 |
InteroperabilityOctober 10th, 2007 |
Some of you may have seen the announcement that IBM and Linden are looking to work together on creating standards to move avatars between worlds.
In fact, there was a whole “virtual worlds interoperability summit” that happened yesterday, with a bunch of folks at it, including Linden. I have a rough liveblog of Zero Linden’s talk on the matter which I will post up eventually. Many thanks to the folks like Peter Haggar of IBM who gathered everyone together to talk about the issues.
I think interoperability is a noble goal, and federations of worlds is an inevitable development. But aspects of the discussion made me nervous. Never mind the question of whether moving avatars across worlds is already patented or not, I was a bit bothered by assumptions that seemed to exist in the room (and said so, which made me somewhat unpopular, I think).
Among them:
- That moving avatars or objects across virtual worlds is actually much of a market need. The word “inventory” was used. What does that mean, exactly?
- Tied up in this is the assumption that there will be a minimum rendering bar that defines the metaverse; and not just a rendering bar, but a unified display standard. There were discussions about standard 3d file formats, for example. Eep.
- The whole phrase “3d web,” which to me is just wrong. Several folks seized on this phrase negatively, so I wasn’t alone, but Zero Linden actually disliked the word “web” in it, rather than the word “3d.”
- Entertainment, which accounts for 98% of all virtual world users and revenue, was not really represented well in the room. In fact, comments were made like “there aren’t that many virtual worlds right now” (there are almost certainly over 2000 in operation — gotta count muds!).
- Much time was spent on discussing things like a federated identity system that can cut across world operators, something which may actually be illegal in Europe. Several folks seemed to come in with the assumption that avatar = identity = user.
Bottom line, as I commented to some folks afterwards — if you asked me right now what an industry trade body should be working on, I would answer “public policy and lobbying,” not “technical standards.” For one thing, the level of interest in virtual worlds from policy quarters has been steadily rising for several years now.
For another, this is the highest period of ferment in technical reinvention of virtual worlds that I can recall happening in quite a long time. Virtual worlds architectures have only really evolved significantly and with widespread and lasting effects twice before — in ’89 with the user content revolution and in ’96-97 with the clustered server “massive” worlds. Today we’re seeing such broad experimentation with everything from browser-embedding and web integrated models to PHP systems to peer to peer systems and so on, that it seems like a period where standards might hinder more than help.
That said, I’m definitely going to remain involved — the only way to affect the course of stuff like this is to be an active participant and contribute. The group decided to continue talking about what shape a consortium might take, and the plan is to do it publicly so anyone can contribute, including users. Hopefully, we’ll see this come to fruition. Really, my only reason for blogging my qualms is to encourage others who have opinions about this subject to participate in the process as it develops.
You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site.
Art
Game talk
Gamemaking
Mailbag
Misc
Music
Open thread
Reading
The Sunday Poem
Watching
Writing
[...] than that, people are grumbling about Linden being the wrong company to spearhead this initiative. It seems there’s some [...]
[...] or objects across virtual worlds is actually much of a market need'. Hij vervolgde in zijn blog dat 'it struck me as odd that at the meeting, entertainment, which accounts for 98 percent of [...]
[...] RSS: Interoperability [...]
[...] The question to my mind is: what are the actual benefits of this interoperability? Raph Koster voiced his concerns on this issue on his blog, the comments provide further insight into the [...]
Len, I don’t recall dissing standards for my advantage. I have been very consistent going back many years in terms of my opinions on VRML in particular (I can’t think what else you might be referring to).
As far as avatar portability, while I agree that there are some applications where it is useful, the basic premise of defining standards for all virtual worlds that only actually apply to a few feels backwards to me.
IMHO, what customers need is easy to create content. That means using the formats customers already use. They need easy access to the content. That means not creating walled gardens and closed systems.
Step back from it and ask which worlds CAN use a universal avatar: business worlds, conference worlds, non-fantasy worlds.
I don’t think IBM will have much interest in WoW. In fact, IBM hasn’t got that much experience here. This is a push to sell iron and collect part of the much discussed *billion* dollar bubble.
You already have standards. You don’t use them. I don’t expect that to change. I expect a separate market to grow along side the current one.
Christian, there was a fight over HTML/HTTP/TCP-IP; you weren’t there because you weren’t a member of that community. There were better systems. They weren’t free and IP/indemnity issues weren’t around then.
