« Multiple Server Hack: Portly NPC Appears Mysteriously in Several Virtual Worlds; Distributes Gifts, Tubers | Main | Howard Dean: The Videogame »

Dec 27, 2003

Comments

1.

I want to set aside the wrinkle that makes reputation-tinkering part of the game mechanics. Not that it's a bad wrinkle, it's just that it makes my head explode when I think about it.

So, just on the issue of player data and its potential uses and abuses: I think this is an area where EULAs and ToS could be written so as to enhance their robustness to external interference. If the EULA says that a) we will indeed collect lots of data on players and allow the software to use that data in ways that we assert will enhance everyone's experiences, and b) players will have explicit mechanisms (i.e. verification checks) to make us responsible for the data's quality and distribution, then I think the world-owner has taken a major step in the direction of accountability and management integrity. I think privacy-related statutes would have much less merit if the ToS had these kinds of privacy protection provisions.

And there's the added benefit that, in collecting the data, the company has some weapons to go after the behaviors that are messing up Sims Online. Example: I took text from Lady Julianna's writing at the Alphaville BDSM site
Black Rose Castle (found through Ludlow's Alphaville Herald BDSM interview) and ran it through a gender-detection engine. OK, so Lady Julianna is a man. Doesn't really mean anything here, but it opens interesting potentials and problems.

World owners have access to an incredibly rich data file on users. Statistical procedures allow the owners to have a very accurate fix on user characteristics. If a user says he's 28 but her behavior says she's actually more like 13, her account can get a flag that prevents access to inappropriate content.

Such discrimination would be a very good thing, it seems to me, but there have to be safeguards too. We'd have to allow a fact-checking scheme like the one Richard proposes. The flag would have to be an anonymous and invisible marker. Important distinctions would have to be made only with extremely high statistical accuracy - "Adult Site; No One Allowed In Unless Their Prior Probability of Being 18 or Above Exceeds 99.9 Percent!"

Companies already use these methods to send us junk mail. We rightfully fear their use by government. But as with all technologies, there are good uses too. Managing content availability (with proper respect for community norms and individual privacy protections) might be one of them.

2.

A blacklist? You think data-protection laws can help you from ending on somebody's blacklist?

I would say that reputation generating software would be permitted, under Norwegian law, as long as it is connected to the character and not to the IP address. Being connected to the character means it is IC, while being connected to the IP address means it is OOC. Being connected to the character means it can't be crossed with other information which belongs in other spheres of the life of the player, while connecting the reputation to the IP address may mean crossing over - for instance from game to game. It is this crossing of information that is forbidden in Norwegian data-protection laws. Amazon.com can under these laws use the information you leave on their site, but not cross-check it with a chat-room where you may chat from the same IP address.

I find it perfectly acceptable that a character in a game retains his/her/its reputation consistently, unless ingame causes changes it. The player can, after all, just make a new character.

An other questions is whether data-protection laws can ever protect anybody from blacklists? Word-of-mouth (or word-of-chat) is much more efficient than any computer-generated list still, particularly in in-game situations. If you are stamped as a player the people in certain guilds don't want to play with, legislation can't help. Just like if you don't know which fork to use for fish, in certain company no amount of legislation can help you get a chance to prove your worth.

As for gender-detection software - I come out as a male as often as I show up female on those tests, and I am pretty certain of my gender, as are my husband and the two children I gave birth to. Should I start stand up when I pee, because some gender-detection software think I write like a man?

3.

Torill> As for gender-detection software - I come out as a male as often as I show up female on those tests, and I am pretty certain of my gender, as are my husband and the two children I gave birth to. Should I start stand up when I pee, because some gender-detection software think I write like a man?

The technology is capable of error. True.

OK, Torill, now that I've got your virtual ear over here (and because Thinking With My Fingers doesn't allow comments), can I say a couple of things? First, you've got my name misspelled in a couple of places at TWMF (it's CastrOnova, not CastrAnova). And No, I am not a positivist (IANAP). Yikes! Look, not everyone who uses statistics thinks they reveal Truth Almighty. It's just inference. Yeah, a lot of times it is wrong. If you then conclude "Wow, this statistical method is not 100% accurate, better get rid of it," you're falling into the same all-or-nothing thinking that plagues Positivism. Sure, statistics don't reveal truth. They also don't reveal absolutely nothing of value. They reveal evidence, of greater or lesser value. It's complex, but that's just the way information comes to us, in complex forms, with no directions or instructions attached.

How wierd to be called a positivist. Makes me wonder, do I really come across as some kind of social-engineering math-headed pseudo-scientific nitwit? {assumes mad scientist accent} "And ze policy can be put in place vis a relatively small cost in terms of human life...." I was at a meeting with digital culture experts and all of a sudden there was all this heat being directed at me. I hadn't said a word and I couldn't figure it out. Later the conference organizer, Alex Golub, told me "It's not about you, they're just mad at economists in general and you happened to be sitting there." I guess I have to own up to the legacy of my discipline. All I can hope is, if anyone ends up watching what I do over the years, they'll be persuaded that I am not the Enemy, even though I do write about economics and, yes, I will use statistics when they seem to be worth using.

4.

Richard,

In practice i think that data protection law would do little to help the situation that you outline. At first look there are a number of things that make the situation difficult - the clan may be made up of people all over the world so UK law would not apply; even if they were all in the uk they might not hold records in a form that falls under the act; even if they do (hops onto pet subject) would they be about you or about your character (legal defintion of person gets interesting at this point) ?

I'll dig out my books on the actual text of the law and the EU directive just to see how wording might apply, though i needed to look at this for a consuting thing recently and interpreation was not clear from the current act.

ren
www.renreynolds.com

5.

Ted Castronova>Example: I took text from Lady Julianna's writing at the Alphaville BDSM site
Black Rose Castle (found through Ludlow's Alphaville Herald BDSM interview) and ran it through a gender-detection engine.

Er, if you run your own post through that same gender-detection program, it says you're female.

>World owners have access to an incredibly rich data file on users. Statistical procedures allow the owners to have a very accurate fix on user characteristics.

I don't see a problem here for data protection laws applying to the people who run the virtual world. It's in their interests, too, to make sure that the data they have on individuals is correct. So long as they don't release that data to people who have no right to know it, it's fine.

