Hi Mike, re: AOL and Satellite
I would have more than a few problems with this, but let's just focus on two of them for now. Latency and Directionality.
While a quarter of a second, approximately, which is what the propagation time is needed for an end-to-end satellite link under ideal conditions, is not much (when you have isochronous services like T1, DDS, etc. to work with), it becomes a huge burden in an Internet model where you have a best effort scenario. In the IP world you must contend with unpredictable latencies such as: congestion effects, such as packet queuing; round-about terrestrial paths employing an unspecified number of hops; policy based inhibitors/accelerators (depending on an eventual QoS/ToS scheme to be put in place; and the transaction profiles of the specific types of applications being offered.
For web lookups, this is not an appreciable problem, since we are by now accustomed to some delays. even very small ones. But for future VoIP and other time sensitive services, I can see where this additional latency will impact those services severely. Keep in mind that the ITU/IMTC VoIP draft committees have established less 175 ms or less as the bogie to beet for VoIP. This means that satellite based VoIP loses, before they even have a chance to get out of the gate.
The second issue is the more obvious one of the return path, and symmetry, or the lack of any potential thereof. Even if very high speeds for best effort text, web page delivery, and streaming video is achievable in the downstream, how will the upstream be provisioned? I don't see any mention of this, so I wont automatically assume (not yet, anyway, since none of the releases are addressing this shortcoming yet) that it will be over twisted pair POTS. Do you know?
Wouldn't it be a rip if it were supported over a cable modem IP-telephony, or a circuit-switched cable telephony, link?
Regards, Frank Coluccio |
