The metaverse has a consistent definition - at least, if you read Snow Crash. It's just not what any company would actually want to bring to the market.
3D game worlds have value, open systems have value, but in completely opposite and contradictory ways. The value is in the entertainment of playing a well-built one, which means there has to be monetization and centralization. In other words, they have "production value". Open systems are outright allergic to that: while you can monetize, say, a webpage; there's no central WWW server, so there's no technical protection (DRM) against you just reuploading someone else's webpages. This doesn't matter so much for the web, because the cost to produce is low and the web has something the game worlds don't: "interconnection value".
A decade and change ago we had Second Life, a 3D game world built like an open system. This meant that instead of just giving you a bunch of levels with characters and game mechanics, they gave you a sandbox and told you to build whatever. They even released the Viewer source under GPL and at one point even promised to do the same with the Simulator server (which never panned out, although they were trialling grid interconnection with IBM at one point). You could build your own entirely independent Grid via OpenSim. At the same time, the game also had a centralized in-game economy based around selling content to build with. Linden Labs' embrace of openness, as incomplete as it was, wound up totally pissing off a lot of long-time users whose content could now be trivially copied, reuploaded, and resold using Free tools. They also had to deal with malicious Viewer forks, which ultimately wound up with them having an approval process for third-party Viewers, walking back some of that openness promise.
(I'm, of course, skipping over all of the other systems that have cropped up before and since - ActiveWorlds, VRChat, and so on. AFAIK none of them had metaverse pretensions to the same level as Second Life did.)
I've heard people claim Roblox is "the metaverse", and they might be close. However, it's far more of a closed system than even SL's open-viewer, closed-grid approach. They wouldn't for a second tolerate, say, someone writing a custom Roblox client or server. They do have an element of the metaverse that nobody likes to talk about, however: copyright infringement. Roblox has plenty of unlicensed fangames and other content that you wouldn't dream of being able to legally ship in a game world normally. This, to me, is one of the core properties of "metaverse-ness"; the ability for content to just exist in the game world without an associated license.
(Epic, you can use the term "metaverse" when I can design my own Fortnite skins and levels. Just being able to play on my PS4 with Xbox players is only a small part of it.)
Someone might bring up Minecraft at this point. This also has some of the aspects of a metaverse - notably, the block-building lets you go crazy with custom content, and it also allows custom skins. Well... mostly. Custom skins are only present in some versions of the game - notably, all Java versions, as well as Windows, iOS, and Android Bedrock. Other versions replace your skin with the default. There's also some licensed skins that are locked to specific versions of the game, that get replaced with the default skin as well. Cross-play lets you join other player's game sessions, but if you want levels to actually move between devices, you need to have a Realms subscription. There are servers, but custom servers are very new (like, in-alpha new) and not allowed on consoles (or at least on Switch - I don't own the Xbox or PS4 versions to test).
So that's another problem with the idea of a metaverse - the current system has way too many barriers in play to make it work. Remember how Sony made your Fortnite account a PS4 exclusive so you couldn't use your lootboxes on Switch? Imagine that, but 2000x worse, if someone actually tried to bring the metaverse to consoles.
A decade and change ago we had Second Life, a 3D game world built like an open system.
After many struggles, Second Life achieved social stability. The company has mostly a hands-off attitude and a minimal "governance" group of about half a dozen people. There are landlords and tenants, with grumbling on both sides, in more or less balance. Griefing is rare. The large-scale scams, such as a fake bank, are gone. Copying and uploading content happens occasionally but is not a huge problem. Land use is all over the place, but there are coherent communities. The main viewer is a third party open source system, and Open Simulator still chugs along, offering an option for people who want an environment under different ownership.
Ease of use is a serious problem in many dimensions. Just becoming a new users is frustrating. Upon entry, you go through a basic orientation and are dumped into a transit point full of jerks and scammers. It's like the Port Authority Bus Terminal in New York City. You've been dumped into a large city, and, other than some signs suggesting various destinations, you're on your own. It really is a virtual world, not a game. What you do in it is entirely up to you. The world is indifferent to you. This throws many new users.
Then they discover their default avatar sucks, that the clothing system is really complicated, and the good stuff costs money, or good free stuff has to be searched out.
If they make it to day 3, they'll start to get the place figured out. But the new user experience is harsh.
Most of the people making metaverse noises are not far enough along to reach these problems, let alone solve them.
Then they discover their default avatar sucks, that the clothing system is really complicated, and the good stuff costs money, or good free stuff has to be searched out.
In snow crash most of the population wanders around with a default avatar to save on cost
Here's Luca Grabacr's view of Second Life as the Metaverse.[1] She's the poet philosopher of Second Life. Her quiet, detailed videos of Second Life at its best show what the Metaverse can be.
