SOP II: IP Panel

Sop2I'm posting live notes from the intellectual property panel here at the State of Play II. My first experience with live blogging, so sorry for the messiness. These will be cleaned up later & links will be added.

David Johnson kicks things off with a hypothetical world where player-created objects can have metadata tags that indicate the player-designated virtual property rights. How would that work?

Yochai Benkler leads off with an argument that the key issue is the directive of the game design. E.g., is the design oreinted toward creating revenues? That would provide one model. But he's particularly interested in collaborative creativity, exemplified by systems like Wikipedia, and he's asking how those models can be applied to virtual worlds. He says that he's pushing toward a model where property rights are *not* asserted, where the results of creation are understood as a shared collaborative output (not property).

Cory responds next. He suggests that as a result of changes in Second Life IP policy (toward player ownership), Second Life has expanded by a factor of 15 since the last State of Play. However, he notes that open source has not been particular productive at creating games. Millions of dollars must be spent over many years to product complex artifacts and art assets. Some of this is not fun. One problem, he says, in creating a commons is that much content in an open world would come from the outside world--so the EULA is needed to get the content to flow in.

Cory points out the significant difference between a designer-created sword "found" by the player, and true player-created and player-coded sword, which is more closely attached to the player.

He talks about the difference between 2D and older virtual worlds (presumably textual), and states that there is a much different dynamic generated by 3D worlds.

Question for Ted about the EULA. Ted says he looks at the end-state equilibrium. If we analogize the competition of virtual worlds with the competition of nations, it seems that the Western model has advanced the premise that capitalist, property-centric systems work best. But now we have the possibility of experimenting. Second Life, by awarding property rights, seems to validate the Western model, but only because it creates growth. The tense question, in terms of property rights, is whether something else may happen in the virtual world space that points to an alternative model? Is it possible to create a space that has no property, that has no legally recognized value? He wonders whether virtual assets decay.

Incentives -- is it possible to create a game environment where the production of value is enabled by the game mechanics instead of standard property incentives? Is it possible to engage this technology to create a new and different social, political, and economic structure?

David Johnson's Question: what is best for me as a game owner -- how should I set the permissions of IP?

David Post: The answer is "What is better?" We all have different ideas about which kinds of rules will better foster creativity -- we can argue about what's best. We can all argue why socialism, or user-selected rights, or coded rights will produce optimal outcomes. Perhaps you can get the objects to inherently (in a coded way) carry their rights with them. Experimentation is the way to proceed -- let everyone try out all these different models. Let a thousand flowers bloom.

Yochai goes up to the podium and starts surfing on Yahoo -- not clear what he's up to.

Cory -- DRM is a waste of time. If you're working on DRM, get a new job. Comparing the technology to the law--technology is not enough to solve the problems we're encountering. Second Life has been working with Creative Commons to set up a system for intellectual property within the game. Compares GPL to a viral propogation -- CC doesn't sit well with the CC system. Goal of Second Life is to maximize creativity, which is why you find a turn to intellectual property law.

Production as fun. Average person spends 35% of time in Second Life. User-created content is tough. You need IP in order to make people labor and create valuable content.

Yochai shows what he was looking for -- how Apache-run servers are beating Microft-run servers. The Open Directory list beats the Yahoo list. Wikipedia produces great stuff. The point is that you don't need IP incentives to produce quality content. Open, collaborative products can produce not just good content, but superior content. The overarching idea: individuals can come together and make small contributions that produce *more.* The growth over the last year is not about IP rights -- it is about Second Life giving affordances that allow players to accomplish what they want. It is not about incentives--we don't know the incentives.

Johnson: Do we need a subset of people in control?

Post asks Cory -- why do you think the grant of rights really created growth?

Cory: If you used Second Life over the last year, you wouldn't be asking these question. The quality is so much better -- a year ago we had to go hunting for the best content. Today the quality of the content has improved. The scale of people working on projects has increased. You see large scale (20+ people) projects.

Post: Well, okay -- but is it IP that changed the behavior?

Cory: Yes. And notes that real world earnings correlate to good content.

Ted: You say you're trying to maximize creativity, but I think you're really attempting to maximize profits. Wikipedia is counter-intuitive. Is it possible that the Internet has lowered the cost of production to the extent that our free-riding problems have been solved. But the question is -- how long will Wikipedia be on top? How old is it? Agrees that the big question, as Yochai identified, is what rules are best? The best answer is to let a thousand flowers bloom. The IRS won't let a thousand flowers bloom.

