You're free to pay for an internet line with an SLA attached, with better support, with redundancy, etc... None of that is a problem under net neutrality.
No, in this example, I pay my ISP to make my end faster, the person on the other pays their ISP to make their end faster.
In the no net neutrality example, the person on the other end pays their ISP to make their end faster and then they also get contacted by my ISP who asks them to pay to make my end faster when talking to them.
So instead of agreeing with my own ISP on how fast I need my traffic to be, my ISP cuts me out of the loop and goes and talks to some third party to decide how fast to make my traffic.
I understand the downsides of extorting specific sites to pay more or be artificially throttled. That is an unambiguous case of NN violation and the kind of thing (per original comment) I'm in favor of prohibiting.
But NN advocates (as best I can tell) want it to go further, and prohibit
A) the other end voluntarily subsidizing my end on the condition that the funds be used for pipes that they (having partially paid for) have privileged access to.
How is that any worse than the paying for a bigger chunk of the ISP's pipes? And it seems economically equivalent to:
B) A content provider P subsidizing a mobile data network's increased capacity on the condition that P's transmissions not count toward customers' limits.
(I know B drives many people mad as a NN violation, while other advocates are like, "nah, cell data is okay.")
Furthermore, B only seems to differ in degree from:
C) An ISP spending customer money to interface with content provider P's system for replicating P's content within the ISP's network [1] so that access to P's content is faster an less expensive -- instead of spending that same money to make the network faster for all customers and all remote servers.
(Case C is like B, but where the "subsidy" is the expense of making Open Connect-like systems work.)
And yet, everyone seems fine with C.
If it seems like I'm nitpicking, let me step back and look at the big picture: NN looks like a bundle of good ideas ("don't artificially throttle specific sites"), married to a bunch of economically dubious hate for (the equivalent of) toll roads and desire that there be no downsides to higher usage -- the classic motte-and-bailey[2].
Like many anti-competitive behaviors, A and B are seemingly beneficial to customers, but in the long run they are harmful to the market by preventing smaller players from competing fairly.
> married to a bunch of economically dubious hate for (the equivalent of) toll roads and desire that there be no downsides to higher usage
I have never seen a net neutrality proposal that includes something that prevents higher charges for higher usage (except when mis-attributing usage, e.g. saying that Netflix uses an excessive amount of consumer ISPs' bandwidth, which is not the case, the people watching Netflix might be overusing their ISP's bandwidth, but Netflix is not using the ISP's bandwidth at all). This is a complete straw-man argument.
>Like many anti-competitive behaviors, A and B are seemingly beneficial to customers, but in the long run they are harmful to the market by preventing smaller players from competing fairly.
But what's the difference between those and C? Or between those and me building a dedicated line straight to my favorite server, just for us? Or are you really against Netflix's OpenConnect as a NN violation?
>I have never seen a net neutrality proposal that includes something that prevents higher charges for higher usage. ...This is a complete straw-man argument.
From a very quick search:
"Net Neutrality is one of only a few tools available to the FCC to keep ISPs in check. Banning data caps and zero rating schemes would be another great way to protect consumers from Wall Street’s insatiable demand for companies to extract more revenue from consumers."[1]
There -- a major advocacy group (stopthecap.com/tag/net-neutrality/) equating net neutrality with a ban on data caps (a species of charging for usage). [2]
The fact that you are too smart to make the argument doesn't mean others aren't, nor that it isn't polluting the discussion, nor that I can't legitimately object to NN if it's read so broadly.
[2] That's on top of the ever-present poster who thinks NN will keep their ISP from throttling their constant bittorrent streams in favor of granny's occasional emails.
Banning data caps is very different from not allowing higher charges for higher usage.
Most data caps are of the type where you pay a flat fee for a fixed amount of data and then pay an outrageously high punitive fee for overages that is vastly out of proportion with the amount you are paying for the data below the cap. In other instances your speeds get dramatically throttled after reaching the cap or are completely cut off if you go over the cap too often. I think that should be outlawed.
Those aren't good faith attempts to charge people for what they are using; they are attempts to shut down heavy users, often because the ISPs are also in the cable TV business which competes with streaming video that is responsible for heavy data usage.
A good faith attempt to charge heavy data users fairly for their use would probably look something like the billing models used for electricity.
>Banning data caps is very different from not allowing higher charges for higher usage.
>A good faith attempt to charge heavy data users fairly for their use would probably look something like the billing models used for electricity.
It's not "very different", it's one way of charging higher users more. And the fact that it's a bad way doesn't mean the NN crowd supports the good way either! They object to the electricity model on the grounds that "bandwidth isn't scarce[1]" and that high users should be just as capable of crowding out low users during peak times, that there shouldn't be any downside to higher usage.
[1] in the true-enough sense of "the relevant scarce parameter is not how much you download in total, but how much of the flow of the pipe you are using at any given moment".
It's still neutral. The reliability doesn't depend on the contents that you're transmitting. Cat pictures sent over a 5-9's SLA line are just as reliable as your video conference.