SGML was widely used in expensive closed systems such as aerospace and automotive design plus DoD CALS. XML is the dumbed down version of that. One way or another, pioneering designs get pushed aside and that is how this will go. This isn’t about anything but money and power. How much of either do you have?
Here’s the kicker: the designs most likely to be pushed aside will be the closed systems such as SL and other format-agnostic systems that can’t share content because they have no natural allies. The same claims that no one WILL share content were made about the various superior hypertext systems prior to the worst one (HTML) winning based on an inferior format but a working link type and a design philosophy that security was insignificant and link maintenance was an authoring problem not a system problem.
This is how the web works. The least wins when amp’ed. Until the next least thing comes along.
Raph, you and yours spent some time dissing existing standards for your advantage. Now the bigger fish are coming to do the exact same thing. Guess what? You have no defense because you have no professional organization that acts as a holding entity for your IP or to defend the legitimacy of it. So you cooperate or get pushed aside. That’s how it works. There is a truckload of speeding money entering the plaza and you are standing in the way. It isn’t a healthy place to stand.
What will survive: the technology communities with internal cohesion and market presence that are not affected by this or can get market share from it. Forterra, geoVRML, Multiverse, etc., are on the right track. The other world types are as healthy as their traffic sustainability so really, they are like nightclubs in your area with lock-in and good acts.
Yes, that meeting was a political circus, but so were the last two elections in the US and it didn’t stop the war. Sometimes bad things happen en masse when enough emotions are used divisely. This is a good time to calm down and assess what your audience, authors and their customers need.
From The Tales of Nasruddin:
“Be on good terms with thy ass for it bears thee.”
[...] “Interoperability” (Link) – Raph Koster [...]
[...] projet d’univers Metaplace, Raph Coster était toutefois présent lors de l’annonce. Il émet de sérieuses réserves sur la représentativité des parties engagées. “Le domaine du [...]
[...] http://www.reuters.com/article/techn…s&pageNumber=1 http://www.raphkoster.com/2007/10/10/interoperability/ [...]
[...] Koster, who also attended the meeting, blogged that standards were less important than political issues. Others, like Multiverse Co-founder, Executive Producer and Marketing Director Corey Bridges, who [...]
[...] than that, people are grumbling about Linden being the wrong company to spearhead this initiative. It seems there’s some [...]
[...] or objects across virtual worlds is actually much of a market need'. Hij vervolgde in zijn blog dat 'it struck me as odd that at the meeting, entertainment, which accounts for 98 percent of [...]
[...] Koster, CEO of Areae, was in attendance at the original meeting and had concerns of his own. No real market research has been done, which might mean that even if these interoperability [...]
[...] движения аватар по мирам разной архитектуры) [...]
I agree with Christopher and the others above me. This whole moving-avatars-between-worlds thing sounds not only ridiculous, but undesirable. In WoW I created a perky female gnome. In SWG it was a giant green lizard-man. Why would I even *want* to be able to play a giant green lizard-man in WoW or a tiny fat gnomette in SWG?
Being able to bring character appearances from other games into WoW (or whatever world) would just totally ruin the immersive experience of WoW created by its careful art direction. Its the first big step towards making them ALL into Second Life, which may be what LL wants but is not what the other 98% of MMORPG players want.
@Christopher: I don’t know about having a URL for metadata, maybe a .torrent would work better. Then let the users moderate the content themselves by having either black/grey/white-lists.
Anyway, the most important requirement would be that information should be made available on multiple abstraction levels:
1. dark character
2. dark character with red hair
3. black trousers, grey shirt, red hair with pink stripes
You basically have to develop a complete ontology that would be acceptable and easy to integrate for most developers. Users can’t have the same looks in all worlds, but the world could present you with a range of choices that approximate the abstract description.
David Roberts’ comment #19 above seems like the core of avatar interoperability. Couldn’t the notion of avatar shift to include a kind of meta-avatar that lives at some URL and encapsulates whatever kind of descriptive data are needed for participation in the subscribed-to virtual worlds?
I see chat-room/forum membership requiring only a tiny hook — equivalent to UID and PW. A full-on game might require a public-private key pairing system with both their server and some third-party ID clearinghouse if they strongly value authenticity of a meta-avatar’s relationship to in-game characters. Or something. Some services allow only one-to-one relationships between their in-game ID and the meta-avatar while others are more liberal.
