Ken, I should have added that in a pinch, it would actually be possible to force fit the universe of ISPs onto the cable channels now being used (or additional ones, for that matter) in some Rube Goldberg kind of way. Heck, it's done all the time at many SP routers in the dialup ISP regime. Note the results of that approach, however. The performance is usually so paltry that it serves as the primary impetus for going for cable modem access and DSL access, in the first place. As Dr. AHhaha has stated already in so many words on the ATHM thread (if my memory serves me correctly), opening up the cable modem pipes for the universe of ISPs at this time would be a Pyrrhic victory for them, at best.
GTE and AOL both recently grandstanded this capability, along with another local GTE-affiliated ISP, I believe, to demonstrate that open access is actually feasible. And, depending on your expectations, it is.
But that demo was in my opinion only a gambit, a major ploy, as far as I'm concerned. Either it was a very embarrassing ploy in the presence of the more enlightened, or someone at AOL had better start using their calculator.
I say this because neither AOL nor any other ISP would want to live with the eventual consequences if such an open access model were made available today under the current state of MSO implementations. If you take my last italicized qualifier into account, then it becomes clear that modifications are required, which are going to be both costly and time consuming, in order to upgrade the HFC model to a state where it could support everyone equally, and with the same feature sets as those currently available (and planned) by the cable industry.
[[I haven't heard any mention of how this would eventually be funded, nor have I heard of any reasonable time frames under which such an enhancement could begin to take place. Everyone, even T (for some inexplicable reason) is avoiding the specifics of these issues. Go figure. The only thing I can think of is that no one want to show their hand right now, keeping their options open, deferring to the possibility of yet additional negotiating ploys down the road. This, whether they can feasibly fulfill on any additional promises, or not. That's the way it's played today.]]
What this all means is this: Yes, it's possible to remove some of the architectural elements which represent barriers to multiple entry, those which are are essential to quality delivery, and open up the conduits for best effort delivery for Internet access by all.
But even if they (the MSOs) opened up four (4) 6 MHz video pipes, each one able to deliver on the average of an effective 10 to 20 Mb/s, as opposed to the usual single or dual channel approach currently being used, this would mean that at best their head ends could support less than 50 Mb/s for an entire serving area of potentially thousands of users per segment or branch. To put this into perspective, this would constitute a line capacity equal to a mere 33 ADSL lines, each operating at 1.5 Mb/s.
Think of it. Supporting access to multiple ISPs (possibly hundreds of them, depending on the locale) in a serving area of 000's of end users with the fire power that would be amenable for only 33 ILEC DSL users.
This would only be the case if an additional three video channels were offered up at the alter in a kind of AOLian sacrifice. Four video channels, in total, as opposed to the usual single or dual channel delivery schemes which are being contemplated, would be a very expensive price to pay from the video channel budgeting perspective, by any measure. Especially for a service that would, by definition, have to forfeit many of the already-in-place (and some, soon to be released) enhanced features afforded by the DOCSIS model. Comments and corrections are always welcome.
Regards, Frank Coluccio
ps - late edit: see the post just put up on the ATHM baord by Eric Davies re a San Francisco official's position on the ATT plan.
Message 10289837
The appointees and regs may have their hearts in the right places, but they are ignorant about the technical details involved, IMO. This, however, could be seen as a way to accelerate technical changes. But I wouldn't bet on this, tho. Once you take away the incentives, the laboratory gets very quiet. |