[Jokes about IRS and TN beards]

Cory: Tensions with the users over IP issues. Most of the concerns about copying were not about property, but about attribution concerns.

Johnson: Is there an issue in applying copyright to VWs -- the fixation requirement that doesn't seem to fit.

Yochai: [Something about attribution in SL -- wasn't clear about the facts of this, so I got lost]
I have no problems with institutional arrangements, but I'm concerned with spillover effects. Asks Peter Ludlow about his intentions with licensing for his Second Life project. Yochai -- opposed to applying IP to Second Life because of the depressing effects on derivative creativity.

Johnson: Q re inspiration for creativity vs. copying of objects

Yochai: If you can distinguish in a robust way between derviative creativity vs. copying -- in the digital environment, most people copy in the course of creating new works. Because of this, you need to be cautious.

Johnson: controlling access to spaces? Performance spaces?

Yochai: that's fine.

Ted: Strong walls are good. You can seal things off, because sealing them off allows them to bloom.

David: Within your wall, Ted, one of the problems with socialism is that you need to prevent things happening within the wall. Within the wall, you need to prevent exchanges, competition with the source of production, etc.

Ted: The EULA is already creating those terrible restrictions. Gives examples of EA EULA.

Ted: What about building one of these virtual worlds in the non-profit center. But when we build the non-profit worlds, we run into ethical issues -- do we have to disclose that we are reserachers?

Cory: Imposing costs and barriers on obvious property is considered rude. But it's not considered rude when it is offshore. Say SL has problems with people who are doing studies and not disclosing this to SL users.

Yochai: Do we think collaborative production will be better or worse if we have password protection for certain subgroups? Perhaps not everyone should have the same rights over the space?

--------------------
Questions from audience:

Peter Ludlow: You see a complaint that we can't see the scripts anymore, we just get eye candy. Perhaps the scripting is being stifled by the IP policy?

Cory: No -- the scripts are better than they were. Economics. Goal is to find the right trade-offs. We want to figure out how to bring a right to tinker back into a digital space.

David: Thinks that Yochai should build the world he wants. But maybe Second Life should have a *Benklerville* where a separate set of rules apply?

Cory: You could do that.

Richard Bartle: You can create sub-worlds within a particular world. People want to create their own atomic world which is theirs. When we're talking about the law and virtual worlds, some are little worlds that are separate and some are little embedded worlds within worlds. Responding to Yochai's point: we can all be excellent authors individually, and when we collaborate we might not be as good. (For instance, if we all voted on a chess game, we could not beat a grand master.)

[missed the question -- about EULAs]

Eric Zimmerman: It is interesting that this seems a lot like games design. Rules serve the function of maximizing fun, profit, reward players, policy (other ideas). A lot of the disagreements are about conflicting philosophies of design decisions. Says that it is interesting with Ted -- scarcity drives player experience. Says these aren't nations -- they're more like health clubs.

[lost track here a bit]


Comments on SOP II: IP Panel:

Timothy Burke says:

I sort of wish we could get away from the constant low-level replaying of ur-capitalist v. socialist tropes in discussions of the future of intellectual property rights, the setting up of a individualist/revenue-seeking structure versus a collectivized, collaborative, communalized, non-propertied structure. This is assuming I am grokking Benkler's observations correctly here, they sound consistent with some of the things he said last year. Can we just throw that script out now? Surely there are not just excluded middles, but entirely different kinds of ways to imagine the distribution of rights, authorities, social architectures, in virtual (and real) worlds.

For one, I think one could borrow a lot from network theory to help out--networks may be collaborative structures that resist singular claims of ownership, but they also have nodes of greater and lesser concentration of activity, rights, wealth, power, what have you. The metaphor (or maybe even empirical reality) of networks strikes me as one (and not the only) way to reimagine property rights and other rights in these spaces that doesn't require replaying the exhausted battle of pure communalism vs. total individualism.

(Man, I wish I was there...)

Posted Oct 29, 2004 10:00:43 AM | link

ren says:

Post asks Cory -- why do you think the grant of rights really created growth?

Cory: If you used Second Life over the last year, you wouldn't be asking these question. The quality is so much better -- a year ago we had to go hunting for the best content. Today the quality of the content has improved. The scale of people working on projects has increased. You see large scale (20+ people) projects.

I wonder about Cory’s assertion. Specifically whether there has been an exponential grown in content creation or a step change. The point being that an exponential change could be explained by a network effect whereas a sep change could be put down to the policy shift.

Posted Oct 29, 2004 10:02:45 AM | link

Andrew Burton says:

I'm replying to this as I read through it, which means it's off the top of my head.

