Slashdot Mirror
ISP Guarantees Net Neutrality, For a Fee
greedyturtle writes "Ars Technica has up an interesting article on the first ISP to guarantee network neutrality. It's called COmmunityPOwered Internet, aka Copowi. The offer of neutrality comes at a higher price — mostly due to uncompetitive telco line pricing schemes — $34 for 256K DSL, $50 for 1.5 Mbs, and $60 for 7 Mbps. The owner claims to need only 5,000 subscribers to move his ISP into the national arena from the 12 Western states where it now operates. Would you be willing to spend the extra bucks for network neutrality?"
The thing is, net neutrality is not really about the last mile. I used to be a sysadmin at an ISP, and utilization of last mile bandwidth was never the issue. Whoever they are leasing the local loops from doesn't care per so about whether my DSL line is saturated all the time or not. Net neutrality is about load on the backbone carriers' networks, and below that, on downstream ISP networks. It can also be just about extra profits. A couple cases to illustrate each:
1) Simple load. There's an overselling of bandwidth formula by which all ISPs make money. If the aggregate bandwidth of all your customers is X, you don't have to have X amount of backbone bandwidth, because they aren't all online at once, or all fully utilizing their links when they are. You only need some fraction of that amount. You've got this all worked out, but then along comes Youtube, IM with voice, Vonage and other VoIP carriers, Bit Torrent, online music and video stores, etc., and in pretty short order, your average user is consuming far more bandwidth than they used to and your oversell ratio just went out the window. To maintain level of service, you can do a couple of things: the first is to throw a bunch of money at the problem, upgrading your bandwidth, your core and edge routers, the whole nine yards. The trouble is, this is expensive, and while the routers are a sunk cost, bandwidth is a running cost. Profit margins are very thin for ISPs, generally, so to remain profitable you would have to raise prices. But Internet access is very price sensitive, and the first one to raise prices is going to see customers walking, in large numbers. The other option is traffic-shaping. You prioritize some traffic over others, and put the bandwidth-hogging stuff like Youtube, BT, online music and vidoe stores, etc, at the end of the bandwidth line. Unless, of course, Google, the stores, etc., are willing to pay you money. Now you have a way to finance that infrastructure without raising rates. Net neutrality is dead, but you're still alive. And Bit Torrent? Oh well, nobody's paying there, so BT is just going to have crappy performance on your network.
2) Greed. I'm a big ISP. I want to get into the VoIP business for myself, so I do. My service is super, and it's cheaper than the phone company. My customers like this. Trouble is, there are VoIP companies out there competing with me, like Vonage and Packet 8. Their service isn't as good as mine (I used to use Packet 8, now use Vonage, and in between had a cable company's VoIP service, so I'm talking from experience here; but Vonage is pretty good), but it's almost as good and it's over 1/3 cheaper. A lot of customers like that even better. What to do? Ah, I know! Traffic shaping! Packets for my own VoIP service get routed at a higher priority than other VoIP services. No their service is no longer almost as good as mine. My customers may or may not really like this, if they even pin it on me, but now my service is worth the premium I charge for it b/c I made the others look bad. Net neutrality is dead, and I can no longer claim with a straight face to be a common carrier like a telco, but I'm making more money and can use it to finance the greater bandwidth demands from case 1, above (along with the fees I'm socking the content providers with to not be traffic-shaped on my network).
This, then, is the problem facing Copowi: They may practice complete net neutrality within their network, consisting of their edge, their core, and the local loops they are leasing. However, if their upstream (be it a major backbone carrier or just a larger ISP who in turn connects to a backbone provider) doesn't practice net neutrality, it doesn't really matter much that Copowi does, except on traffic local to their network, which isn't a whole lot.
Of course, if their upstream starts traffic shaping on VoIP, P2P, whatever, and Copowi wants neutrality, they do have an option: pay to have no shaping on traffic going in or out of their network. And lo and behold, this appears to be exactly what's going on
Having started an ISP with famous investors too dumb to put their money where their mouth is I can tell you why I worry about these guys. Certainly, if you are on Comcast and can move to them, go for it. The problem is, you know that $200 Billion people are talking about? The 200 gigabucks that went up in smoke? Look this isn't Cheech and Chong. Money doesn't fly away. What if the big boys actually did invest in fiber and equipment, but they just don't want to roll it out unless they are dragged out and screaming? That's a lot of money. The big boys are waiting to see how far they can push it, and when something starts to look interesting, if they can they will smash it. Welcome to the ISP business.
Now if these guys are going to try and tie in last mile people with great service and maybe value added (how about 2 free locally served movies a month, etc.) then they might have a future. Or if they could spam access to people wirelessly with some amazingly cheap technology, maybe. Maybe they could also have a chance if they are spinning off the hardware to someone else and they just have to sell "virtual" service. And maybe if they build a nationwide grassroots league (a federated little league if you will) peering with similar companies, they could even offer higher speeds and lower latency possibly. Or maybe if they could get some nice deals with municipalities or academia. Well maybe. I'd go with them if I was unhappy with my U.S. provider, though I'm not in the U.S. now, but long term? Their website says how it will be good for the long term. Personally, I've seen costs drop every 3 months, if it makes sense in the short term and you are getting really hassled with your ISP fine. But I think the only way to get good service is to legislate it. There are too many maybes, and too many big boys with big bank accounts who are just playing a cynical game until you show up on their radar.