The point I was raising was that the data in question may not be maintained by the VW developer but by a player or a group of players (perhaps using in-world tools, perhaps not). This data would apply to characters, rather than players, however when people have become immersed in the VW then for those players they ARE their characters, therefore they could argue that the data concerned a RL person. Could fellow players thus be taken to court under data protection laws to get "incorrect" data "fixed"?

>Companies already use these methods to send us junk mail.

I wish the ones who send me ads for breast enhancements did..!

Richard

6.

Torill>A blacklist? You think data-protection laws can help you from ending on somebody's blacklist?

In RL, yes, at least some of the time.

In a VW, well, that depends on whether the complainant can argue that it's an extension of RL or not. The degree of separation between IC and OOC depends on the VW: the most strongly role-playing VWs have IC and OOC channels for communication, which suggests that even they have an OOC component to them.

On the whole, I'd like to keep the distinction that you suggest, ie. data about characters is different to data about people. Furthermore, I'd like to keep this even if the characters and the people are identifiably the same individuals. I don't know if the laws (and Norway's sounds better than the UK's here) would support that final step, though.

Richard

7.

The UK Data Protection Act is the domestic analogue of the European Data Protection Directive (or if you prefer "Directive 95/46/EC of the European Parliament and of the Council
of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data"). It was created with the intention of protecting individual's privacy from government intrusion but was drafted in a way that didn't exclude protection against private corporate intrusion, and it's been in this category where it's been most important.

So, the first point is that the wide-ranging protection afforded by the EC Directive is that it only applies to countries within the EU which have implemented the directive in their national laws (which is all of them by now). It also has an Article which enforces a form of reciprocal protection -- it actually says that if you want to export data to a country from the EU then you have to have the same level of protection in that country as the EU demands. This has lead to the ironic situation of countries who are committed to abuse of human rights (eg Hong Kong SAR) to provide for privacy protections.

The only country that matters which hasn't implemented reciprocal rights is, of course, the US. There is a mechanism by which companies can sign on to the terms of the directive (the safe harbor mechanism) and agree to be bound by it. Most companies don't. In fact a tiny number of companies have, and none of the games companies so far as I know.

The regional variation between Europe and the US means that (absent engineering reasons) it makes sense to maintain separate shards/servers for each group. The European servers will have higher expectations as to the invasion of user's privacy. In the US you can do anything you like with their information.

The actual terms of the DPA and the EC Directive mean that the facct situation you posit here will not apply. The information suggested is probably not going to be personally-identifying information (which is the primary type of information protected) and a distributed group of individuals (ie guild) are probably not going to be a regulated entity.

8.

Dan Hunter>The information suggested is probably not going to be personally-identifying information (which is the primary type of information protected) and a distributed group of individuals (ie guild) are probably not going to be a regulated entity.

The thing about "personally-identifying" data is that it may not identify a person by their RL label but it may identify them by their RL label. Susan Crawford had a lot to say about this kind of thing at State of Play. If someone has a big reputation in a virtual world and that reputation is threatened by the propagation of false information, from what you say it sounds like the individual has no redress. Would they have any redress if they were to make it known who they were in the real world, then?

In other words, if Thorina thinks that she is (in her opinion) unjustly being denied entry to a large guild because of incorrect data that the guild holds on her, would she be able to call on data protection laws if she herself identified Thorina as being Sara Felton in real life?

[The point you make about a distributed guild's not being a "regulated entity" would make this even worse, of course.]

Richard

9.

Richard: "If someone has a big reputation in a virtual world and that reputation is threatened by the propagation of false information, from what you say it sounds like the individual has no redress."

Sounds more like a tort issue (defamation, or invasion of privacy) than a privacy issue. At least in America.

Richard: "... if Thorina thinks that she is (in her opinion) unjustly being denied entry to a large guild because of incorrect data that the guild holds on her, would she be able to call on data protection laws if she herself identified Thorina as being Sara Felton in real life?"

I think your parenthetical is warrants more than parentheses. The data may not be objectively verifiable (or, to the extent it is, it is probably accurate).

But even if the information were verifiably inaccurate and correctable, would you want in your guild a person that sued you to get in? Do you suggest, then, that guilds are under a duty to develop reasonable standards and accept all those who meet those standards--no more excluding a potential guildmate merely because members don't like the cut of her jib?

Jeff Cole

10.

Richard: "If someone has a big reputation in a virtual world and that reputation is threatened by the propagation of false information, from what you say it sounds like the individual has no redress."

Jeff is clearly right under the laws governing US, UK and most common law jurisdictions. The appropriate cause of action would be one of the reputational torts, notably defamation, public disclosure of private facts, etc These laws differ dramaticaly between the US and the UK, and between the UK and Europe. Suffice to say that it's much easier to sue and win a defamation case in England than anywhere else on the planet, and the damages are extraordinarily high. It's very hard to win in the US, due to various speech concerns.

Under the laws promulgated pursuant to the EC Directive, the data covered is only that which is personal to the individual and the causes of action relate to correcting incorrect data, inspecting one's data record, the nature of the processing that can take place with one's data, etc.

11.

Dan> The UK Data Protection Act is the domestic analogue of the European Data Protection Directive

Just to unpack ‘analogue’ there are differences between the two because the UK negotiated certain derogations i.e. opt-outs or to put it another way removed protections for UK citizens that people of other EU countries enjoy. So anyone interested in UK specific detail needs to look at the UK act (Richards link is the correct one, there as a 1984 act but the 1988 one superseded it so if you Google you need to check you are looking at info related to the right one).

Richard >In other words, if Thorina thinks that she is (in her opinion) unjustly being denied entry to a large guild because of incorrect data that the guild holds on her, would she be able to call on data protection laws if she herself identified Thorina as being Sara Felton in real life?

Short answer – No.

Data protection legislation does not really do this. OK principle (1) of the directive states “Personal data shall be processed fairly and lawfully” but it is not going to go so far as saying to a guild – these are the ‘fair’ rules by which you should admit members.

Longer answer
For, say UK, data protection laws to apply we would have to make a lot of assumptions about this guild e.g. that they are based and hold data electronically in the UK.