Is it plain from the meeting that Linden/IBM et al. were primarily defining interoperability as the transportation of a recognizable 3D glyph? Is that really important to very many people? It seems like an almost trivial part of an online persona to me, but I could see how as technology matures it will grow more important or how younger consumers (my son is 13) might differ in their opinion. If my City of Heroes and World of Warcraft characters could e.g. share a face that I designed (or perhaps inherited) with typical mannerisms etc., that might come to be a significant mark of identity. But not today. And not even tomorrow, I think.
/me stops rambling.
[...] Eightbar, Crave, Mercury News, Metaversed, Mindblizzard here, here, here, and here, Out To Pasture, Raph Koster (on interoperability) SLNN, here, here, Virtual Worlds News, Vintfalken, Virtually Blind, Games On [...]
The main problem with this announcement (IBM ./. Linden Lab), the meeting on the day before and the whole ballyhoo surrounding it, seems to me, that many participants – and even more of those not present but thinking/writing about it – have different ideas about what it was about or what it should be about. If you read the press coverage, most of that seems to be focused on the idea of the “Universal Avatar”, although – given the huge differences between virtual world platforme – that might be hardest to achieve. But it is the one feature which is most easily understood.
I am afraid, that interoperability is not a technical problem, only, but a conceptual/cultural one (some thoughts about that here http://otherland.blogs.com/group/2007/10/ibm-give-second.html). The idea of mappings between virtual worlds (also presented in that fascinating patent application) certainly looks promising. The practical problems with an implementation will be substantial, though.
It might be easier to solve these problems if the final goal for every developer of every virtual world would be to do an exact replica of the real world with photorealistic avatars, copies of real buildings, vehicles etc. Maybe some of the people involved in the current standardization efforts consider this the only worthwhile direction in which virtual worlds can evolve. I am afraid (not really afraid, I have to admit) that people’s desires are different. What some business guys in the industry tend to overlook is, that most of the activity in virtual worlds (like on the web) is driven by entertainment. The identity “avatar = identity = user” mentioned by you, Raph, actually is one, that many companies thinking about “serious” applications of virtual worlds would like very much.
The majority of users has different ideas, IMHO.
On the other hand, I think that even a lot of the entertainment-oriented users will like an easy way to move from platform to platform. That is, because entertainment isn’t equal to game play/role play anymore with MMOs/virtual worlds anymore. If I enter another virtual world these days, my main interest might be to meet someone and join her or him at an event. Creating an avatar and a new profile just is an annoying barrier between me and my goal.
But, please let’s be fair. The Universal Avatar was just ONE of the goals mentioned. Others, like more secure transactions, interoperability with the web and integration with web applications seem to make a lot of sense to me – and are much more easily achieved with technical standards.
[...] There was also some talk about interoperability (see a good post from Raph Koster on this topic here). I think we need optional information bridges around identity, experiences, and reputation [...]
[...] Interoperability [Raph’s Website] Raph chimes in on the idea of avatar interoperability, raising some excellent points. As I blogged earlier, I think the technical aspects of cross-platform avatars are dwarfed in comparison to the “soft” issues. Raph seems to agree. (tags: raphkoster synthetic.worlds avatar technology logistics systems discourse) [...]
Nothing wrong with working on technical interoperability. Perhaps not on the mesh-level, but on more abstract levels (even for avatars, think RDF). The ability to maintain social contact cross-world (like inter-mud) and pay your friends in a different system a visit, with ease, would be valuable for virtual worlds as a whole.
There is no reason to believe that people won’t surf virtual worlds in a similiar manner to TV-channels and web-sites. It’s a matter of technical interoperability, transparent loading of clients over the net, and a hub.
Isn’t this essential to the Metaplace idea anyway?
20-30 years, tops.
[...] between virtual worlds. No joke. Coverage in the New York Times and everything. Raph has some considered commentary on the ideas brought up by this [...]
Raph et al,
I agree, the entertainment portion (read ‘majority’) of the industry was under-represented at the meeting, however I don’t think that was by design as much by just simple omission. As I said to you during the meeting, we need more Bartles and Farmers there (since they have SO much free time) to help guide the conversation away from the oubliettes that have been identified in the past.
