OT - Technical: Open ISP Access Arrangements Over Cable
Jay, an interesting set of reads on the open cable access issue. Thanks. I continue to find this subject intriguing from a growing number of perspectives.
There is the obvious (to me, at least) prospect that regulators would try to impose the same kinds of openness rules in the cable sector as they do in the wireline telco/carrier sector.
But this doesn't appear to be the case, as Kennard continues to dole out double standards, despite the fact that equivalent types of services will be offered... and the fact that the same kinds of monopoly status causality exists from one model to the other.
Where the latter is untrue today, it will readily become true when telco-like services are eventually enabled on the cable systems. As I've stated before, cable may actually become the provider (carrier) of last resort, in some locales. Then what?
The MSO/ATHM relationship is not unique in every sense, since there are parallels elsewhere. RCNC/Erols-Ultranet comes to mind, along with a couple of others, where there are tight operations-level bonds and exclusivity arrangements between the facilities-based cable provider and their subsidiary (or subsidiary-like) 'net partner/department.
So, what happens to TCI will inevitably have some direct, and in some cases indirect, impact on the fate of other operators in this space, as well.
Through RCN's acquisition of Erols in Washington D.C, and Ultarnet in Boston, they now claim to be the nation's seventh largest ISP, in addition to their growing CableTV presence, and their support of any-distance voice services, as a bona fide CLEC. Haven't we seen this model emerging somewhere else, recently?
Assuming for a moment that the FCC is overturned on this issue, and AOL and the others get their wish, how does one go about implementing this open Internet access arrangement over cable facilities? Several options are available, but none of them would be elegant right now.
There's the matter of physics, in the face of a limited set of potential solutions to an unanticipated set of demands. Assuming that each ISP wants to have its own slice of the spectrum, or at least some control over its channel allocations in such a model, has anyone here seen any proposed methods as to how this would be achieved? Several options emerge as plausible, however intractable they may be.
If we ignore the obvious conflicts of interest for a moment, the cable operator could act as a kind of gatekeeper, and create a front end for all ISPs in a non-proprietary and open way. Something on the order that Compuserve or AOL now permits,while facilitating access the public Internet.
Hey, here's a novel idea: Why not allow ATHM and Erols to do this (as they have been all along, in fact), and just go on with Business As Usual? There are some obvious, and some not-so-obvious, reasons why AOL and the others would prefer otherwise.
How about creating a second and third slice of the spectrum to be used by the second and third takers (ISPs) who want in? As you can see, this wouldn't scale very well with thousands of ISPs out who would jump at the chance to join in, before it begins to eat into and totally consume the regular program TV channel bandwidth allocations.
But assuming that this additional channel option is seriously considered, and it could only be done on a limited basis, it would appear then that some allocations of spectrum could conceivably be auctioned off, or lotteried in some manner which is deemed to be fair. I cough at this, ah hem, but then again, dumber things have taken place when the regs have stuck their collective noses into the stew.
Another option is to broaden the existing (singular) upstream and downstream channel allocations incrementally, allowing all ISPs to then share a single channel pair. [This would be no trivial task in itself, since the die has already been cast for many standards initiatives to use the parameters that are now in place.] Heck, that's equally bizarre.
But it makes more sense than allocating the entire breadth of the cable spectrum to individual Internet access arrangements, in the process forcing lower capacity constraints on the likes of ESPN and CNN.
Let's play this option out. If all ISPs are utilizing the same spectrum with discrete logical feeds at the network layer (Layer 3), then how do they administer traffic flows and classes of service over a common infrastructure in the face of unknown traffic profiles being supported by any of the other competing feeds?
Along this line of logic, what kind of head end arrangement would be required to support this meeting of the ISPs in a common routing environment? In other words, how do ISPs effectively "co-locate," either physically or logically, within the confines of a head end?
While this is actually a doable configuration, it is at the same time extremely untenable, given the volumetric constraints that now exist in the current channelized delivery method.
It would be foolhardy, IMO, to expect that any one of these ISPs, including ATHM, would be able to control their own traffic for any length of time in a commercially viable manner in such a pluralistic setting. Even in the absence of this Major-Kludge-In-The-Making, they'll have problems controlling traffic patterns and service quality, as time progresses.
But getting back to the plurality issue, the sector will not be able to do this without drastic architectural changes taking place first, to an HFC architecture which continues to prove itself (again, to me, at least) to be obsolescing itself before its time, in the face of changing and heretofore never-in-your-wildest-dreams requirements and expectations.
Of course, these are only my opinions, and they are all dependent on the assumption that openness will ensue among the cable operators, forcing them to allow entry to all ISPs, and eventually, to CLECs and DLECs, as well.
On the other hand, if Kennard does get his way and prevents the AOLs and the rest of the world from getting in, then I think that there is still some breathing room left here, but not a whole helluva lot.
The above scenarios have all been thought out, rather exhaustively I would assume, by all of the players involved in this dilemma, and the prospect of a viable open architecture must seem to them, as well, to be an elusive one. For, no one who I am aware of has come forward [please speak up, someone, if I am mistaken about this] to offer a proposed solution that makes sense to all.
All of those looking in from the outside here (like AOL) must make noise in order to dispel or distract from the progress being made by their foes (ATHM, in this case), if nothing else, while they continue to contemplate their options. Do I hear a ping from out there? Hello?
Regards, Frank Coluccio |