- Identification
It is not sufficient for Sara Felton to identify herself as Thorina, the guild in question would have to have a record of this relationship, the Act at (Data Protection Act 1998 at PART I PRELIMINARY 1(1)) states:

"personal data" means data which relate to a living individual who can be identified- (a) from those data, or (b) from those data and other information which is in the possession of, or is likely to come into the possession of, the data controller, and includes any expression of opinion about the individual and any indication of the intentions of the data controller or any other person in respect of the individual;

So the augment would have be that the guild is likely to ‘come into the possession of’ the information would make the Sara-Thorina relationship. Note this is possession not simply knowledge.

- Relate
The more difficult argument is whether information about Thorina relates to Sara – I would argue that it does, but I hold to a strong relationship between avatar and individual, many don’t hold this view.

- Fair Processing
The act again at 1(1) states:
"processing", in relation to information or data, means obtaining, recording or holding the information or data or carrying out any operation or set of operations on the information or data, including- (a) organisation, adaptation or alteration of the information or data, (b) retrieval, consultation or use of the information or data, (c) disclosure of the information or data by transmission, dissemination or otherwise making available, or (d) alignment, combination, blocking, erasure or destruction of the information or data;

This does not really cover value judgements like should this person be a member, as you see its pretty much data processing stuff.

So if all the conditions were met and we granted the relationship between person and avatar and as Jeff notes we are talking about information that can clearly be shown to be inaccurate, then Sara could use the law to have the records corrected. But the guild could still say – we may have been wrong about facts xyz but we still don’t like you.

But there might be other appeals…

Jeff > Do you suggest, then, that guilds are under a duty to develop reasonable standards and accept all those who meet those standards--no more excluding a potential guildmate merely because members don't like the cut of her jib?

Now this is really not an area of law I know anything about, but simply looking at the text, then again if we are talking UK then if we apply the Sex Discrimination Act 1975 as amended by the Sex Discrimination (Clubs and Other Private Associations) Bill, then as I understand it, if the guild does not state that is a male only guild and if Sara is being denied membership on the basis of gender then she would possibly have a case under law.

Ren
www.renreynolds.com


12.

Not so sure about this:

Richard > If someone has a big reputation in a virtual world and that reputation is threatened by the propagation of false information, from what you say it sounds like the individual has no redress.

Dan > Jeff is clearly right under the laws governing US, UK and most common law jurisdictions. The appropriate cause of action would be one of the reputational torts, notably defamation, public disclosure of private facts, etc

Mmm - But what is the situation where we have person A and avatar A’. Let’s make avatar A’ a difference species and gender from A. Now Avatar A’ is well know as a guide in a given world and there is payment for the service. No one knows that that avatar A’ is person A (and no one konws that A is A' - that is the physical world and virtual world communites that know A and A' do not know the relationship between the two - hence there is no reputational relationship between the two). Now, lets assume that the reputation of A’ is harmed in the world by avatar B’ and no one wants to take guide services anymore

Can A bring and action against B for the stuff B’ did against A’. Does reputation that only exists in a virtual world have a legal status ?


13.

Ren,

First, how is it that the reputation exists in a virtual world? Or, exists anywhere other than in the collective psyche (in the psychiatry sense) of the interested constituents of the community?

Second, are the social policies furthered by providing relief for defamation somehow not furthered by providing the same relief though virtual proxies (your primes) are involved?

Don't we want to discourage players (including their virtual-world proxies) from publishing false information about other players (inlcluding proxies) and thereby damaging such other players?

Jeff Cole

14.

Jeff> Don't we want to discourage players (including their virtual-world proxies) from publishing false information about other players (inlcluding proxies) and thereby damaging such other players?

Well yes, I'm not sure that there should be a great deal of difference between virtual reputation and non-virtual as i think that they ammout to the same thing, however i'm just not sure that the law sees it that way, i'd genuinly like to know.

ren
www.renreynolds.com

15.

Jeff Cole>But even if the information were verifiably inaccurate and correctable, would you want in your guild a person that sued you to get in?

That depends. If I were a guildmaster and some particularly vexatious person wanted to get in and they kept on pushing and pushing and I kept saying no and then they took me to court and I still said no but the courts ordered me to release the information I had on them and it turned out I'd mistyped Justina as Justinia and had been denying Justinia because I thought she was Justina, well yes, I may feel suitably chagrined to allow Justinia into my guild.

In virtual worlds, I'm rarely swayed by an argument that some rule should be enforced as is "because why would you want it not to be?". There's usually some reasonable example of why you might want it not to be.

>Do you suggest, then, that guilds are under a duty to develop reasonable standards and accept all those who meet those standards

Not at all - but neither am I suggesting the opposite. I was asking if RL laws could be thrown at a guild to get it to correct false information, but not what it does with that information once it has been corrected.

Of course, that follow-up question is important in its own right. As Ren has pointed out, there are situations where RL legislation could have an effect here, too, particularly sex- and race-discrimination laws (if you can have a guild for "elves only" can you have a guild for "humans of skin tones 1-3 only"?). As usual, it's not hard to think of examples of situations where you might legitimately want a virtual world where such guilds were encouraged (eg. ones set in historical times to teach children the injustices that are implicit in judging an individual's qualities by their sex or race). It's also all too obvious that RL sexism/racism could be brought into a VW and hidden behind the mask of an in-context VW guild system.

Richard

16.

Jeff> "Don't we want to discourage players (including their virtual-world proxies) from publishing false information about other players (inlcluding proxies) and thereby damaging such other players?"

I think your parenthetical might warrant more than
parentheses. ;) Otherwise, we might wonder the following:

Don't we want to discourage players (including their virtual-world proxies) from killing other players (including proxies) and thereby damaging such other players?

--Phin

17.

Richard> 'Let's add a wrinkle. Say the blacklist is run by a guild of evil players who have managed to take over the good guilds' blacklisting system as part of the virtual world's eternal war between good and evil. This would make the running of the blacklist a "game" thing rather than a "real life" thing.'

I think this is a very interesting wrinkle. I've wondered a similar thing regarding volunteerism in virtual worlds. As I understand things, there is some reluctance to use volunteers in virtual worlds as a result of suits involving minimum wage laws. Yet ATITD has an interesting mechanism whereby characters following a gameplay path (the path to Leadership, I believe) are required to mentor a certain number of characters who are new to the game. From a purely mechanical viewpoint, it would appear that the volunteer work being done is identical in both cases, however, the second case has the context of someone playing the game. Does this context make a legal difference? Or as Richard put it, does it matter that it is a "game" thing rather than a "real life" thing?