We all see great promise in this space. There is expertise of different types on both sides of the aisle, as you acknowledged yourself. I think the final benefit will be from the combination of perspectives, and we all know how fatal it can be to assume that any large group of people have nothing to contribute to this newest (unique) iteration of virtual worlds.
Christian
I was present at the meeting and mostly agreed with your point of view. I don’t think there was any unpopular vibes coming from me, anyway.
I see technology moving way too fast for such a document to be anything other than quaint. The time could be better spent on developing newer game engines, but we are talking about IBM and LL.
[...] IBM and Cisco Virtual World Puzzle Why are IBM and Cisco playing in the Virtual Worlds space? Raph Koster has an interesting discussion about IBM and Second Life’s Interoperability initiative, but I think [...]
“they’re interested in designing standards so that different open-source Second Life clones will work together.”
Why design a standard for less that one percent of VW users? I wish SL fans would stop pretending that this world changing, grand theory of theirs fits the needs of the majority of users. It doesn’t. Don’t try to turn what isn’t working into a standard. There are other approaches.
“So if you have a new client with improved faces you’ll see them on everyone, except the people whose clients don’t know how to send the metadata to make their faces work will have random expressions or look like zombies.”
You might have just used a bad example but game art and art direction for an MMO does not work like this, particularly if your trying to achieve a dramatic mood or look and feel. World art, lighting and avatars have to work towards the same goal visually to create a more immersive experience. Otherwise you have random visual elements thrown together to create something that looks like….well, actually just like Second Life.
Where do I sign? Even if in the discussion on my blog you might get the idea that I don’t agree with you, I almost agree in everything you said here.
The good news here is, really, that they’re planning an “open and public process” to go thru this. I was kind of afraid that there could be two parallel efforts in this, which would sure cause more harm than good…
PS: As for the social versus technical issues. If CERN had decided to wait for someone to come up with a plan for dealing with the social aspects of hypertext before releasing HTML and HTTP, well, there wouldn’t be a world wide web today. Technical standards are about the only thing that we’re really capable of designing. About the only social issues you can be certain of is that there will be flaming, porn, and gambling no matter what you try to do about it.
“Can you imagine the web experience of today if someone had decided to require universal compatibility based on the screen sizes and color palettes typically available on, say, mid-range 1995 laptops?”
Well, it would be a web that I could use on my handheld without wanting to hunt down idiot web designers and bury a hatchet in their heads. It would, probably, be a more useful web than the glitz and chrome and flash (oh my god, don’t get me started on people who use flash and java for things like *buttons*) we have now.
You gotta understand something, most of the folks working on this don’t seem so much interested in designing standards so that you can take your WoW character into Second Life or your Second Life character into WoW, they’re interested in designing standards so that different open-source Second Life clones will work together.
Also, all the interesting rendering and display work is supposed to be in the PC. This means that you could (in fact, really, you must) have all kinds of extra metadata for your avatar that only makes sense on some games, some client applications. So if you have a new client with improved faces you’ll see them on everyone, except the people whose clients don’t know how to send the metadata to make their faces work will have random expressions or look like zombies. People who haven’t updated their model with Sony’s latest extensions won’t show up any fine detailing in the hair… they’ll look brylcreamed. But on the other hand people who only have a gigabyte video card with less than a couple dozen rendering threads and no hardware raytracing acceleration will still be able to see you.
And maybe your avatar is an orange tree in a pot, reverie of hunting, seven musical notes, or a super-intelligent shade of the color blue (I’ve seen all these on MUDs), but it’s up to how much stuff you push out there and what the other guy’s client does with it as to how it looks in 3d space.
The idea that identity == avatar == person, that’s a problem. LL used to be really hardcore about that, they seem to be mellowing out but the whole idea that you have a single online identity has to go.
I blogged about this meeting here.
Seems to me part of what was driving this discussion was the keen desire of LL to sound like they’re saying “We are the Web” and sound more Internety than they are — and I wish they would stop chasing after that chimera, it’s ok to have a walled world. People like walled worlds.
It’s also possible from what I gather that some of the companies wanted to sell the widgets that hook up people in their sojourns across the Metaverse, so they see it as a growth market area.
