My first thought was "Wow, somebody made that work!" After reading their site, it's clear they have a long, long way to go.

In the early days of There, Inc., I was recruited by them, and was in on some early design discussions. They wanted to build a planet-sized 3D virtual world with physical simulation, with all the scaling problems that implies. My comment was "Now let us all join hands around the world", which produced groans. (Doing physical simulation across a cell boundary is hard enough, but if the users can force you to have to do it across N cells, you have a real problem.)

For that system (and for Second Life) the assumption was that all the servers were in the same data center, with low inter-server lag. Trying to do this with home servers and lag in the 100ms and up range is going to suck. However, a distributed system where all sites are hosted in data centers on the same continent with low-latency connections might work.

Now they have to deal with the security problems, the delay problems, etc. They also need a really big MMORPG to exercise the system. Something like a planet-sized version of Minecraft would be a good test. The rules are simple and users create the content. A tougher test would be to create a simple, but huge, driving game, using Open Street Map, "seamless.gov", and topo data. Then load it up with automated traffic and let people drive in it.

I really do feel the need to ask, these days: am I the only technologist completely uninterested in transforming real life into Snow Crash? As far as massive-world 3D games go, I still see more artistry in the Assassin's Creed series (for all its flaws) or even something more fantastic like Pokemon Snap (safari games are the idea that stay with me for VR implementation: make something that does not already exist rather than copying meatspace) than in these weird attempts to create a "virtual reality" or even just large-scale game world modeled off... Times Square.

Not only can I already visit Times Square in real life, I choose not to. Ditto on the more branded, corporatized, ad-infested portions of the internet (ie: the "downtown internet" approved by official, polite society).

Many scifi novels (insofar as Snow Crash can even be called "fiction" rather than "prediction" these days :-() are written to make fun of wacky technologists and our just-somewhat-off-target ideas, you know.

Now, to end on a zog note rather than zig or zag, here's an idea: someone figure out the massive-game or VR equivalent of 4chan. There's something people will actually want to use. (For all the official disapproval from all sectors, 4chan and 2chan are massively popular and influential. They are offering something users want: anonymity, obscurity, simplicity, and freedom!) EDIT: http://kazerad.tumblr.com/post/96020280368/faceless-together

The more important test is not the technological one, but if the idea is useful at all. So far 3D worlds haven't proven themselves as necessary or something that leaps forward communication. It provides all kinds of constraints with few upsides. Just because the world is in 3d for my normal life doesn't mean that constraint is natural or useful by extension -- when I read I absorb language and abstract thoughts mixed with imagination without any "real world" basis. 3D avatars might look cool but they also visually become the hero when that might be exactly what we want to get away from in a community context -- the potential for new forms of representation that dont resemble the physical world. Second life was predicted by some to have already taken over all of business world, and thus far it had the most uptick and chance of all of these systems. Yet am I reading HN via some dragon emitting nerdy fireballs? No I have a simple web page with the ability to hyperlink me straight to interesting content that is more text typically. And it's working way better in this form than a 3D world could because the concept doesn't make sense to add that extra graphical dimension and navigational constraint.

Having been in on two previous attempts, let me state my wry amusement at this go-around :-)

On the back is gibberish explaining how he may be reached: a telephone number. A universal voice phone locator code. A P.O. box. His address on half a dozen electronic communications nets. And an address in the Metaverse.

Sounds like High Fidelity: https://highfidelity.io

(No affiliation other than having tried it a couple months ago and written a little bit of code for it.)

If you try to go extreme on all possible requirements simultaneously, the chance of failure approaches one hundred percent. The key to success is to pick one or two areas where you want to go beyond state-of-the-art, while staying within it in other areas.

I strongly recommend deleting the requirement about seamless visualization across servers. Of all the items on the list, that strikes me as adding the most difficulty for the least value.

Can someone explain what this is for? I watched the videos, and I still don't get it.

And no, I haven't read Snow Crash. I shouldn't have to read a novel to get excited about a project...

Calling this an "operating system" just seems like misdirection from the actually cool problem of visualizing distributed systems. I mean the whole thing about the Metaverse is that it's simply a vivid realization of what's already happening on the internet --- it's not some extra layer on top.

This technology is pretty impressive, and I get the marketing speak that will appeal to certain folks, but I'd rather call it what it is --- an impressive, distributed, rendering engine.