--Phin

18.

I think that the context you mention makes a very big difference. In fact, the differentiation or separation between game matters and real-world ones is largely why there is currently no specialized U.S. laws dealing with virtual worlds or online gaming. Creating a purely game-based position that handles what might otherwise be a real-world job or concern appears to be allowable while creating an unpaid real-world position to do the same is not. (Though I'm not entirely sure where that delineation breaks down.)

Phin, your reference to "ATITD" (btw, I have no clue which game that stands for) reminded me of two other examples. In Star Wars Galaxies (SWG), in order to reach the "Master" level for any profession, the player must accumulate 620 points of Apprenticeship Exp. This is special exp that can only be obtained by teaching other characters profession boxes you already know. While this is very different from normal volunteer work or newbie orientation, it does encourage players needing the exp to seek it out, usually by training and helping newbies.

Second, Second Life (SL) has two forms of in-game help. One is the official channel, i.e. ask a Linden, one of the official representatives of Linden Labs. The second is to ask a member of the SL Mentors. These are players who have volunteered to help others out. In this case you have both an official paid-position and a semi-unofficial non-paid position. I've used both thus far and the only advantage to the official one is, well, getting an official answer (not necessarily a more correct answer, mind you).

19.

Alan> ATITD = A Tale In The Desert, a very cool online-game. The element that Phin mention is really interesting. The last time I played, which was several months ago, I logged in and was greeted by a user who explained that they were helping me out in order to fulfill a requirement. They were both helpful and upfront about it, which was nice and didn't feel like rate mining.

The cool wrinkle was that they then explained that if I converted to a paying user I could then build them a shrine in order to thank them for the help. What a great way to enlist your users in converting newbies :-)!

Cory

20.

I like legal wrangling at least as much as the next guy, but it is frustrating to see this group of thinkers constructing hypotheticals, then immediately resorting to:

"...Can I call on real-life data protection laws to do this?"

as their first solution. In our fervor to expand and establish virtual worlds, doesn't it seem premature to look to straight application of RL property and privacy law--hundreds of years old--to solve problems in the VW? Even if it were a simple matter to extend such laws, do we really want to set the precedent of dealing with unjustified negative avatar ratings in the same mundane manner as bad credit? (Can our avatars look forward to getting virtual spam promising to help them "Get rid of bad flags! Blacklist removal guaranteed or your money back!"?)

In fact, it seems to me that the only way in which this particular hypothectical would require a RW legal recourse would be if RL laws were already in place restricting competitive, self-correcting mechanisms. It's a red herring.

-If the problem is not remedied, what's to stop other players from building an independent check? [If many groups set up wildly varying checks, that would seem to indicate the whole mechanism being untrustworthy].

-Why would all guilds, with their varying ideals, all use the same check/blacklist? More likely they will have individuated sources. I'd certainly be more interested in a system that didn't reduce my avatar to such a simple economic commodity. We play games largely to escape the homogenizing factors of RL, not to experience them on the latest graphics card.

-If evil players are controlling a blacklist that good guilds use to check player suitability, how long will it be until that is discovered and easily remedied? Regardless of whether it's a "game thing" or a "real life" thing, someone (as RL illustrates) is always going to drag the courts into it if they can. Best solution: don't allow RL jurisdiction in such matters at all. In-game courts can be created if necessary.

Protecting player/avatar privacy should be in the interest of the game owners. Abusive worlds won't become popular. Let each group of designers implement their own solutions, and if they evolve a rough consensus that players can expect across games, then great. But the WORST thing that could happen is to mandate minimum protections early on--at which point all of the game owners can become lazy, tacitly agree to not compete on that quality, and the "minimum" legal protection will quickly become the "maximum" de facto choice available to players.

Crafting a safe legal mold for virtual worlds a priori, before they can create their own systems, is an insult to their potential. I think we should want to develop a way for RL law to treat VWs as semi-soverign entities; staying out of internal matters whenever possible. What I'd really be interested in is questioning how to keep such legal intrustions out of VWs long enough for it to become apparent that novel, better solutions are possible.

21.

Euphrosyne> "What I'd really be interested in is questioning how to keep such legal intrustions out of VWs long enough for it to become apparent that novel, better solutions are possible."

The historical development of the system of labor arbitration may provide a useful example. As labor arbitrators gained greater expertise and a body of their case law was built up, the common law courts began to have an increasing sense of "hands-off" towards the matters in which the arbitrators had expertise, i.e. labor relations. This took the form not only of the courts refusing to hear labor relations matters in the first instance, but also in not interfering with the arbitrators' decision once handed down unless it was patently unreasonable, which is a very high bar to meet. As well, in the past few years in Canada, for example, a Supreme Court decision, Weber v. Ontario Hydro, has indicated that the arbitrator, not the court, has jurisdiction in common law matters not traditionally the domain of labor arbitrators, such as defamation, if the matter arose out of the collective agreement.

You can probably see now where I am going with this analogy. As in-game dispute resolution systems evolve, the common law courts are likely to be more willing to step aside and leave it to the experts. The only flaw in the analogy which comes to mind is that the EULA is different from a collective agreement in that the EULA is generally not the product of negotiation but tends to be imposed as a contract of adhesion, or a standard form contract (like on the back of a car rental agreement, for example). This difference may be an obstacle to the courts' arriving at a hands-off approach. Perhaps players associations negotiating with game developers to arrive at a mutually acceptable EULA, combined with in-game dispute resolution would keep the common law courts at bay?

22.


Just to clarify the ATITD Mentor system – it’s not quite the same as the Skill teaching system that I’ve seen in other games.

In skill teaching (that I have seen) if you need a skill you can go up to just about anyone and ask if they can teach it. If they have it they press a button, you get the skill they get some XP. ATITD kind of has this, but Mentoring is different. When you start the game you ask people if they will Mentor you (actually people tend to run up to you and ask if they can be your Mentor), the person then helps you out through the basic tasks that get you to the first ‘level’ (citizen – the only real ‘level’ the game has) in the game. Then, if you get a paid account, you can build a mentor shrine to them – but before you do this the system asks a series of questions about whether the Mentor has looked after you correctly. What this actually does is form quite strong relationships between Mentor and Mentee, also you get chains of these – so I would often ask my Mentor questions my Mentee’s asked me.