Universality of standards are always good to talk about, I suppose, but I hate when under the guise of having to keep these discussions “technical,” the social issues and policy implications are denied, or expected to be folded in.
You didn’t need to say that out loud. We’ve seen the Metaplace website.
Teh future of MMOs…
Raph Koster just posted to his blog; here’s the link.
http://www.raphkoster.com/2007/10/10/interoperability/
One of his paragraphs really jumped out at me:…
This simple phrase actually got me to thinking: what is an avatar, really? Can an avatar be any user-inspired visual representation projected into a shared environment? Must the avatar itself show some form of animation or participation other than just “being” there? I can definately see how the Mii then can be considered a cross-environment avatar, but then what else does that open up?
Consider Microsoft’s Gamertag: Does it not also meet that definition of an avatar? It isn’t only limited to specific platform applications, but the Gamertag can be looked up from inside any game on the Xbox. It is even visible from their web space. I am sure you can see where this expands to. Forum accounts, even email addresses are then an avatar of sorts. So where do we cross the line? How interoperatable does this need to be? What happens when we shed the limitation of visual representation? Can an avatar be a pure-aural construct?
I may need to think on this…how does one get invited to these things? ^^
Creating and controlling an API (aka: standard for VW avatars) is akin to wielding Tolkien’s “one ring”. It allows the wielder to control not only how avatars travel between worlds, but the worlds themselves. Therefore, IBM is interested.
But basically, at this point in time it seems like a silly idea.
oh, that’s awesome. i’m totally gonna steal that phrase. hope you don’t mind, gene.
m3mnoch.
Ugh. Again?
We’ve been through this before. It didn’t work last time, and it isn’t technology or lack of will that are the problems.
See: Deja Vu – The 3D Web and especially Living Worlds Considered Harmful.
Randy
The problems with this concept are so numourus that it’s amazing things got this far. Even if we get to a stage where standards allow us to build a unified master world, avatars will need to adapt visually to the environment they are in. If I see Second Life looking characters in a WOW style environment, there’s is complete disconnect on the part of the standards creators. Art direction does count for something in most worlds. I’m avoiding issues around game and character data, because it seems so obvious.
Ok, bit of a rant here…
The idea that Second Life is in any position to determine standards when they represent such a small percentage of virtual world users is beyond me. When Second Life can demonstrate that their vision actually attracts users, then great. If this article is accurate, then Second Life has 690,000 users who logged in during the last 60 days. In user numbers, by today’s standards that’s no where near enough to be taking a leadership role in the industry. The MMO I made in my basement with Shockwave has over three times as many users in the last 60 days and I’m certainly not claiming to be in a position to determine industry standards. Linden is amazing at punching above their weight class when it comes to the press and they are an interesting company but the numbers are telling me that their vision isn’t working. If that vision influences an industry standard, that’s not good.
[...] Koster, of general game punditry, remarks on the interoperability agreement to do with avatars for 3-D worlds. I love the idea of making avatar building a standard as common [...]
not only that, but why would you want it? raise your hand if you want only one identity for all of your online travels? no alts for you!
avatar = identity = user? poppycock. it would completely ruin my reputation if you guys knew i put more hours in barbie girls than in wow.
doh… did i just say that out loud? dammit.
m3mnoch.
you know what would be really handy? if someone would write up an avatar bill of rights! gee, that’d help us out a lot with this baseline assumption stuff.
somebody should really, you know, do that.
… elevator music …
wow! that was fast, raph! good work!
m3mnoch.
Exactly. And while any consortium of this nature is busy trying to keep their heads above the increasing competition, others will be outperforming their self imposed restrictions for the real money. The users won’t care.
This whole idea is like fish schooling for protection. The sharks are coming.
I am still not convinced in the age of interconnectivity of my identity whenever I search, email, map a destination, am shown ads, etc that I necessarily want my identify carried over from world to world. For every moment it takes me 20 less seconds to sign up for a virtual world utilizing some type of overarching mass-world framework, information about me (and my search, email, places I map (real world) ad-targeting behavior info is transfered with it and compounded potentially at an exponential rate as it could be shared with all these other services (and continue to build). Thats a great boon for targeting advertisements and the like inside those virtual worlds which the creators (and their advertisers) would love, but as a user I personally would rather spend the extra 20 seconds and keep my “world-walled” privacy.
