You raised a lot of issues here and I'll try to answer them.
1. I think, but honestly am not sure, that the $175/mo. for HDSL from Interprise includes Net access. But it might not. I think the ADSL commercial service in Chicago is over $200 for the ADSL line including the Net, not to mention equipment and install costs. Which is why I also think ISDN is the mass market service for the next few years. Any form of DSL is going to be niche marketered for several years, because average people won't be able to afford it.
2. Cable modems are mostly press releases. Very little of the U.S. coax infrastructure can handle high-speed data. I think even the cable modem advocates would agree with that, or at least the person I talked with today from TCI Cable agreed. The battle, for a while, will be on the telco side.
3. I don't think VOD is commercially viable right now. Those who come at it from a technology angle completely overlook the business of renting movies. Video stores, pay-per-view and movie sell-through on videocassette, and soon DVD, will make the business case for VOD over the telco infrastructure a very weak case.
Notice that the telcos have abandoned their links with Hollywood studios, i.e. TeleTV and Americast. I wonder if it ever occurred to telcos that Hollywood doesn't need another pipeline to consumers to market their hit movies. What Hollywood needs is a way to market their dogs. Video stores, which promise you hits but deliver empty shelves and consolation-prize dogs in their place (who, after all, EVER leaves a video store without a tape in their hand?) are absolutely perfect for this.
Prediction: come 2010, we'll still be going off to the video store to look for hit movies and coming home, half the time, with our second choice. If you think RBOCs are going to spend tens of billions to get into the business of charging three bucks for a movie, I think you'll be sadly disappointed. If any alternative delivery mechanism gets the business it will be satellite dishes.
4. Yes, HDSL is now the winner in DSL. Interesting that as recently as a few months ago no one was saying that. I think HDSL will be ahead for the next 2-3 years and maybe longer in the DSL space. In the mass market, BRI-ISDN is going to have a much longer run than people think.
5. Asymmetrical architecture isn't the issue with the network backbone. The issue is more bandwidth in the backbone, and high-bandwidth switching, i.e. frame/cell based switching. Not much use in handing out megabit services to homes if it can't be switched and you're nailing up connections to a packet-based Internet most of whose servers are connected with T-1s.
For the next several years, the only uses for DSL service will be LAN extension and small business connections from the LAN to the Net. For these apps, HDSL beats the pants off of ADSL.
Ultimately, I think ADSL evolves into VDSL and becomes the drop from a fiber-based digital loop carrier infrastructure. We'll see that in 10-15 years. Meantime, ADSL strikes me as a laboratory phenomenon plus press releases and trials.
But, hey, who knows what any of us is overlooking in the fast-paced world of telecom? If I'm wrong, I think it will be because I'm overlooking the speed that Frame Relay/ATM backbones arrive, or perhaps some wireless broadband methods. |