IDK, maybe I'm thinking about this all wrong.

Anyone know if the source code is available? Their post says it's an open source project -- would be cool to explore.

This looks like it would be very good tech for MMO games.

Well, there is are several metric tons of examples out there.. Hard to resist coming up with my own ideas though since this is such an interesting topic.

Maybe we don't have to have the most advanced open source 3d networking game engine out there as a protocol.

Maybe we can start by building off of Named Data Networking (NDN) http://named-data.net/doc/NDN-TLV/current/

Would be nice to have everything including content represented in Manchester OWL syntax but that's probably not realistic.

So perhaps use the NDN protocol with TLV encoding, NDN name format/URI scheme, order. Use the interest packets for subscribing to rooms in the metaverse. Then exchange NDN Data Packets with updates to object positions etc.

Maybe for data contents use something with YAML for a change from XML.

    name/smith/john50/homeworld

    base: org/sim
    libs: dim/1, objs/1, phys/1
      dimension: real
        - terrain:
            heightmap: maps/hills.png
        - objects:
          - house:
            windows: 5
            type: simple
            bedrooms: 3
          - avatar:
            model: models/man.dae
            name: name/smith/john50/me
            pose: walk 2
            position: [50, 50, 2]
            velocity: [0.5, 1.0, 0.1]
            owner: name/smith/john50
          - portal:
              world: name/wallace/liz10/homeworld
              type: simplecircle
              position: [123, 435, 123]
              dimension: real
          - webview:
              name: John's Smart TV
              url: http://en.wikipedia.org/wiki/Metaverse


    org/sim/objs/1

    base: org/sim
    dimension: real
      - procedures:
        room:
          - for: [1, walls]
              - makegeom: [cube, 1.0, 0.5, 1.0, 0.0, 0.0, 0.0]          
        house:
          - for: [b, 1, bedrooms]
              - room:
                  walls: 4
          - for: [b, 1, windows]
              - cutgeom: [quad, 0.25, 0.25, 0.25, b, 0, 0]             
   
         
    name/smith/john50/update102
    timestamp: xxxxxxxxxxxx
    which: name/smith/john50/me
    update:
      operation: impulse
      vector: [ 0.25, 1.0, 0.0 ]
      startstate:
        velocity: [ 0.5, 0.0, 0.0 ]

The idea with the "procedures:" thing is to take advantage of YAML's flexibility in order to create a language independent representation of an algorithm for doing simple procedural generation. It would be designed in such a way as to make it easy to generate code from the data tree in different languages like C, C++, Nim(rod), Java, Python, whatever. That would be compiled on the fly and loaded as a sort of plugin.

Then rather than doing simple state updates, it uses operational transformations like saying "at time A force B was applied to this object and its velocity before that was C".

So my idea uses an emerging standard (NDN) for efficiently distributing information and updates about the metaverse, an easy to read standard in YAML for describing the details, includes the concept of portals linking between worlds (you gotta have portals, they are one of the coolest things about the metaverse), includes the idea of embedded webviews (too much awesome stuff on the web to leave that out of the metaverse browser), a YAML-format for encoding algorithms that can be easily translated to common languages and compiled on the fly, and operational transformations for informing other clients about operations on the world state.

Again? :)

It does look quite cool though.

It's another CLR monstrosity (see the code on their Facebook page)?

Why!!!???

I despise your use of YAML instead of JSON, but I applaud your suggestion of a flexible procedural markup language.

It allows (theoretically) for us to be sharing the same space, but I see a modern office while you see a Japanese dojo. Effectively, Spatial Style Sheets, right?

The big problem (one we still run in to everywhere else) is that people are going to want to be guaranteed that the dragon dildo they proudly display in their 'verse is seen the same by every visitor, and that's only accomplished by low-level data such as triangle meshes, UVs, and images.

Still, a very intriguing thought.

Not letting me edit it anymore but wanted to fix something:

  base: org/sim
  dimension: real
  procedures:
    room:
      - for: 
          range: [w, 1, walls]
          block:
            - makegeom: [cube, 1.0, 0.5, 1.0, 0.0, 0.0, 0.0]          
    house:
      - for: 
          range: [b, 1, bedrooms]
          block: 
          - room:
            walls: 4
      - for: 
          range: [b, 1, windows]
          block:
          - cutgeom: [quad, 0.25, 0.25, 0.25, b, 0, 0]