While server virtualization is a great advantage for anyone who wants to make their systems run more efficiently, it presents a long term challenge to server and component manufacturers. While the take up of virtualization will boost sales as customers purchase new platforms to host technology, inevitably in the longer term the mathematics of server consolidation points to fewer units being sold. The challenge for the computer industry to avoid this shortfall is to find new applications to induce greater product demand.
I attended the Intel Developer Forum (IDF) in Taiwan this week, and in a keynote presentation Intel Senior Fellow and Communications Technology Lab Director Kevin Kahn gave a clue about Intel's vision of a future application that will require increased workloads for servers, clients and the infrastructure in between. In his keynote Dr Kahn spoke passionately about the rise of the 3D Internet. His message was that virtual worlds like Second Life and World of Warcraft represent the beginnings of a 3D Internet experience that over time is bound to become more ubiquitous and will transform much of today's 2D web into a truly interactive, user driven experience presented in a fully rendered 3D virtual reality.
The virtual worlds of today, with Second Life as perhaps the most well known example, are certainly proving popular, with an estimated 60 million users worldwide in 2007, according to figures from Wired Magazine cited by Dr Kahn. In a short time they have become a platform not only for social interaction, but also for commerce and information dissemination. But in order for such a rich graphical environment to become standard on every desktop, an increased computational load will have to be borne by servers, desktops and networks alike. And it just so happens that Intel's processors, chipsets and other components can be found in the majority of products in these three categories.
In order for a rendered 3D experience to be ubiquitous, server, clients and networks alike will need to be able to do a lot more work than they do today. And they do a lot of work today. Also, universal broadband access will need to be vastly improved over today's standards in many parts of the world. Client PCs will all need to have the computational power necessary to do the 3D rendering, which is a resource intensive workload as any hard core gamer or graphic designer will tell you. This should also ensure that the personal computer stays with us for some time yet; thin client technology is not yet suited for 3D rendering across such a widely distributed network as the Internet.
Another key message from IDF was Intel's vision of mobile devices having the same internet experience as desktop PCs. If we are to combine this with Dr Kahn's 3D web, then the logical extension of this thinking is that future internet users will be engaged in a beautifully rendered 3D Internet on their mobile devices. That would be something to behold.
The web of today, apart from being a primarily text based medium, is still largely driven by server generated static content. Proponents of a transformed internet, including Dr Kahn, are calling for the balance to be shifted to more interaction and for much more user generated content. This paradigm, which has driven the Web 2.0 concept, will be integral to the next generation 3D web. Further to this, content providers will need to re-think how their services can most effectively be represented in a 3D virtual world. An online bookstore could let shoppers stroll amongst virtual shelves, or in a virtual auction room bidders' avatars could stare menacingly at each other. However, maybe not all Web content would be better delivered in this way. For example I would probably prefer to see my banking details in 2D text rather than try and converse with a virtual teller.
So is the concept of a 3D Internet inherently superior to that of a 2D one? While I think that a 3D virtual world would certainly be more immersive and engaging that today's experience, in my opinion many of the things we do online today would not necessarily be improved by being presented in 3D. But the laws of supply and demand dictate that if users want a 3D Internet, then someone will find a way of providing one. And that demand would be good news indeed for hardware manufacturers like Intel in the post-virtualization world.
The best use of 3D may _not_ be to simply emulate ordinary or even extraordinary reality. Rather 3D may be most useful for all kinds of data visualization e.g. visualizing cyberspace or other large data sets.
Mail me directly for more details.
BobLQ
Posted by: Robert La Quey | April 08, 2008 at 01:52 PM