This doesn't sound particularly different from when I was active on there a decade ago.
Does Linden Labs still have a weird habit of banning people running large "Internet meme"/chan-culture sims? (Actually, TBH, that almost seems forward-thinking nowadays...)
I don't play fortnite, but I get the sense it's closer to the vibe of Snowcrash's metaverse. Interoperability and freedom to create is cool, but stuff like Roblox doesn't strike me as a connected experience. Making games interoperable won't fit the bill unless it's about you, as the player, in the game; as opposed to you playing a character in the game. That's hard to do, because most games benefit enormously from having a structure and plot and characters. Even competitive FPS games to some extent.
You have to do stuff as yourself/your avatar. MMOs are close, but you never feel like you escape the world of the MMO. Fortnite has some purely social stuff like concerts and I feel like that's closer to the mark.
Epic released a very high quality character creator called MetaHuman recently. It's a great free resource to devs. But if they really want this to be a metaverse kind of thing they should be opening up unreal service apis to let players take their character models and port them into games. That would be a big step in the right direction.
HTML and HTTP are open standards, but they're not the metaverse. The "metaverse" is supposed to be an open standard for interoperable 3D game worlds.
You could build a 3D game world inside of a browser, and people have been doing that since the 90s. We've had multiple different open and proprietary technologies that allow this: VRML, Java applets, 3D Groove SX/GX, Shockwave 3D, Flash Stage3D, WebGL, and WebGPU. However, most game worlds built this way are less like "websites but in 3D" and more like Facebook. Yes, Facebook uses HTML, but it's not part of "the web" - most of it's content requires a login and can't be scraped or searched by external sites.
An actual "metaverse built on the open web" would be possible; in the same way that comment sections, trackbacks, Gravatar, and so on made blogging into a shared space decades ago. You would need standards for a whole bunch of problems unique to 3D game worlds, including presence, chat, avatars, asset inventories, and so on. All of these would have to also be integrated with existing content-management systems (WordPress, Drupal, etc), which would also have to serve up a viewer application that actually handled our "web metaverse" thing (in the same way that browsers don't need to know about comment sections, they just need to know how to submit a form). That in and of itself has significant UX problems: maybe, you go on one particular metaverse site, and suddenly you have entirely different player movement from the last site you were on, because it loaded a different viewer.
Comments
The metaverse has a consistent definition - at least, if you read Snow Crash. It's just not what any company would actually want to bring to the market.
3D game worlds have value, open systems have value, but in completely opposite and contradictory ways. The value is in the entertainment of playing a well-built one, which means there has to be monetization and centralization. In other words, they have "production value". Open systems are outright allergic to that: while you can monetize, say, a webpage; there's no central WWW server, so there's no technical protection (DRM) against you just reuploading someone else's webpages. This doesn't matter so much for the web, because the cost to produce is low and the web has something the game worlds don't: "interconnection value".
A decade and change ago we had Second Life, a 3D game world built like an open system. This meant that instead of just giving you a bunch of levels with characters and game mechanics, they gave you a sandbox and told you to build whatever. They even released the Viewer source under GPL and at one point even promised to do the same with the Simulator server (which never panned out, although they were trialling grid interconnection with IBM at one point). You could build your own entirely independent Grid via OpenSim. At the same time, the game also had a centralized in-game economy based around selling content to build with. Linden Labs' embrace of openness, as incomplete as it was, wound up totally pissing off a lot of long-time users whose content could now be trivially copied, reuploaded, and resold using Free tools. They also had to deal with malicious Viewer forks, which ultimately wound up with them having an approval process for third-party Viewers, walking back some of that openness promise.
(I'm, of course, skipping over all of the other systems that have cropped up before and since - ActiveWorlds, VRChat, and so on. AFAIK none of them had metaverse pretensions to the same level as Second Life did.)
I've heard people claim Roblox is "the metaverse", and they might be close. However, it's far more of a closed system than even SL's open-viewer, closed-grid approach. They wouldn't for a second tolerate, say, someone writing a custom Roblox client or server. They do have an element of the metaverse that nobody likes to talk about, however: copyright infringement. Roblox has plenty of unlicensed fangames and other content that you wouldn't dream of being able to legally ship in a game world normally. This, to me, is one of the core properties of "metaverse-ness"; the ability for content to just exist in the game world without an associated license.
(Epic, you can use the term "metaverse" when I can design my own Fortnite skins and levels. Just being able to play on my PS4 with Xbox players is only a small part of it.)