However, [Cory] notes that open source has not been particular productive at creating games.

Cory's wrong. Second Life uses C/C++, Perl, Python, OpenGL, MySQL, and Linux; how much more Open Source can you get when creating a game?

Is it possible to create a space that has no property, that has no legally recognized value?

City of Heroes, as of yet, has nothing ownable aside from your costume. It has Influence, sure, but there's no way, that I've seen, you can leverage Influence the same way you can leverage Lindens, Pyreal, or Dollars.

DRM is a waste of time.

I'm lost, which Cory is this? Doctorow? ;)

User-created content is tough. You need IP in order to make people labor and create valuable content.

This is what I don't grok -- and maybe I'm going off on Second Life here, but it applies to all future games. Why don't you see this kind of mentality popping up in other games? When I play City of Heroes as a Iaxibot, I'm not overly concerned that some other group is going to come along, start teleporting people, and take away my business. I'm playing a game, having fun; if someone else wants to do what I do, they're welcome. What I see more of, though, instead of competition is people recognizing the Taxibot brand, saying it's a cool idea, and joining up. No intelelctual theft there, just viral marketing.

Going back to what you said, though, why do games need defenses for people who spend all this time "laboring" to create something? In the OSS community, if something someone creates is cool, they'll get credit, praise, often times help, and maybe even royalties. What makes virtual worlds so diferent than that? (I also want to make a snarky remark about laboring to create something in Second Life being kind of silly, but here's not the place.)

I'll give you my favorite example. In Second Life there are these gaudy contraptions known as "Dance'o'Matics" that let people syncronize their steps. I know who created them, one "Kris Ritter," because I keep up with her LiveJournal. I also know that she's complained about people stealing her idea and building their own DoM clones. The social network of Second Life is setup so I know who created the original. What more IP rights would she need, aside from something like the DRM, which is already stated as a bad idea.

The growth over the last year is not about IP rights -- it is about Second Life giving affordances that allow players to accomplish what they want.

I want to have Yochai's children! Thank you for saying that!

But the question is -- how long will Wikipedia be on top?

Ten years ago people asked the same thing about the web: how long will this be around, why would I want to have my own web page? Three years ago it was the blog: what good is this, it's just people ranting? I'm sure someone said, way back in 1903: don't these Wright Brothers have bicycles to fix? Wikipedia is just an evolution in collaborative tools. The Wiki may not be around, or on top, forever, but something like it, from it, or because of it always will be.

Most of the concerns about copying were not about property, but about attribution concerns.

We're working on that.

Posted Oct 29, 2004 10:35:17 AM | link

Kirk Job-Sluder says:

I have been thinking a bit about this this week, after a tremendous burst of creativity, has resulted in an RSI relapse, forcing the back to the single application that I have not been able to find a viable open source replacement for, speech recognition. As much as I like open source, I think that it falls flat with developing some application types. If you look at the most successful open source applications, they started small and grew through incremental patches submitted by multiple independent users. They also started small and grew slowly over time.

The basic problem that open source software faces is that the open source models that the most successful so far don't apply well to applications where you need a large amount of research and development upfront. Because quite a bit of rich media are produced in an extremely short period of time, and because skeletons of rich media are not especially engaging or fun, because a high degree of stylistic and content consistency needs to be maintained over the entire product, it makes a lot of sense that open source models have not been extremely effective in this area.

The fact that a game system uses software such as Python to develop its content does not necessarily mean that the content itself will be free or open source.

It bothers me a little bit that wikipedia is hailed as such a radically new method of producing content. The Oxford English dictionary and Encyclopaedia Britannica were collaborative efforts. In fact, the revolutionary idea of the encyclopedia was of creating a collaborative work that encapsulated the knowledge of the day rather than publishing the expanded writings of the single author. Wikipedia is just the next evolutionary step.

Posted Oct 29, 2004 11:23:29 AM | link

Ola Fosheim Grøstad says:

The main issue isn't IP rights to it's fullest extent, but the psychological effect: "a sense of authorship". Wikipedia retains this in the same way that grafitti retains it: "Yes, that is the text I wrote." Linux retains it: "I wrote subsystem X".

Most bottom-up designs can retain a sense of authorship because the creator can do whatever he feels like doing. He makes the decisions of how it fits in with the rest of the system and can feel proud of the piece he made all by himself.

Top-down designs become much more problematic for voluntary collaborate development strategies.
I don't believe you can easily get a top-down project to work well without a clear system that makes sure that you are being credited for your work. After all, you are just doing what you are being told to do. I might be willing to do that if I get some recognition for it.

