Apr 19, 2010

Defining the Metaverse - Revisited

"It's a construct of the mind, brought to fruition by our dreams and desires. A collaborative hallucination."

- William Burns


In 2006, I was a contributor to the Solipsis Decentralized Metaverse project on an intellectual basis, and as to how such a system would fulfill a greater definition of a true Metaverse system. Later, in 2007, Solipsis did a presentation which included the very definitions and explanations I had contributed in 2006 concerning the true nature of a full Metaverse, and in 2008 they gave an updated presentation at IEEE even going so far as to directly quote me for a full page with the likes of Philip Rosedale and others in the industry.

The ANR-RIAM Solipsis presentation given in 2007 can be found here:
http://bots.netofpeers.net/riam_solipsis/LOT6/slides/ANR-RIAM_Solipsis_ColloqueSTIC_2007.pdf


The 2008 IEE Presentation version can be found here:
http://www.andromedaunderground.com/darian/Solipsis - A Decentralized Architecture for Virtual Environments.pdf

In the latter, pages 5 and 30 are of interest for this post. Page 5 on the latter is the definition of (Meta)Data, MetaWorld, MetaGalaxy, and finally Metaverse. These definitions, and more importantly the line of succession, are from the following article written in 2006 (and reposted here with some updates). Page 30 of the Solipsis presentation is a quotation from me concerning the actual nature of a full Metaverse. the reason I make a big deal of this today in 2010 is because years later, companies like BlueMars, Linden Lab etc are still ignoring the original points from the Solipsis research (and research points I've made since 1999, and have been made since 1990 by Morningstar and Farmer).

The original article can be found here:
http://wburns.blogspot.com/2006/09/its-construct-of-mind-brought-to.html

Definition

An electronic 3D representation of a real world or fictional environment, populated by real people and programs (known as bots or daemons). Within such an environment it is possible not only to interact with the scenery as you would in real life, it is also possible to interact with other system users in 3D real time.

The Metaverse, as defined by the book Snow Crash (Neal Stephenson), is representative of a highly evolved Internet system in which the standard 2D interface for the GUI is all but replaced by a lush three dimensional interactive system with approximately sixty million simultaneous users (referred to as "in world").

What Classifies a Metaverse?

A common misconception concerning the fictional Metaverse was that it consisted only of a single world. It is my personal observation that the world known as "The Street" was in fact just one of many worlds within the Metaverse, though the most popular as it was the default starting world for the Metaverse within the incarnation of the book Snow Crash. There were obviously other literature concerning the fictional Metaverse structure and what it's purpose was, starting with the father of the cyberpunk genre himself, William Gibson, whose short story Johnny Mnemonic (1981) was popularized as a movie in 1995 and was actually just a short story taking place within the same time line and space as a larger set of stories.

The idea here is that, the William Gibson idea of the Metaverse differed greatly from Neil Stephenson, and while the ideas of Cyberpunk differed across authors there was still an underlying element of a Metaverse structure. Each of these ideas of a Metaverse differed, but they encompassed different aspects of the vision. As a result, I get the understanding that there wasn't exactly a single, giant, virtual planet like in Snow Crash, but a wide decentralized system with centralized gateway servers to handle it. In this aspect "The Street" from Snow Crash really represents a single virtual world among a decentralized Metaverse, but something like The Street would have been managed as a gateway to the Metaverse, and thus most popular.

Due to these fictional systems arising within the same culture, I would safely say that in a real world environment it would make sense to incorporate all of these aspects as a total Metaverse concept.

Another argument in favor of this is that the very meaning of Metaverse itself demands there be more than a single world approach; as we see with the terminology of (Meta)Data, MetaWorld and MetaGalaxy and finally Metaverse wherein it defines a MetaData Universe. Indeed the very term Metaverse derives from the word MetaData, and so would follow in this correct line of logic - MetaData, MetaWorld, MetaGalaxy, and Metaverse when outlining the progression of the virtual space each term would inhabit. Also of note is that Neil Stephenson when writing Snow Crash was influenced by the Unix operating system as his view of computers of the day - hence we see a lot of Unix type references in his book, so I couldn't imagine that he meant Metaphysical Universe instead of a Unix styled Metadata Universe. The first being a more holistic approach while the latter sticking with the computer and Unix styled theme.

