godel" in-world.)

[higbee]

32.

You got it =) Good arguments!

[Hellinar]

33.

Higbee>Having some structure is unfortunate but desperately important; i.e. a small TOS, or rules, etc.

An external set of rules is vital if you are trying to get your collaborative world to converge on a particular correct solution. In the Army’s case, an effective way to train recruits. In the Wikipedia’s case, good information about the topics listed. But if you let go of the need to have a correct solution, then you can use an emergent world design. That is, a world that starts with a set of rules, and a means for exploring the resultant world possibilities. The collaborative act then becomes a question of the community deciding which possibilities are worth preserving.

I’d argue that that such a world design is open to more people than one that converges on a particular use. In the case of a world with a particular use in mind, you can only collaborate with people who agree on that use. In a world based on exploring the possibilities of a ruleset, many different communities can focus on aspect they consider the most interesting. Thus the dog breeding community has explored possibilities of the dog development ruleset over many centuries. And have come up with dogs for a wide variety of uses or esthetic sensibilities.

Mind you, any practical world would require some external rules and a means of detecting outlaws. But in a sense, you give people more freedom by embedding rules in the structure of the world, than by mandating a particular goal or use for the world. If the rules are complex, and allow emergent behaviour, then Players can reach goals never imagined by the world creators. This possibility of unexpected discovery represents a particular kind of freedom to me.

[higbee]

34.

> I’d argue that that such a world design is open
> to more people than one that converges on a
> particular use.

This is actually the most important point I was really trying to make Hellinar. Thank you for clarifying. I was actually challenging the definition of a "wikiworld" more than anything. I was also searching for reasons why a "wikiworld" would be superior to a world like SecondLife. It is easy enough to open my browser while I am in SecondLife. I can take classes on PHP in-world, while I am surfing the internet for facts and functions. I feel like this is just like using multiple browser windows. My computer performance is a little slower but I am using something graphic intensive, so that doesn't really surprise me.

I think everybody here agrees that a joining of powers is what's most important for true expansion and limitlessness. One thing that Andrew and I definitely do agree on... SecondLife will never win the battle of the Metaverse unless Linden Labs makes it OpenSource.

[AFFA]

35.

I wonder if Wikis (or games based on this method) would be subject to "memetic" attacks.

I.e. in a future where people rely more on Wikis than traditional sources, would it be possible to spread misinformation more rapidly (such as email spreading urban legends and scams more rapidly)? If bad data is on the Wiki long enough to convince a majority of users, what's to stop it from becomming "fact" if the most trusted sources are other Wikis who have the same corrupt data?

[Brask Mumei]

36.

Your point is interesting AFFA. However, please keep in mind that we already live in a world. Even worse, you don't have to convince the majority of users, you only have to convince a very small minority.

A lot of academia involves shovelling around references and footnotes, no few of which are wrong or taken radically out of context. It is not uncommon to try and track down a bibliographical reference that showed up in a lot of papers, to find out that the reference is incorrect. The *first* paper may be excused as a type - the others clearly just cut and paste.

At least with a wiki one has the ability to try and rectify these issues when discovered. So, I could go to the wikis and fix the citation with the appropriate note explaining why it was wrong. It is rather more difficult to go to print journals and fix citations :> I guess, however, in cases like this the most trusted source isn't the other wikis, but is the original cited material itself.

I'm not sure where the most trusted source would ever be wikis. If it is a physical fact, the trusted source would be the initial experimental reports. (Or, repeating the experiment oneself, a la "A ducks quack never echoes", http://www.straightdope.com/classics/a5_071.html) This leaves subjective results, which you will likely always find various wikis contradicting one another. Much like you can find various learned journals contradicting one another.

The thing I have learned about memes is that they are highly overrated :> Humans have been living with memes ever since we learned to talk (if not earlier). Thus, it is safe to assume we have remarkably robust immune systems for dealing with memes. If we didn't, humanity would have long since been wiped out by one meme or another. So, a "bad meme" will have greater problems spreading than a "good meme" or "harmless meme", as the people it infects will have greater difficulty infecting others. A truly effective memetic attack would have to target the long term - set up some simple false fact as truth today which in 50 years can be used as a lynch pin for your attempt at global domination. "Free Market" may very well turn out to be such a meme. :>

- Brask Mumei