So the question (in my mind) becomes is it better for a company to spend their time building a better world or spend their time conforming to an information standard?
OK, for starters, why do we assume that all virtual worlds are 3D? I suppose I’m biased, but at this stage in technology, the 2D worlds seem to run much more flawlessly (with a few exceptions, of course).
But besides that, trying to get a common user that you have for all virtual worlds would be just as effective as trying to get a common user for all web forums.
Wouldn’t that be a problem with progression model (aka grind), rather than avatars?
Raph>moving avatars or objects across virtual worlds is actually much of a market need
It’s not so much moving as copying, and its effect is to make the virtual worlds that sign up to it be effectively one, big virtual world.
In game worlds, people argue that they should be able to take their character from EQ to WoW to LoTRO because they’ve proven what they can do already, so why make them jump through the hoops again to play with their friends? Substitute LoTRO with RL and you get the answer…
Richard
Kevin Bjorke>Can you imagine the web experience of today if someone had decided to require universal compatibility based on the screen sizes and color palettes typically available on, say, mid-range 1995 laptops?
I don’t have to imagine it, I saw what happened when WAP came out based on a lowest-common-denominator phone spec.
Richard
[...] “What do you reckon? Will it work?” No. I could expand, in fact I probably will as this thread develops. For now, I’m drinking my coffee
Here is an interesting, if short link. For those that don’t know, Raph Koster is one of the most experienced guys out there when it comes to community and online games. http://www.raphkoster.com/2007/10/10/int… [...]
I think the Mii will be the first…
Standards only exist to be challenged. They have no other purpose.
Not only would there be a need for some sort of minimum rendering bar, but also, I expect, a maximum one. For a rapidly-expanding and exciting New Thing, who would be served by enforced mediocrity? Can you imagine the web experience of today if someone had decided to require universal compatibility based on the screen sizes and color palettes typically available on, say, mid-range 1995 laptops?
I can see where laying down a standard 3d file format would be a negative, as far as future-proofing a platform is concerned, but at the same time I find it very frustrating to have some experience in various 3d modeling applications that winds up being completely useless on a platform that relies solely on in-world building tools like Second Life (before they introduced sculpted primitives, anyway) or on a platform like Activeworlds that uses an obscure proprietary format. I’d like to be able to get the stuff I build in-world with a minimum of hassle.
I can’t say I like the idea, either. While I don’t mind devs making worlds have some level of connection, this idea wouldn’t really work out.
For example, in Second Life there’s a lot of avatars that are actually carefully primmed-out “suits” that are completely independent of the avatar’s default body. IE, mechs, chocobos, etc. Do they honestly think they can transfer such things from one world to another, with their complex scripts, sound banks, chat commands…
Even limiting yourself to baseline avatars, what about the primmy clothes? What about custom animations? A lot of baseline avatars have special hair, too… what about nude avatars? What about avatars wearing only lace underwear? What about a hundred other details that various worlds can’t or don’t want to support?
The transfer would be like trying to experience a movie while only able to see the closed captions, don’t you think?
This is putting aside the fact that some people will push for limits in order to insure interoperability…
I don’t think I like the way this is going… it’s making me very nervous.
It’s funny how the real world paranoia attributed to globalization and illuminati, translate to virtual worlds.
Interoperability [Raph’s Website]
IBM vill ha öppen standard i virtuella världar Läs mer här Open borders sought for virtual worlds (Reuters) och från enbloggsom bevakade “virtual worlds interoperability summit”.
Virtual World Interoperability” (Link) – Julian Lombardi “Open Avatar Announcement a Great Move” (Link) – Tim O’Reilly “IBM, Virtual Worlds and Standards – a roundup” (Link) – IBM’s Roo Reynolds“Interoperability” (Link)- Raph Koster “Linden Lab and IBM Collaboration on Virtual Standards” (Link) – Second Life Insider Posted in cyberspace | No Comments » October 4th, 2007
” stands for “Multi-User Dungeon”) – how much more difficult might this be when trying to convert 3D graphics and scripted objects? Raph Koster, CEO of Areae, was in attendance at the original meeting and hadconcerns of his own. No real market research has been done, which might mean that even if these interoperability standards were functional, they might not be something consumers want in the first place. Most of what was being proposed was based on assumptions that weren’t