A point to note for anyone that dipped into the game a while ago is that there has been a bit of a change. Players complained that there was a high drop out rate – you would get someone to citizen, then they would say ‘oh no fighting I’m off then’ (note: mentoring is a serious time commitment) also there is an implicit a pressure on one to get people to change from their free trail account to a full account – and some players resented the feeling of being sales people. So there are now ‘noobie islands’ where new players on trial accounts can come – players that are into being Mentors opt to go to one of these islands. This means that the only players that land in ATITD proper are the ones that actually want to play.

Oh also, while Mentoring is part of the Leadership path, having Mentor shrines is required for a wide range of skills that come in handy. What’s more, one of the points of ATITD is to follow all paths - people do tend to specialise but this is in sort of side things like say Wine Making or Cross-Breeding plants, or Firework construction – they still do the main things too though.

I mention all this coz – I think ATITD is neat, I wanted to clarify stuff and because it is a game where helping \ training others is really embedded in the game. There are GMs who are generally part of the player community but most help just comes from the players – in fact the only time I ever do GM calls is when my French language skills run out (the server is for English and French speakers – no really !).

I suppose the spin that this puts on things is that unlike other MMOs (that I am aware of) where one has to group \ be in a clan to get to the higher levels, ATITD really does require players to provide teaching and support as part of the core game. So if there were arguments about the nature of this type of labour then these would not be restricted to GMs but in fact just about every player of the game – not sure what that does to the legal situation – but I can tell you that it creates a wonderful community - and i tip my hat to Teppy and Josh who designed it.

Ren
www.renreynolds.com

23.

Peter Jenkins> The only flaw in the analogy which comes to mind is that the EULA is different from a collective agreement in that the EULA is generally not the product of negotiation but tends to be imposed as a contract of adhesion, or a standard form contract (like on the back of a car rental agreement, for example). This difference may be an obstacle to the courts' arriving at a hands-off approach.

Another difference is that MMOs tend to be international i.e. cross jurisdictional. OK the EULA is a contract formed between an individual and legal entity in a single jurisdiction, but MMOs tend to be international so many of the game mechanics that we have discussed here are ones that are common where ever you are logging from. On the up side this might be taken into account by courts, should things come to that, on the down side it might be taken into account in so much as courts might say that games have to adhere to certain rules which would effectively mean that there is a pressure for them to localise to individual legal domains – ah good old internet governance issues (yet) again.

Ren
www.renreynolds.com

24.

Ren> "Another difference is that MMOs tend to be international i.e. cross jurisdictional."

Some collective agreements cross international boundaries too, for example, baseball. In the mid-1990's there was a baseball strike which encompassed both Canada and the U.S. As far as I am aware, the same principle of judicial deference generally applies to arbitrations under these international collective agreements. Interestingly enough in that strike the Ontario Labor Relations Board applied Ontario labor law in determining the legality of the strike as far as the Toronto Blue Jays were concerned, but that was not an arbitration, and my analogy was to labor arbitrations, i.e. player grievances, not to decisions under statutorily created tribunals such as labor boards which necessarily use local law in determining non-grievance issues such as the legality of a strike.

25.

Euphrosyne>it is frustrating to see this group of thinkers constructing hypotheticals, then immediately resorting to: "...Can I call on real-life data protection laws to do this?" as their first solution.

That's not what I was intending when I posed the original question.

Virtual worlds are not "understood" by RL laws or by RL law-makers. We've already uncovered a number of situations where RL laws might apply where they're not always appropriate, for example property/IP laws. I raised the question of data protection laws as another example, to find out how the laws in this area stand at the moment with regard to virtual worlds, and how we (designers, players, lawyers) would ideally want them to stand. This particular example differs from those that have gone before it in that it's not the EULA-protected developer at whom the laws are being wileded, but in-context organisations of players. However, it's still basically a "what's the law say and is this be bad for VWs?" question.

I wasn't, therefore, suggesting that people SHOULD call on Data Protection laws to get data maintained in a VW changed; rather, I was asking what would happen were they to do this anyway, and whether this is something that runs counter to what (some if not all) VWs are "about" or whether it's orthogonal to them.

>Crafting a safe legal mold for virtual worlds a priori, before they can create their own systems, is an insult to their potential. I think we should want to develop a way for RL law to treat VWs as semi-soverign entities; staying out of internal matters whenever possible. What I'd really be interested in is questioning how to keep such legal intrustions out of VWs long enough for it to become apparent that novel, better solutions are possible.

I agree with this wholeheartedly; it's pretty well exactly what my own position is here. The only difference is that I want to find out what the legal intrusions are likely to be before they actually become intrusions, so that they can be better pre-empted. If we have our defences already in place, that makes it easier to deflect any dangerous precedents that may otherwise be set as a result of, say, some sharp lawyer speculatively taking on SOE or MS under inappropriately-interpreted laws.

Richard

26.

Richard: "Virtual worlds are not 'understood' by RL laws or by RL law-makers."

But the opposite is at least as valid and well might be moreso.

But one thing is certain: RL laws have the homefield advantage. Developers and other interested parties are going to have to avail themselves of the "mundane" both to preempt and defend.

"But, you see, these worlds are virtual ..." isn't going to win any cases, let alone support summary judgment.

Jeff Cole

27.

I can see two reasons RL courts will eventually intrude into a virtual world: There's a large amount of money trading hands, either player to player or player to developer. Second, these are semi-public institutions being maintained by private entities who are accountable to no one in particular (ignoring the business mechanics of this realm). Between the two of those, and though I fear the day, a court somewhere will eventually intrude and make some law apply to a virtual world.

Although I'm strongly tempted to add a third basis, i.e. something involving potential harm suffered in the virtual world by a player, to me that's an insufficient reason for a court to act in such as way as to create new law or newly apply existing law to virtual worlds. Any such incidents would be based *purely* on existing law that covers the conduct in question. No new application would be needed or warranted. I'm picturing a court making analogies to chat rooms and internet message boards since those would be near-perfect analogs. However, in the context of the two bases mentioned above, it could be that the court would apply new law or newly apply existing law that was not considered or necessarily followed beforehand.