Anyway, being credited as the author is a right that cannot be given up by contracts in my country. So the option of deliberately not crediting people for their works seem to be limited to only some countries.

And of course, Open Source games are widespread. You don't get the 120 man years projects, except perhaps some MUDs that allow bottom-up or stratified designs (areas). The problem is coherency. People want to make a contribution which they find interesting, not just follow orders by some art director.

Posted Oct 29, 2004 12:18:08 PM | link

Andrew Burton says:

If you look at the most successful open source applications, they started small and grew through incremental patches submitted by multiple independent users. They also started small and grew slowly over time.

As opposed to what? This is even the business model that Philip Rosedale points to when talking about Second Life: emergence.

The basic problem that open source software faces is that the open source models that the most successful so far don't apply well to applications where you need a large amount of research and development upfront.

Such as? I'm willing to agree with you at the moment, because you don't see as many things like OSS virtual worlds, OSS voice synthesizers, etc. -- they exist, but aren't abundant. However, as you said, OSS tends to start small and build larger. I can see OSS voice synthesizers and virtual worlds becoming mainstream someday when the tools reach a certain point.

Posted Oct 29, 2004 12:24:19 PM | link

Cory Ondrejka says:

Andrew> Cory's wrong. Second Life uses C/C++, Perl, Python, OpenGL, MySQL, and Linux; how much more Open Source can you get when creating a game?

SL uses open source tools but has developed a world using closed source methodologies. The world of Second Life also leverages much of what is good with open source wrt user created content. However, open source methodologies have not been successful at creating games.

Posted Oct 29, 2004 1:39:10 PM | link

Flatfingers says:

I'm glad Greg provided these notes -- thanks! (I wish I could have attended, too. Copenhagen was more interesting than I expected; it would have been fun to hear more of what some of TN's "beardy" folks are thinking.)

WRT the overall topic: I wish the question had been more clearly framed. There seems to have been a lot of confusion over whether the panel were discussing the foundations of IP in virtual worlds, or the virtues of specific IP systems.

People enjoy debating the latter (as I'll do in a second), but as noted, it's been done. Since the origin of "let a thousand flowers bloom" bugs me, I'll agree by saying that we should let the market decide. (As, I think, it already has done. Go, invisible hand!)

It might have been more interesting/worthwhile to take a more foundational focus: what can designers do to help would-be developers experiment with different IP systems? Are there design tools or collaborative agreements that can be constructed that would make it easier for developers to try out alternative IP systems?

I wish the panel had gone down this somewhat less traveled road.

...

And because I can't resist, either:

"Yochai -- opposed to applying IP to Second Life because of the depressing effects on derivative creativity"

Where does this objection come from? I'm not in SL (because where I live I can't get the broadband connection that SL requires), but the rebuttal on economic grounds seems reasonably obvious.

A robust IP rights system generates more primary creativity than weaker systems because it does a good job of guaranteeing that creative effort of value to the community will be fairly rewarded. This results in more creative effort being applied than in other systems. In turn, when others within the system gain expanded access to creative products (because more such products are being created), they become more likely to create products themselves; first, because they also see their creative property rights being protected, and second, because seeing what someone else has done sparks additional creative effort. ("Hey, I could do a better job than that!")

Greater primary creativity thus stimulates additional derivative creativity. So if SL's strong IP rights mechanism does a good job of promoting primary creativity, then it should see greater derivative creativity as well, not less.

The only danger would be if somehow a few people could create a majority of the potentially creatable objects in the SL universe. But this wouldn't seem to be the case. (It'll be even less likely with extensions to SL's creative language/tools.)

If you want to encourage people to undertake complex acts of creation (which contributing some text to Wikipedia isn't), you need to assure them that their expenditure of time and energy will be rewarded in some desirable way. A secure IP system provides such assurance, and I'm glad (but not surprised) that it seems to be working well for SL and its subscribers.

--Flatfingers

Posted Oct 29, 2004 3:00:56 PM | link

Andrew Burton says:

SL uses open source tools but has developed a world using closed source methodologies. The world of Second Life also leverages much of what is good with open source wrt user created content. However, open source methodologies have not been successful at creating games.

Whatever
you say, Cory. If that's the case, why does Philip say there is "no reason why SL tech shouldn't be open?"

I can see why it might take closed source methods to get the ball rolling, but if OSS is not "open source has not been particular [sic] productive at creating games," why does Philip think "[Second Life is] already very aligned to open source?" Maybe I'm confused, but the two statements sounds opposing.