I am highly confused at the twisting and misrepresenting of the term Metaverse by companies such as Second Life in which they are trying to use the old "bait and switch" approach to the word. Instead of its original meaning of MetaData Universe, they instead imply that it's meaning derives from Metaphysical Universe.

While this may work for promotion and advertising to the public, it does a grave disservice to the word itself and also misleads the public into believing that the product (Second Life) is indeed a full scale Metaverse, when instead it qualifies as only two MetaWorlds (Adult SL and Child SL). Even with their misleading advertising, they still do not qualify as a Universe structure.

When probing further into the subject, one would also find that There.com was also misclassified as a Metaverse by the public (more than likely due to the misrepresentation of the word to begin with), while in fact even There.com only classified as a single MetaWorld.

To gain a better insight on these classifications, I will present a step by step example of each:

  • Metadata: A single virtual object (standalone VRML display of a single 3D object) or simply 2D websites. [picture]
  • MetaWorld: Perceived single location space (world). Either single user or massive multi-user. [single user] [multiuser]
  • MetaGalaxy: A group of MetaWorlds, more than likely massive multiuser, interconnected (not standalone) - ActiveWorlds
  • Metaverse: Multiple MetaGalaxy systems linked into a perceived Virtual Universe, although not existing on a central server, (Morningstar and Farmer: Lessons Learned From Habitat - Decentralization) [no system yet qualifies]
Since I have already classified both Second Life and There.com as a MetaWorld structure, I move onto another of the contenders in this field: Active Worlds.

Active Worlds as a platform is the closest to an actual Metaverse classification that currently exists today. While not entirely a Metaverse, there is a much higher probability that it can become one if properly implemented. As of this writing, the Active Worlds Browser (AW Browser) is currently available in server configurations that reach only as high as a MetaGalaxy (Multiple Worlds, Interconnected, Centralized). This despite their server classifications of Sol, Galaxy and Universe (which again are incorrectly labeled based against proper classifications).

The Active Worlds server comes in a Sol configuration, which by all standards is simply the same as a MetaWorld configuration and no interconnectedness to other systems. Sol servers are small Metaworld configurations with a stand alone browser to access it (usually also very small user access limits as well).

The Galaxy configuration is essentially the same as the Sol configuration in that it is a single MetaWorld configuration though the differences are that this configuration facilitates much larger virtual land and simultaneous user limit options (whereas the Sol configuration does not). The misleading aspect here is that this is not actually a MetaGalaxy (or a galaxy by proper definition which would include many planets and even stars). Instead the so-called "Galaxy" from Active Worlds is nothing more than a larger MetaWorld configuration than a Sol.

Which brings me to the "Universe" configuration. By definition a Metaverse should be many Galaxy systems linked together as a decentralized network of Metaworlds. What we find here with the Active Worlds Universe server is nothing more than an actual MetaGalaxy configuration, regardless of how far it can expand or how many worlds can exist within it. It is still a centralized system containing many worlds (and not decentralized with many metagalaxies).

So by proper definition of the terms in use, the Sol configuration for Active Worlds should be merged with the Galaxy server option and renamed MetaWorld Server, while the Universe configuration should be renamed MetaGalaxy - leaving open the Metaverse option for actual implementation once again, though not an option for purchase because a Metaverse constitutes decentralized massive multi-metagalaxies linked through a single interface.

What would a Metaverse look like?

Under current configurations of an Active Worlds system, there lies the innate ability to evolve into something much larger. While it is considered physically impossible to link multiple worlds or galaxy structures within the Second Life platform (and even There.com) whether this is by limitations of software, hardware or by business decision, with the Active Worlds system there is a latent ability to evolve into a full scale Metaverse.

There already exists many MetaGalaxy configurations for Active Worlds (Sol, Galaxy and Universes by their definition) and all are currently stand-alone in nature. A full scale Metaverse would evolve should the client browser be given the ability to traverse all of these independant MetaWorlds and MetaGalaxies to create a perceived Metaverse space.

