Ken, since we're speaking Gilder here, I'll assume we're not talking about this month or next year necessarily, but over the visible horizon. That being the case...
...the Internet will sieve through any perforation it can find, or, more to the point where users are concerned, any perforation that they can economically afford.
Today's HFC CM architecture is substantial, very nice for today's applications, granted, when it's sized and maintained properly.
But beware the coming video over the net onslaught and other very large applications which are perched to greet tomorrows users in the time frame we are speaking about.
Shared pipes of the sort we have today, may, and probably will, find the Video/IP and other massively large flow applications overtaxing and fall apart, once adoption rates on shared segments grow, and the latter app types begin to multiply. Ditto for wireless in certain of its shared forms, where contention domains are used. This should not surprise anyone here, as I am sure most viewers here will recall when their first 14.4 was seen as a blessing from heaven at one time. And, that was not so very along ago on our increasingly accelerated obsoleting time line.
Scaleable links (either physical or virtual) with deterministic qualities will prevail in the end, I firmly believe. Again, today's CM is but a very good interim measure, until a more partitioned or oversized methodology - from a spectrum standpoint - for delivery, is established.
OR, until the overall b/w windows on modem interfaces and routers, not to mention the spectrum allocated on the medium to each host, itself, is increased, substantially.
And for the answer to this puzzle, I revert back to Gilder's FiberSphere, or to a [much] lesser extent, the exchange carriers' Full Service Area Network, or FSAN, model, which partially takes advantage of what they refer to as an All Optical Network Delivery Layer.
Preferrably, the Fibersphere, however, whatever that turns out to be, and whenever it arrives.
The line is blurring, incidentally, between what the classical FTTC model was where Cable was concerned, and where today's FTTCs for DSL are going. I wouldn't even classify tomorrow's dedicated options as one or the other right now, except for the distinction that one delivers program grade video, and the other... for whatever reasons, and there are a lot of them, doesn't.
But architecturally, we'll see a convergence between the two.
I don't think that we should have any doubts about this: The FSAN (which is one of those FTTCs that looks like the old Cable FTTC model) is on the way. Just give the incumbents some time, as they sense the build up to the convergence crescendo nearing, to establish their packaging and billing arrangements.
There are a lot of factors behind this line of thinking, not the least of which being how they will treat other players' services which ride over their facilities... witness today's FCC decisions concerning DSL services re: one of GTE's rollouts. Shades of FON's ION offering appear on the screen when we speak of this, as well.
Speaking of which, how do you all think last week's DSL ruling will impact other DSLs and the cable modem delivery model? Will it have any impact at all here?
See:
Message 6250146
Message 6250544
[Thanks, Stephen.]
Comments welcome.
Regards, Frank Coluccio |
