5 ms·
And then everybody would switch to this new and cheaper service. Then your website, which isn't on the list, wants a fast connection too. For a fee the ISP woul
by pbz 12y ago
And then everybody would switch to this new and cheaper service. Then your website, which isn't on the list, wants a fast connection too. For a fee the ISP would gladly make your service faster. However, many could not afford this and would go out of business. This, ironically, would limit your "choice" on the other side.
- greggyb 12y agoNo, those consumers who judge that that level of service is a better deal for them would switch. To step back a bit, what I posited was only a hypothetical to show how such a choice might limit competition down the line. To step back even further, your post makes it clear that we have an ideological difference here. I do not view rate-limiting as inherently bad. There is a ceiling above which I would not be willing to pay for a completely unthrottled connection. There also exists a price at which I would be happy to take a throttled connection. My general preference would be for an unthrottled connection, but my preferences do not exist in a vacuum. The price difference where I would choose the throttled connection would have to be quite large, and the throttled vs unthrottled rates would also come into play. Further, some of my favorite websites would not be much affected at all. This site, for example places a miniscule load on my connection and minimal data transfer in a month. I don't generally frequent multimedia sites. I utilize an RSS reader for blogs, so that bandwidth is also fairly minimal (especially if I choose to omit images). Email is text for the most part. A website could compete very well with minimal bandwidth for any of these purposes. Furthermore, you assume that the bandwidth would be throttled always, and I've gone along with that. There are other rate-limiting strategies that might be chosen. One might be a data transfer cap, beyond which the bandwidth is throttled. Another might be peak-hour throttling, in which case my late-night browsing would be unaffected (and I do a lot of that). One plan that would be pretty tempting for me is unlimited Netflix and Pandora with a reasonable data cap (I haven't tracked my monthly transfer rates lately so couldn't say what would be reasonable here). Further, I've not suggested that this be the only plan available. I merely suggested it (hypothetically) as a manner in which a new entrant might gain enough market share to establish itself as a player. The more ISPs there are, the more likely one is to offer service you like. The more you restrict the terms an ISP may offer, the harder it is for them to compete. I hope this thought process helps clear up where I am coming from. Finally, I ask whether you consider toll roads immoral, or whether you hold your cable (television) provider responsible for the (production/acting) quality of the television shows you watch, or if you want to stick up for HBO and how downtrodden it is by NBC? The provider of the means of access has no moral imperative to provide equal access to all content.
- pbz 12y agoIt's not a question of morality. What it boils down to is which side are you on. Do you care more about ISPs and their ability to do whatever they want in order to maximize their profits, or do you care more about the health of the ecosystem (Internet)? In your hypothetical, if an ISP could come with an offering that would hurt the ecosystem (by favoring X against Y, and hence not allowing the Internet to be a leveled playing fields) but offer some savings for its customers then I would be against it. It's far more important to maintain a leveled playing field than save a few pennies here and there. However, looking at other developed countries and their offerings, we can certainly conclude that what we get now is the worst of both worlds -- expensive monthly cost and throttling -- so there's plenty of room for improvement (cheaper service) while maintaining net neutrality.
- greggyb 12y agoI do absolutely agree with you that ISPs are in a poor state in the majority of the US. I think a lot can be done to improve it. I disagree strongly that net neutrality is the most important or most effective lever to push on. As I have made clear, and as I think is fairly obvious to anyone who has moved between various ISP markets in this country, I believe that the quality of internet service is positively correlated with the number of providers in an area. I have cheap and reliable internet with three ISPs. None of the three throttle their connections in this market, and one has just introduced 10Gbps fiber to the home. The fiber provider is actively expanding, but even if you don't live on a street with access, you can get cheap reliable service from either of the remaining two. I think it would be better for any individual in the US who cares about their internet access to move their markets in a direction from one ISP to more ISPs (like the area I live in) than to pass a law enforcing net neutrality and forbidding any kind of rate-limiting. I do care about ISPs being able to offer me as many types of service as possible, and me being allowed to choose the service I prefer. Again, my general preference is for unlimited internet, but my preferences are bounded by costs. Another hypothetical: Suppose I am willing to pay $15 for internet access, and the only service an ISP is willing to offer at that price is throttled in some way (data cap, certain sites slower, whatever scheme). If you eliminate this as an option, then you have eliminated my access entirely, because $15 is my limit. In this way, marginal customers may be harmed by net neutrality. I am not convinced by the argument that the internet will be ruined if some internet service contracts are rate-limited in some way. The biggest threat to the next (Google|Facebook|Twitter) is not any ISP, but (Google|Facebook|Twitter) existing and having a large network effect for its services. And to be honest I have not seen much more argument beyond this point. Even if I were to accept this argument, I think the pathetic internet connections that some people in the US have (there are people still on dialup and <5Mbps isn't that rare in rural areas based on my experience) would provide a bigger impediment to the-next-big-thing reaching all of its potential market than any rate-limiting. Multimedia content, especially, is awful on a slower connection. And no regulation of rate-limiting will do a thing to increase speed and reliability of networks - at absolute best it has 0 effect. Forgive me if this is a misinterpretation, but you seem to jump from my hypothetical of _one_ ISP out of >1 in a market offering _a_ plan that is rate-limited in some way (what if it is just a data limit? This does not favor any portion of the web over another.) to there being no un-rate-limited service contracts available. I find this incredibly unlikely, primarily due to the huge unpopularity of rate-limited plans among a vocal minority of internet users. Finally, I am quite confident that two ISPs fighting for my dollars are more likely to offer me something I want than two ISPs fighting for a favorable interpretation of some regulation.