This follows well the writings of Morningstar and Farmer (Lessons Learned from Lucasfilm's Habitat) where they explain that a decentralized system is a must due to bandwidth constraints. The idea here is that while it is feasible to house a great multitude of virtual inhabitants within a single perceived environment, the cost to user ratio skyrockets as these figures reach the hundreds of thousands and even millions of users. Though within a decentralized system (many linked MetaGalaxies) the perceived space of all of the worlds and galaxies metaphorically creates a seamlessly integrated and single Metaverse structure.

The first area to investigate involves the elimination of the centralized backend. The backend is a communications and processing bottleneck that will not withstand growth above too large a size. While we can support tens of thousands of users with this model, it is not really feasible to support millions. Making the system fully distributed, however, requires solving a number of difficult problems. The most significant of these is the prevention of cheating. Obviously, the owner of the network node that implements some part of the world has an incentive to tilt things in his favor. We think that this problem can be addressed by secure operating system technologies based on public-key cryptographic techniques (Rivest, Shamir and Adelman, 1978; Miller et al, 1987). - Morningstar and Farmer: Lessons Learned From Lucasfilm's Habitat


Admittedly the Active Worlds Universe configuration is supposed to handle a mere 65000 simultaneous users before reaching a need to essentially "clone" the server for a mirrored version capable of handling, again, only another 65000 simultaneous users. This configuration is for the entire server itself, and is split among the amount of worlds currently within the server. The main drawback here is that under current configurations for Active Worlds, both cloned servers would require a separate install of essentially the same browser in order to visit either. Two copies of the same program for two systems that are essentially the same. Imagine trying to visit the Metaverse right now under these constraints - each section would be a separate download and install of essentially the same browser.

This idea is like downloading Firefox and finding out you can only visit 500 websites, and to add the functionality of only another 500 websites you must download and install a separate copy of the same program (continuously). This concept is the exact opposite of what the Internet itself was meant to be.

Under this configuration, it would be impossible to host ten very popular worlds within a server effectively.

Popular websites play host to millions of users per hour, and under this idea, if one were to implement a MetaWorld using seemingly the most capable system (Active Worlds) this MetaWorld would essentially be completely obliviated within the first 15 minutes (or less).

The idea of allowing the client to traverse all servers (Sol, Galaxy and Universe) across the globe makes a Galaxy (in active worlds terms) less important, as a Metaworld configuration (currently known as a Sol/Galaxy) could be configured to as much as 65000 users for that single Metaworld, while access to this Metaworld could be accomplished through normal web means (or even teleport linked in world).

This would create a nearly seamless (or if done correctly, seamless) transition across many multitudes of server configurations across the real world, and thus creating a real MetaData Universe wherein the servers are not centralized, and the perceived space for the user is infinite.

What is stopping them?

While much of the hardware and software exists to truly create a full Metaverse system, the vast majority of companies involved with systems capable of doing so are keeping their programs closed to expansion, and thus hindering the greater potential that their software systems could produce.

Some "homebrew" applications are popping up across the Internet, interestingly enough, in places like Sourceforge.net in order to attempt to tackle this problem. Unfortunately no homebrew software campaign has managed to sucessfully capture the ease of use and power of a real metaverse system.

Points of reference - "Snow Crash" : Neal Stephenson "Diamond Age" : Neal :Stephenson

Internet 2.0

The future of the Internet itself has yet to be determined, but one thing is for sure - it includes ever increasing bandwidth options. While the fiber optic internet of fiction has yet to be realized, there are ongoing plans to utilize high bandwidth systems for our immediate and distant future.

It's a matter of mathematics to say that technology in most respects will double every 18 months via Moore's Law. This is not only limited to the speed of processors, but indeed extends into every facet of technology connected to the CPU. Years ago, it was stated that it would be impossible to imagine the consumer needing anything more than 640k of RAM for all applications, and yet we are currently using gigabytes of RAM for the latest software, operating systems and games.

So too, the demand for bandwidth must follow as our online experience matures and the rate of data transfer increases. Where once we had simple HTML text websites with a few hyperlinks, we now have media rich sites utilizing streaming audio and video. But these too will mature as the bandwidth increases; there is only so much a person can experience via video and audio.

