- Top Features:
- War Room
- Glenn Greenwald
- Broadsheet
- Joan Walsh
- Camille Paglia
- Since You Asked
- Follow Salon:
Read other letters about this article
Because it becomes impossible to sell packet prioritization and VoIP is extremely sensitive to QoS (Quality of Service) isochronous packet priority ordering.
I've been noting that for many years now. Despite early hype, IP telephony was much delayed in getting out of the starting box, and a large part of the reason was these technical requirements (lacking guaranteed latency delivery of low-bandwidth but continuous [isochronous] content, echo cancellation goes to hell and you get the sound quality of talking down a well ... or worse).
IP in general operates better under less than maximum utilisation, and tends to fall apart as usage approaches bandwidth capacity (and not nicely). For web content, you get ised to it, and just hit the reload. But for voice, the conversation gets to be totally useless with any significant dropouts (the commercials to this effect are hilarious).
With all the IP video nowadays and streaming media, the bandwidth starts to approach tolerable capacity for these apps, but that's far above what makes VoIP two-way communication intolerable.
The solution is the evolving QoS protocols and different classes of service. But as one of my buddies (and a frequent contributor to telecom spec/tech meetings) pointed out, all that does is make your VoIP look like circuit-switched signaling, and if so, then why bother switching; we already have billions of dollars in circuit-switched technology equipment deployed.
One solution to the "conflict" between "net neutrality" and the requirement for grafting a "voice-friendly" QoS scheme onto the IP network is to allow for "class of service" prioritisation, but not for prioritisation between users of a certain class of service based on user or provider identity. IOW, neutrality of access within each type of service.....
Cheers,