Someone might bring up Minecraft at this point. This also has some of the aspects of a metaverse - notably, the block-building lets you go crazy with custom content, and it also allows custom skins. Well... mostly. Custom skins are only present in some versions of the game - notably, all Java versions, as well as Windows, iOS, and Android Bedrock. Other versions replace your skin with the default. There's also some licensed skins that are locked to specific versions of the game, that get replaced with the default skin as well. Cross-play lets you join other player's game sessions, but if you want levels to actually move between devices, you need to have a Realms subscription. There are servers, but custom servers are very new (like, in-alpha new) and not allowed on consoles (or at least on Switch - I don't own the Xbox or PS4 versions to test).
So that's another problem with the idea of a metaverse - the current system has way too many barriers in play to make it work. Remember how Sony made your Fortnite account a PS4 exclusive so you couldn't use your lootboxes on Switch? Imagine that, but 2000x worse, if someone actually tried to bring the metaverse to consoles.
A decade and change ago we had Second Life, a 3D game world built like an open system.
After many struggles, Second Life achieved social stability. The company has mostly a hands-off attitude and a minimal "governance" group of about half a dozen people. There are landlords and tenants, with grumbling on both sides, in more or less balance. Griefing is rare. The large-scale scams, such as a fake bank, are gone. Copying and uploading content happens occasionally but is not a huge problem. Land use is all over the place, but there are coherent communities. The main viewer is a third party open source system, and Open Simulator still chugs along, offering an option for people who want an environment under different ownership.
Ease of use is a serious problem in many dimensions. Just becoming a new users is frustrating. Upon entry, you go through a basic orientation and are dumped into a transit point full of jerks and scammers. It's like the Port Authority Bus Terminal in New York City. You've been dumped into a large city, and, other than some signs suggesting various destinations, you're on your own. It really is a virtual world, not a game. What you do in it is entirely up to you. The world is indifferent to you. This throws many new users.
Then they discover their default avatar sucks, that the clothing system is really complicated, and the good stuff costs money, or good free stuff has to be searched out.
If they make it to day 3, they'll start to get the place figured out. But the new user experience is harsh.
Most of the people making metaverse noises are not far enough along to reach these problems, let alone solve them.
In snow crash most of the population wanders around with a default avatar to save on cost
Here's Luca Grabacr's view of Second Life as the Metaverse.[1] She's the poet philosopher of Second Life. Her quiet, detailed videos of Second Life at its best show what the Metaverse can be.
[1] https://www.youtube.com/watch?v=zdvGlwKPVk4
This doesn't sound particularly different from when I was active on there a decade ago.
Does Linden Labs still have a weird habit of banning people running large "Internet meme"/chan-culture sims? (Actually, TBH, that almost seems forward-thinking nowadays...)
I don't play fortnite, but I get the sense it's closer to the vibe of Snowcrash's metaverse. Interoperability and freedom to create is cool, but stuff like Roblox doesn't strike me as a connected experience. Making games interoperable won't fit the bill unless it's about you, as the player, in the game; as opposed to you playing a character in the game. That's hard to do, because most games benefit enormously from having a structure and plot and characters. Even competitive FPS games to some extent.
You have to do stuff as yourself/your avatar. MMOs are close, but you never feel like you escape the world of the MMO. Fortnite has some purely social stuff like concerts and I feel like that's closer to the mark.
Epic released a very high quality character creator called MetaHuman recently. It's a great free resource to devs. But if they really want this to be a metaverse kind of thing they should be opening up unreal service apis to let players take their character models and port them into games. That would be a big step in the right direction.
I'm really not convinced that what we're communicating on isn't already the metaverse
HTML and HTTP are open standards, but they're not the metaverse. The "metaverse" is supposed to be an open standard for interoperable 3D game worlds.
You could build a 3D game world inside of a browser, and people have been doing that since the 90s. We've had multiple different open and proprietary technologies that allow this: VRML, Java applets, 3D Groove SX/GX, Shockwave 3D, Flash Stage3D, WebGL, and WebGPU. However, most game worlds built this way are less like "websites but in 3D" and more like Facebook. Yes, Facebook uses HTML, but it's not part of "the web" - most of it's content requires a login and can't be scraped or searched by external sites.
An actual "metaverse built on the open web" would be possible; in the same way that comment sections, trackbacks, Gravatar, and so on made blogging into a shared space decades ago. You would need standards for a whole bunch of problems unique to 3D game worlds, including presence, chat, avatars, asset inventories, and so on. All of these would have to also be integrated with existing content-management systems (WordPress, Drupal, etc), which would also have to serve up a viewer application that actually handled our "web metaverse" thing (in the same way that browsers don't need to know about comment sections, they just need to know how to submit a form). That in and of itself has significant UX problems: maybe, you go on one particular metaverse site, and suddenly you have entirely different player movement from the last site you were on, because it loaded a different viewer.