Hello Jay,
I thought we weren't going to get to justice, let alone fairness, until we were done with truth.
When you ask "who will pay for it" you make the assumption that "it" exists.
"It" doesn't exist.
No one has come up with a widely accepted interface (which was referred to as the point of interconnection, or POI, in the Canadian paper I posted two days ago) approach thus far that will satisfy all of the players and all of the regulators.
In fact, I've not seen Version 1.0 proposed yet that merits serious attention, come to think of it. I don't regard the GTE demo in FL as fitting this bill, since it was overly simplistic and never demonstrated how hundreds of new SP entrants would be supported on top of an already potentially overloaded condition (when ATHM begins to approach fully subscribed potentials). Beyond those basic reasons, it also didn't show how interactive voice and video would be supported in a secure and reliable means, either. Let's not even get into QoS or Security issues on this one, 'kay? Thanks.
All I keep hearing is two years out, two years out. I suppose they feel that it's going to take them two years to come up with a viable solution. Otherwise, I would have serious cause to wonder if T would be leveraging the exclusivity issue as much as they have, in part because they "don't" have any solutions to offer, despite their maybe wanting to. In a way it protects them in more ways than one, since they don't have a viable means of accommodating OA now, in any event. Leastwise, not in such a way as to preserve any semblance of quality in the delivery of those services, be it theirs or anyone else's, if they had to take a shotgun approach from the hip.
When a solution, or solutions, are ultimately presented, then it will be possible to put a label and a price on it. Until then, expect to read a lot of rhetoric from Kennard's offices, and posturing on the parts of the principals.
Did I say, yesterday, that machines don't know what the heck they're doing? In some ways the same goes for the regulators and the operators in this instance. The current HFC/DOCSIS model took years to develop and it was only intended for one facilities based operator and their chosen single platform. To put this into perspective, say, it was designed to allow a cable operator such as Cox to administer a cable modem service such as ATHM, or TWX to administer RR.
Several more ISP services can also be "squeezed" onto these systems, granted. There are ways to partition the flows, and cater to more than one provider. But there isn't any viable means on the current architectures which exist today to allow all comers who want in, not by a long shot. It was never intended in its current form to support 6,053 ISPs over it at once. Today's HFC/DOCSIS is not an 'end state' design for such purposes.
In order to open up the pipes fully to allow for "internet" like openness, DOCSIS and the coaxial segments will need to be either radically changed, or supplanted by fiber and an open IP scheme. Lightwire is a step in this direction, as is the contemplation of removing the DOCSIS devices and replacing them with Fast Ethernet network interface cards, or NICs. Comments and corrections welcome.
Regards, Frank Coluccio |
