Thanks. This last statement in the first of those two articles caught my attention:
"I personally believe that it's not one vs. the other," he says. "They are all providing the end user with the benefit of this killer application -- the Internet. Whatever technology it is, the content enrichment of the Internet will continue to drive broadband deployments."
One has to wonder about the all-importance of the Internet going into the photonic future. Right now the Internet serves as the great economizer, where resources and routes are short in supply if everyone were to build their own routes. We have shared utilization on the 'Net, instead. Going forward, however, I can see where end user control of their own lambdas (or, more specifically, of carrier lambdas which are dynamically leased much the same way that IP addresses are leased today) will be made to target remote nodes and distant end points more easily than before. Where OIF addresses the macro model, some end user-specific framework will be developed here, as well.
Some would argue that we already have this in the ways of Layer 3 switching and Layer 2 tunneling protocols. So be it. But these primarily electronic-, IETF- related protocols have to first be adapted to the optical model, before we can say that they already exist. This is being done already under various fora, between the core and the edge. But not to the end point, per se, from what I can see.
Setting up temporary paths and virtual circuits that bypass the greater 'net, in other words. Just some thoughts... comments and corrections welcome. |