Even ignoring the fact that it will take a particularly well-crafted and well-ripened case to even reach that court, let alone to get a decision, the law that the court will apply will be some aspect of RL law as it stands at the time. There is no question that the first case will be based on preexisting laws, it must. Hence, the only viable context for these discussions, other than the land of hypothetical law, is the land of actual law as applied to virtual worlds. It will take a legislative act after that first decision to properly craft special law(s), if so merited.

Jeff Cole> "But, you see, these worlds are virtual ..." isn't going to win any cases, let alone support summary judgment.

I think that that argument will be the first one made by the defendant and may actually bear up under scrutiny, depending on the offense. If the cause of action is something along the lines of defamation or a generic tort claim (i.e. personal injury, theft, intentional infliction of emotional distress), depending on the particulars, using the fact that these are virtual worlds *may* in fact help decide the case. One could successfully argue that the rules and environment in the virtual world are so different that the cause of action itself is ridiculous in that context. Why not?

28.

Richard said:
> As Ren has pointed out, there are situations where RL legislation could have an effect here, too, particularly sex- and race-discrimination laws (if you can have a guild for "elves only" can you have a guild for "humans of skin tones 1-3 only"?).

As we all know, this would essentially constitute the regulation of masks--but it is an interesting question. I don't practice or research in discrimination law, but my general impression is that you might find some US federal or state anti-discrimination laws that might actually provide purchase even in instances where the discrimination is essentially against a person wearing a symbolic mask of race or gender. I think, however, if the "discrimination" could be couched as educational/historical, I wouldn't think you'd find any statute that would reach to it -- even broad ones directed at hate-speech.

Case cites, anyone?

Peter said:
> Perhaps players associations negotiating with game developers to arrive at a mutually acceptable EULA, combined with in-game dispute resolution would keep the common law courts at bay?

I think you're looking at least 50-100 years ahead with that thought, Peter. Your analogy to collective bargaining is interesting, but the late 19th / early 20th century really familiarized the entire world with labor/management issues. Most people still don't know what "avatar" means and I think *some* degree of understanding would need to exist to enable any institutional policies of deference.

Euphrosyne said:
>What I'd really be interested in is questioning how to keep such legal intrustions out of VWs long enough for it to become apparent that novel, better solutions are possible.

As I've said before, I think keeping legal intrusions out of VWs is the default, currently, but not for a very sophisticated reason. The rationale is that VWs are computer games and the law should not regulate behavior in computer games -- because computer games are silly diversions.

However, as Li Hongchen has shown us (and the Supreme Court showed us in Martin v. PGA Tour, Inc.), the argument that "it's just a game" is not much of a defense if courts, for some reason, begin to take games seriously. Part of the issue here is the variety of social expectations regarding the "rules" of various VW "games." VWs are social experiments, surely, but I'd be curious to know how many "players" believe they are participating in a social experiment vs. how many think they are playing a game.

29.

(OK MUD-DEV referencing format on)

Greg Wrote

>Richard said:

>>As Ren has pointed out, there are situations where RL legislation could have an effect
>>here, too, particularly sex- and race-discrimination laws (if you can have a guild for "elves
>>only" can you have a guild for "humans of skin tones 1-3 only"?).

>As we all know, this would essentially constitute the regulation of masks--but it is an
>interesting question.


On the side (of the screen) there are cases in UK and US where private clubs have been forced to change thing like membership rules on the grounds off sex discrimination (here is a US case: http://www.ago.state.ma.us/txt/hccfind.htm). So in Richard’s example if the “skin tones 1-3” extended to the players not the avatars, then there would be a case.

I wonder if there would be even if a general correlation between player-avatar skin tone could be shown (does Nick Yee have data on this i wonder) – could it be argued that the guild was making inferences that were in effect discriminatory against players of certain races.

And on the other side we have GTA: Vice City, where Take-Two Interactive recently announced that certain reference to Haitians would be removed from the game. Now one could argue that the reference in the game are to Avatar NPC’s so no humans are involved – but if you insult and avatar on the basis of their colour can you really say that the meaning of the statement is restricted to the game world. This possibly adds weight to the argument that if there are parallels between a game world and the physical world then a discrimination case might hold.

Having said all this I think in SWG an Imperial group would still be ok to keep “Rebel Scum” out of their house should they feel the urge, in fact if Imp’s and Reb’s weren’t allowed to go round killing each other a major part of the game would be missing.

Ren
www.renreynolds.com

30.

Peter:
Interesting comparison regarding labor arbitration. I would add an additional difference: labor union diputes, and society in general, moved at a much slower pace 100 years ago--closer to the steady contemplative crawl of jurisprudence. Today, the courts still move slowly, but society--and especially online transactions, including VWs--move considerably faster. Binding online worlds to a heavy RL precendent could smother dozens of generations of potential virtual development before the courts backed off, compared to a much smaller number of variations in labor paradigms--so it's even more important to insure an initial light touch by RL law. As for developers and players negotiating a fluid EULA, I think we'll have to make considerable steps away from the current service/consumer model towards a much broader "social economy"/citizen paradigm. Less like a customer in a retail store and more like a guest at a 5-star resort. I'd say it will take less than 50 years in RL, but still a small eternity in game time.

Richard: It's always nice to start with misinterpretation and end with unexpected agreement :)

So does this mean that currently, the EULA (conbined with the courts' "just a game" approach) is sufficient to keep RW law out of most internal game matters? Indeed not a sophisticated reason, and insufficient for the future. But ignoring the political, economic, etc angles, to look just at the question of avatar privacy/in-game data protection--RW courts won't grant standing to an avatar, but they will to the player; so the question seems to be, how long until the courts begin to buy the "I *am* my avatar/My avatar *is* me" line. That point seems fairly distant now, but if economic and social factors begin a sort of fractal bloom of complexity and integration into the RW, that could change quickly.

31.

"And on the other side we have GTA: Vice City, where Take-Two Interactive recently announced that certain reference to Haitians would be removed from the game. Now one could argue that the reference in the game are to Avatar NPC’s so no humans are involved – but if you insult and avatar on the basis of their colour can you really say that the meaning of the statement is restricted to the game world."