So what form will our future Internet take? The obvious answer for this is a Metaverse, or more aptly, Multiple Parallel MetaGalaxies. With broadband rapidly expanding from all directions (cable broadband is now up to 6 and 8Mb/sec) we are expected to double the information transfer capacity accordingly. It's not viable information that we will experience in the future, but instead passive information - wherein objects representing concepts (metaobjects) will be downloaded on demand and instead of a flat webpage, we will experience an entire online virtual space, collaboratively.

The future of the Internet is indeed broadband. And not the 3Mb/sec you see today, but upwards of 90Mb/sec and higher. Of course, current hardware will not support the onslaught of data transmission, and nobody expects it to. Just as the bandwidth increases, it does so in response to ever increasing storage capacities, cpu seeds, GPU (graphics processing units) and onboard RAM requirements. see also Abilene Backbone [Internet2]

Some companies today are in possession of systems that can easily qualify as a Metaverse (with some minor modification), though not nearly as high definition as the ones described in fictional literature. Everything has a start, and even the most advanced technology meta-morphs into something truly amazing every year it exists. What we have is the beginning of the Metaverse as we know it. Conceived in the late 1980's (Habitat), advanced in 1995 (Active Worlds), and finally gaining acceptance in 2006 with companies like Active Worlds and SecondLife.

As well it should. While the real world grasps the meaning of what a true Metaverse is, and how it should effect their life - I write here the actual definitions and explanations of what they are and how they should be implemented. In the end, a Metaverse structure is nothing more than an entirely new media form.

What this boils down to is that the content contained within even a MetaWorld (no matter how well linked into a universe structure) can be described simply as a collaborative 3D Website. A Metaworld is nothing more than an advanced, collaborative, real time 3D website. It is an entirely different media form altogether, but understanding it in this light helps a great deal.

A Metaverse can be described as a 3D Internet, plain and simple, wherein the content contained within can be seen as a 3D Website or Webpage. With this in mind, even an experienced technology expert (right down to the average person) should be able to easily grasp what this is all about. It is wholly incorrect to classify such a system as a Game, or simply and Educational tool, or even just a chat-room. Just like the Internet now - a Metaverse can be everything you see on the Internet today, in a wholly new and remarkable form of media.

The browser chosen for this writing was due to a few discerning factors:

  • The Active Worlds platform is massive multiuser.
  • The browser can be embedded into a webpage
  • The environment is easily scriptable and editable.
  • Multiple "metaworld" capability

While other companies such as Secondlife indeed have impressive systems, they are missing half of those requirements to qualify as a true Metaverse. They are both categorized as a single Metaworld, despite what they may tell the media and are incapable (or unwilling) to create multiple Metaworlds in a structure linked together in order to create a Metaverse.

Also listed, Blaxxun Interactive, was supposedly trying to create the Metaverse using the VRML standard and adding it's own back-end client to add support for massive multiuser - but fell short of all of the requirements by not including scriptable and editable environments. (VRML is not a live language, but instead a precompiled world standard).

Possible Issues In The Future

Since 2006, the issue of piracy has become a big talking point and ISPs in the United States (and elsewhere) have been pushing to implement a capped bandwidth structure for their customers as a result. Instead of upgrading their network capacity to alleviate the network congestion, many ISPs (including larger names with a stranglehold on their markets) are quietly imposing a bandwidth limit on their users which would drastically hurt the Metaverse going forward. This is shameful behavior at best, and until this issue is taken care of, it will effectively impede the creation of effective Metaverse systems going forward.

What this boils down to is the need for developers to design better and more efficient systems which require drastically less bandwidth while retaining the high quality going forward. this seems to be a contradictory requirement, and even impossible at first glance, but the main purpose of Andromeda Underground is to research such methods, and after a few years doing so we've come to the conclusion that it is completely possible to accomplish. Not just possible, but we've figured out exactly how to accomplish it.

And so, while companies employ Cloud Servers to pre-calculate the scenes (BlueMars), or Simulators which chew up bandwidth centrally (SecondLife), we continue to look forward to the day when a true metaverse will be created while enjoying the follies and misguided actions of current generation systems.

William Burns

Project Leader for Andromeda Media Group

0 Comments:

Post a Comment