Posted Oct 29, 2004 3:25:20 PM | link

Ola Fosheim Grøstad says:

Cory Ondrejka> However, open source methodologies have not been successful at creating games.

nethack, DIKU, LPMud, XPilot, Crossfire, all-kinds-of-board-games, emulators, Doom clones etc...

The incentive for making new single users games is low though as you can always get them for free on the sharing networks.

Posted Oct 29, 2004 4:47:29 PM | link

Kirk Job-Sluder says:

Flatfingers: Greater primary creativity thus stimulates additional derivative creativity. So if SL's strong IP rights mechanism does a good job of promoting primary creativity, then it should see greater derivative creativity as well, not less.

The question of what system works better for compensating creators remains to be seen. Corey Doctorow argues that his royalties have increased after releasing his books under a Creative Common license because as a writer who is not a household name, marketing is a major issue for him. For those people who like hard copies of novels, a professionally bound copy is almost as cheap as a print-out. (Especially with ink prices.)

But one of the big problems with existing ideas of IP is that there there is a lot of hostility out there in regards to secondary creativity. Independent mods, like fanfic, only exist by the benevolent grace of the publisher. Granted, quite a few publishers see mods and fanfic as a good thing in terms of marketing, but if any game publisher wanted to pull a Disney and shut down all of the modding sites, there is nothing to stop them.

Posted Oct 29, 2004 7:55:05 PM | link

Kirk Job-Sluder says:

Andrew Burton: Such as? I'm willing to agree with you at the moment, because you don't see as many things like OSS virtual worlds, OSS voice synthesizers, etc. -- they exist, but aren't abundant. However, as you said, OSS tends to start small and build larger. I can see OSS voice synthesizers and virtual worlds becoming mainstream someday when the tools reach a certain point.

Well, one example close to my heart happens to be continuous speech recognition. There is CMU Sphinx but it is the outgrowth of a DARPA-funded project. It's one of those applications that's hard to get right, and does not work well in incremental stages. (Unlike say, Linux, which started as a minimal kernel and added drivers and features.) It also looks like the next generation GUI systems (looking glass and longhorn) are coming primarily from big development processes rather than grass roots development processes.

Things might change now that Big Blue and Sun are on the bandwagon. But even there, I'm seeing more of a loss-leader strategy at work. Eclipse and OpenOffice are carrots leading organizations to adopt the platform contracts. Some of the examples cited such as idsoftware also seem to be painless loss leaders. The cash cow for QuakeIII has largely came and went.

I certainly feel that we are going to see a trickle-down effect as innovations are reverse-engineered or released under OSS licenses. But I don't think that OSS development paradigms have effectively dealt with the problem of types of development that require heavy investment just to get started.

Posted Oct 29, 2004 8:14:55 PM | link

Cory Ondrejka says:

I'm not sure how citing a Microsoft marketing campaign defends open source games. id releases their tech, not their content, and still has many controls over the products of using their tools. Text game examples don't work either, since my point is exactly that artistic development is where open source fails. Open source succeeds best on projects that are modular and that benefit from small, independent improvements. Games rarely fit those constraints.

However, in the long run -- as Philip and I have both said many times -- there is enormous, positive pressure to make SL open source. It isn't something that we can do quickly, but I'm pretty sure that we all agree in the open source direction for SL, since SL itself -- as opposed to the content built within it -- can be made modular enough and will gain from small, independent improvements.

Posted Nov 8, 2004 10:50:24 AM | link

Brask Mumei says:

"The incentive for making new single users games is low though as you can always get them for free on the sharing networks."

I can't really think of a single time in the history of computing when that wasn't true.

As for OSS game development, I agree strongly with the comments about the importance of artistic direction. The reason why you have to pay artists is because you have to convince them to work under *your* vision.

While many games are developed with OSS licenses, most successful ones nonetheless were developed by an individual, or, less often, a small team. This is because that is the largeset group that could be motivated behind a coherent vision. (Nethack, many would argue, lacks a coherent vision precisely because of its relatively ready acceptance of user generated patches)

The question is not one of open source or closed source. The question is one of "Do I pay my artists/programmers a salary or not?" So long as you are willing to pay the artists wages, you can get them to sacrifice their own vision in exchange for your vision. The licensing of the resulting content only matters when it comes to trying to figure out how to recover the money spent on wages.

Indeed, the requirement of a vision can actually be a good reason for a solitary developer to avoid the OSS licenses. One does not have to worry about being tempted by patches that distort one's vision.

- Brask Mumei

Posted Nov 8, 2004 7:43:02 PM | link