NNSquad - Network Neutrality Squad
[ NNSquad ] Re: Skype works. Competition is hard to create
Who said Skype doesn't work in the US? The only trouble I ever have with VoIP is calls to places like Peru or Internet cafes in rural Chile. Video Skype is somewhat more challenging, but it "works" more often than not. I think the point we were discussing - or the one I was discussing - was immersive video conferencing, which is quite far beyond Skype. Regarding your thoughts on competition, well, gee. Unbundling has never motivated a single network operator to invest a dime in pulling fiber to the home, so if that's the end goal (or a milestone) we're going to need a fantasy other than unbundling or hyper-regulation to get there. Networks have major economies of scale that push toward consolidation. In the UK, BT has over 60% of the ISP business despite unbundling, for example. The "miracle" of cheap broadband in France is confined to the parts of Paris where the amazing 500 year old sewer system serves as a free conduit for fiber; France as a whole is a DSL nation with slower downloads than the US on average. US regulators are happy with DOCSIS, a technology that doesn't lend itself well to unbundling or to fancy anti-QoS rules, but does support managed services quite nicely. An "all packets are equal" mandate would make me buy stock in cable; it's a license to print money. Richard Bennett On Aug 28, 2010, at 8:38 PM, Dave Burstein <daveb@dslprime.com> wrote: > Folks > I don't write to this list often, but the comments that Vonage and Skype don't work on most U.S. broadband networks are plain wrong. I spend a lot of time on the phone and it's all Skype and Vonage. When we asked a class at Columbia how many use Skype, more than half raised their hands. (Lots of international students.) > > I essentially never have congestion issues in Skype, Vonage, Netflix, and Hulu on either Verizon DSL or Time Warner Cable. That's certainly not true for everyone, but what I'm hearing from teh network engineers at half a dozen of the large networks and it's typical. > > I've seen the network data from several large U.S. networks and congestion is remarkably rare. They've even solved the upstream cable p2p issues in the last 12 months almost everywhere with interim upgrades. > > This is the kind of thing to answer with data, not speculation. > Separately, I like Gilmore's idea of solving things thru competition, which clearly worked well in France in 2003. It's working in British retail as well. > > But it's a big jump to think it would work in 2011 in North America. The telcos and cablecos have huge scale economies and hence much lower costs. In 2011, nearly everyone not poor but interested in broadband is connected. It's far more difficult (and expensive) to persuade someone to switch providers than to win customers when people are first starting up. > > Which implies it would be hard as hell for a new ISP to reach efficient size in 2011 than in 2003. Wall Street tells me they wouldn't finance it. Unbundling in the U.S. still remained possible through UNE-L (equivalent to what most competitors in Europe are using.) No one has developed in the last nine years big enough to make a difference. > > Interestingly, neither has anyone in the major European countries except for those who already have a huge number of customers for a piece of the quad play. Specifically, wireless carriers Bouygues in France, Vodafone in several countries and Carphone Warehouse/TalkTalk in Britain have come in with some success. So has BSkyB, the British satellite TV company. > > But other than those, I can't think of any successful new entrant in any major country since before 2006. They are all already in the game, with huge customer bases to draw on and a good reason to lose money on broadband for years so as to hold on to the customers, as all of these companies have. > > So saying "the problems in the U.S. should be solved by competition" is very attractive and not totally impossible. But it's highly unlikely any likely form of unbundling - even full structural separation like England and Australia - will have major impact in the U.S. unless one of four companies decides to jump in: Sprint, T-Mobile, DirectV (which almost did five years ago) or Dish. Sprint and Dish couldn't raise the capital and I don't think the others are likely. > > I wish it weren't so, but it's very unlikely that we'll get high speeds except from telco and cableco in the U.S. in volume enough to make a difference this decade. > > So we need policy that works if we only have two highspeed players rather than dreams of competition. db > >