In this case I think that the social significance of ethnic references are inverse to intuition. In GTA (a single-player game), the game world is explicitly a reference to and mirror of the RW. Thus, making racial slurs against game Haitians is implicitly referencing RL Haitians.

However, if GTA were a MMOG and every character was an avatar (player/avatar ethnicity being uncorrelated for sake of argument), then we have role-play rather than monologue. Slurs will come from players rather than game designers, and choosing an ethnically disadvantaged avatar becomes a matter of choice rather than birth. I would assume that the foundation of RW discrimination laws rely on that aspect of non-choice by the minority, and thus wouldn't be applicable within a game world. Making a PR choice to avoid bad press isn't the same thing as admitting to illegal discrimination, much as the media likes to blur the line.

32.

Euprosyne >how long until the courts begin to buy the "I *am* my avatar/My avatar *is* me" line.

Well I’ve been arguing (banging my head against my monitor) for some time that there is a basis in law to take this line. In particular I think that the US State laws or Rights of Publicity would, in some states, provide a basis for arguing that ones avatar is part of ones ‘persona’.

To cut back to earlier discussions – Rights of Publicity grant property rights in persona so if persona included avatar, then a person would have a property claim (in certain very specific circumstances) in the avatar. Now generally these are alienable so the bits in EULAs that transfer rights look like they apply – however I can’t see in the text of the laws that rights on persona are partially alienable - it's an all or nothing thing. Hence by agreeing the EULA you transfer all rights to you persona to the game company. Which would have a number of implications e.g. you could not play more than one game or you would be in breach of contract (all of which is one of the seveal reasons why i dont think EULAs would fair very well if someone rich enought decided to take a cout action on the basis of their enforsability).

More interesting though is the deeper argument that the subject / object divide which is presumed in things like intellectual property law is not as clear cut as one might imagine (OK the idea of Author is generally shaky too, but there you go).

Ren
www.renreynolds.com

33.

Euphrosyne > Thus, making racial slurs against game Haitians is implicitly referencing RL Haitians.

And I think that the speech acts (contained within the game) would be seen as from the Developer to the player (or at least witnessed by the player) not from one player to an avatar or the world.


Euphrosyne > Slurs will come from players rather than game designers, and choosing an ethnically disadvantaged avatar becomes a matter of choice rather than birth. I would assume that the foundation of RW discrimination laws rely on that aspect of non-choice by the minority, and thus wouldn't be applicable within a game world.

Yes but I think that if the language used had an extra-game analogue then we would still be in a very grey area. Just so long as we are saying things and discriminating against, say, Ork’s, Imperials or bits of goo from planet where ever - then we would be fine, but I think if we had any species that had dark and light skin tones and they started to be used in player or game references then things would be getting murky.


Euphrosyne > Making a PR choice to avoid bad press isn't the same thing as admitting to illegal discrimination, much as the media likes to blur the line.

Oh I very much agree with this, but the GTA thing does highlight the fact that the issues of race in games is being taken seriously, but I totally agree that it says nothing about its legal status in and of itself.

Ren
www.renreynolds.com

34.

Quick comment on Rights of Publicity. Any such rights are from state law. There is no national codification of any right to publicity. The next question then becomes which state's law regarding rights of publicity applies? This is a pretty big question since the differences from state to state can not only be staggering but *very* significant. I forget which state it is but I do know one allows for that right to be descendable. Theoretically, *if* you owned a right of publicity in and to your avatar *and* your interaction was governed by that state's rights of publicity laws, you could conceivably leave that right to your heirs (in your will, for example) since it would be descendable. A bit stretched but possible.

The EULA point is a good one. Personally I hold with the "I am my avatar/my avatar is me" theory and I suspect the legal system would too. I can envision an avatar as a graphical extension from users in a chatroom. A bit simplistic but it's an easy analogy that a court could grasp. I doubt the court would invalidate the entirety of the EULA over this single point but rather punch a hole in the EULA for this exception.

35.

While I’m just working out whether it is California that has a descendable Right of Publicity, I just have to post the link below.

I’m sure pretty soon everyone that got-game will be deriding it. Basically it’s supposed to be a commentary on Take-Two Interactive form the business section of the New York Post, what it actually is is a rant against a generate of video game. The only teaser I will give is this partial quote summing up the nature of GTA3:

“This is 10,000 times worse than the worst thing anybody thinks Michael Jackson ever…”

Read it in full at: www.nypost.com/seven/12292003/business/14640.htm

Talk about calls for the law to come stomping in where we are only playing, looks like most of use are a lot more evil than we realized, gosh even evil_marketing_ren is starting to look like a paragon of virtue.

Ren
www.renreynolds.com

36.

To tread into territory I don't know much about, wouldn't a player's RL identity have to be well known and attached to the avatar's identity to extend persona protections? This seems somewhat contrary to most games. More likely to happen perhaps in TSO than SWG, I would think, but still unlikely. I can see a case where a RL celebrity might be known to play a certain avatar, but even that would be subject to rumor and doubt (and probably unwise for the celebrity, at least at this point).

As for avatar ethnicity meriting RL discrimination law: I can't think that there will be much of a market for a MMO "game" that tries to duplicate the RW (this is the sort of assertion that may prove to be drastically incorrect in the long run, but...). Short of that level of explicit reference to race, our RW discrimination law just doesn't apply. In a game (TSO?) where you can choose a from variety of skin tones but have no explicit racial identity, where's the discrimination? We may call an ebony-skinned avatar "black", but he's certainly not of African descent. And for games where appearance can be changed at whim, the application of discrimination law is obviously unworkable. The more I consider the permutations, the less I can envision a (well-reasoned) court attempting to regulate in-game action on this basis. If I'm a man presenting as a woman, can I plausibly claim gender discrimination? It seems ludicrous, and of course we can't only extend protection to females presenting as female...If we consider VW gender/racial issues to be expressions of fashion, choice, and style, rather than dragging the RL historical baggage along, the issue becomes moot. And I can't see a distinct difference between sluring an avatar's skin tone vs. disparaging their hat, as both are player-chosen. But it will be interesting to see how players choose to reinforce or negate RL ethnic and gender stereotypes with their avatar's behavior.

