3 ms·
Differential Pricing of Traffic in the Internet
- awalton 9y agoStudy pretty firmly starts with the assumption that all content providers are fungible, and that's clearly not the case. In fact, it's almost exactly the opposite of reality, apart from aggregators and news outlets, which could use kilobits of data to transfer but are usually bloated with an extra meg or two of markup, fonts, and javascript (and now, more often than not, video content that is literally some local news reader reading the written content...) If you charge a billion extra dollars to transit YouTube or Netflix, those businesses go under. The economics doesn't require a mathematician to figure out.
- kfk 9y agoSo here is what I understood. Price differentiation will impact the equilibrium as content providers (CPs) are now competing also on connection speed. Lower quality CPs have incentive of paying ISPs to get better speed for their sites. If lower quality CPs pay ISPs more than the good ones everyone gets worst off. In this scenario CPs are also worst off as now they are diverting money to ISPs instead then profits or investment.
- gumby 9y agoGood analysis. Having only one provide is de facto the situation for most of America, FCC maps not withstanding.
- wskinner 9y ago“Most of America” as in most people, or most land area? I have lived in cities my whole life and have never had less than 3 choices of provider at purchase time.
- wtallis 9y agoAre you counting wireless ISPs? They're not a viable alternative at scale.
- drdaeman 9y agoWhat's wrong with wireless? (Unless that means LTE or WiMax or some broadband tech like that) I've always had FTTH, but at current residence (non-US) I have a radio bridge for the last ~500 meters - there's a railroad nearby so my ISP can't lay fiber directly to my house. I've expected worse (esp. in a bad weather), but have zero problems whatsoever. Besides an extra half millisecond to what it could've been - which is insignificant - I have no issues.
- wtallis 9y agoBroadcast wireless (including LTE and WiMax) obviously doesn't scale. Point to point wireless can make far better use of the spectrum, but requires a lot more equipment at the endpoints and doesn't scale to large numbers of customers as easily as broadcast wireless. It's much closer to wired connections both in terms of cost and performance, but is fairly rare because it doesn't have a compelling business case except in cases like yours where the simple straightforward reliable wired solution is unavailable.
- nugi 9y ago7 states in 8 years. Only 3 of 7 homes did I have more than one option for high speed, many with 1.5down on a good day. Outside developed and high income areas, the situation is still quite dire in the USA. Oddly, I found suburbs to be much better served than rural or downtown areas. One rural area had no options at all (beyond hughesnet and dialup), and I was forced to find a different area. When we complain about rural folk being backwards, and extending none of the future to them, we only have ourselves to blame. Quit forgetting about the rest of the country that does not live in cities.
- gumby 9y agoMost people. Here in Palo Alto, for example, my choice is Comcast or AT&T, but AT&T offers only dsl and, having tried it, the fastest I could get was 768kbs even though I’m less than a mile from the CO. My vacation house offers the same choice: Comcast of AT&T but, like the cheese shop, I can’t actually choose AT&T, only Comcast. And that’s in town.