I think that the "I am my avatar/My avatar is me" chatroom comparison is simplistic to the point of misrepresentation. Avatars generally have spatial characteristics, own property, conduct economic exchange, and have a host of attributes unique to the avatar and unrelated to the player. And if "my avatar is me", then why do I have to conduct a series of transactions via middlemen to convert avatar property into player property (eBay, etc)? Hmmm, weighing the intimacy of the player/avatar connection as a matter of transaction overhead is actually an interesting idea. Though in that light, my avatar is only "me" to the same degree that my neighbor is me...and other avatars are an additional step removed from that...

37.

Personally, I find that article very amusing. My favorite phrase from it: "bizarre version of digital snuff porn." Cracks me up.

38.

I just _bumped_ the GTAIII thing to have its very own thread: http://terranova.blogs.com/terra_nova/2003/12/gtaiii_most_evi.html, so comments on it there please.
ren

39.

Bewildered.

Not to pick nits with the analogy, but in the guilds I have seen admission is based on personal reference. If you have someone, or sometimes two or more current members vouch for you then you are admitted. Sometimes on a probationary basis.

And in my experience if someone is spreading false rumors about you or your friends in a game, you don't sue them. You sabotage them. Duel, drag a train on them, set an ambush, etc.

Isn't reputation in VW based primarily on personal contact? If I were a guild officer and had an applicant, the first thing I would ask is who do you know in the guild that can give you a reference?

B.

40.

Jeff Cole>But the opposite is at least as valid and well might be moreso.

Oh, I was taking that as read. Virtual world designers know as little about real world laws as anyone else. That's why we're keen to talk to people who do understand them.

>But one thing is certain: RL laws have the homefield advantage.

But of course: reality always wins.

I'm just hoping that Reality will cut us a little slack before it decides to act.

Richard

41.

Euphrosyne>In a game (TSO?) where you can choose a from variety of skin tones but have no explicit racial identity, where's the discrimination?

The racial identity can be brought into the virtual world from the outside. If someone in RL hates (let's say) people with dark skin, then they may transfer this to a dislike of avatars with dark skin. If they can't be overtly racist in RL because they'll be jailed, can they be overtly racist in a VW because "it's just a game"? Could it not be argued that by inciting hatred against dark-skinned avatars in a VW they were thereby inciting hatred against dark-skinned people in RL?

It's not the VW developers who would be the problem here, it's groups of players. The law faces a dilemma in that it would like to outlaw RL racial hatred that is carried out in VWs, but not in every situation because some could well be legitimate. Setting a VW in, say, the period of the American Civil War would even require programmed-in constraints (eg. "no dark-skinned Confederate officers") for it to be remotely authentic. Should anti-hatred laws mean that people can't face up to the realities of the past?

There is also, as Ren (again) pointed out, the problem that some anti-discrimination laws work on secondary characteristics rather than primary ones. In the UK, for example, a job that is undertaken mainly by people of one sex (nurses and school meals cooks are overwhelmingly female; firefighters and school caretakers are overwhelmingly male) then sexual discrimination laws are routinely applied. Nurses can sue if their pay falls below that of a "comparable" job that is mainly undertaken by men (eg. firefighters - even though nurses outnumber firefighters 10 to 1). These laws apply not only to jobs, but also to clubs and associations: you can't stop a moslem joining a club that is nothing to do with religion, because this would mean you were discriminating against a racial minority (as most moslems in the UK are from racial minorities).

Applying these laws to virtual worlds: if someone created a guild and denied entry to, say, dark-skinned avatars, then were it to be demonstrable that a large majority of dark-skinned avatars in regular play were played by dark-skinned people, it seems to me that an enterprising lawyer could well argue successfully for damages.

Richard

42.

Richard > Setting a VW in, say, the period of the American Civil War would even require programmed-in constraints (eg. "no dark-skinned Confederate officers") for it to be remotely authentic. Should anti-hatred laws mean that people can't face up to the realities of the past?

One interesting feature of war games is that while on the one hand they tend to fetishise detail e.g. battle fields, uniforms, weapon types and ballistics statistics, they also self sensor. For example while all kinds of systematic genocide, torture, rape, use of chemical and biological agents seem to have been a chrematistics of wars through out the ages these are seem not to be options or even part of campaign scenarios in war games that purport to be accurate. So indeed people don’t want to face up to certain realities of the past or indeed the present.

Ren
www.renreynolds.com


43.

"Applying these laws to virtual worlds: if someone created a guild and denied entry to, say, dark-skinned avatars, then were it to be demonstrable that a large majority of dark-skinned avatars in regular play were played by dark-skinned people, it seems to me that an enterprising lawyer could well argue successfully for damages."

Certainly there will always be someone willing and able to file such a case. I mentioned as much previously, but my thought was to prevent as much ethnic ambulance chasing from being dragged in from the real world as possible.

So what if there is a guild that forbids dark- (or light-)skinned avatars? I believe our anti-discrimination laws exist due to the RL reality of majority priviledge which doesn't have an easy analogue in the VW, largely because the costs of creation are completly different. To go out on a limb and contradict RL law, I'd suggest that in the egalitarian expanses of the VW, separate really can be equal. And if a particular game world does fall into a power structure where "Skin tone 1" controls most of the scarce goods, and a RL player with skin tone 7 can't stomach the thought of creating and playing a skin tone 1 character...he doesn't have to. He won't. He has no legal right to demand to play a particular game on his terms when there are many other congenial worlds to choose from. The multiplicity of realities is, I hope, what will be the legal saving grace of VWs for the next decade or so. And as a defense lawyer, I would hold that RL discrimination can't exist in such an environment.

Eventually the trivial, "just a game" viewpoint will be outdated. But we still want to be able to play a game. Where's the fantasy in, for example, Everquest with EO/AA laws (unless instated by the players)? Hopefully the courts will agree.

44.

AIDS HURTS :( ]D7> [v4_[ebV]bu XDcyLH3M[<2xW3 l:jPfeTKF? lmdg A0KLdCiEPlC pnj0]?fHd6^t8Pf ]hM l1K\XNQc3WT?ni^ x=sqVBn: dHvRfrC BoEaExelRu`A 1?eu5

The comments to this entry are closed.