I guess they most be using something in front of the code [0] (which do document the double slash issue). Probably should have used PathUnescape[1] on line 185 though.
I don't like it either. What I'm saying is that this is something people who support net neutrality has to be able to refute.
Most consumers only want slow traffic to many websites and fast access to a few websites, but the are paying for a lot more. The example of "sites" might be a bit convoluted. We could as well make the differentiation based on geography, which is sort of already the case. Get 1 gigabit domestically and 10 mbit internationally. Now it up to people who want to host something to buy transit into the country. (This is sort of how it works in China, where a lot of people don't speak English anyway).
> Back to your scheme, the customer would not pay less. Costs have to be recouped somehow.
People who use the Internet less would presumably pay less and people who used the Internet more, including setting up servers, accessing services far away or anything else that require more infrastructure, would pay more. Which is sort of the case today with upstream bandwidth.
My overall point would be that the open Internet costs money and I'm not sure people can be convinced to keep it ones there's an alternative.
The poster said "OCSP requests that use the GET method use standard base64 encoding, which can contain two slashes one after another", so I'm wondering if that's that case or it's actually "clients don't encode their base64". I didn't read trough the RFC, but judging by the later statement of "OCSP request using the GET method is constructed as follows 'GET {url}/{url-encoding of base-64 encoding of the DER encoding of the OCSPRequest}'" that seems to be the case.
I'm not sure I understand how slash collapsing is affecting this. Slash is a reserved character and presumably if it the data was correctly encoded should never have ended up in the URI in the first place?
> First, it confuses technological innovation (impressive to engineers) and economic innovation (valuable to consumers). Second, it confuses gross and net benefit — of course, when government does X, we get more X, but is that more valuable than the Y we could otherwise have had?
[...]
> And don’t forget Bitcoin and the underlying blockchain technology that has the potential to decentralize everything from web apps to logistics.
The latter here seems like a rather good example of the former. EU is about to introduces the revised payment services directive (PSD2) which is set to force banks to interface with third parties. I would expect this action to have far greater value to people than the blockchain.
What you're describing is how net neutrality works, not how the Internet works.
> First, bandwidth is limited. It's whatever download and upload you sign for.
I said that I don't encourage bandwidth limits, not that it wasn't limited. The point is you usually sign a flat fee, you don't have universal access and then pay for traffic, more like in a country like Andorra.
> Second, there is no such thin as "full speed access to Netflix" or whatever. It's limited by your own bandwidth at least.
It's only limited because the ISP has limited it, it's entirely possible to give someone with a slower overall internet connection a higher speed to certain services.
> They don't have to update their infrastructure every time a new web site comes up. As such, access to Netflix, Youtube, or http://loup-vaillant.fr are not separate services.
I don't see you point. ISP do peer directly with services, but I never said they had to. You just have to have a separate exchange for full speed services.
> ISPs that treat access to particular web sites as separate services are just plain lying. The real "separate service", for which they actually spend money, is the throttling/filtering/censoring infrastructure they have to put in place to charge different web sites differently. Text book malicious features, not unlike DRM.
I don't agree with throttling, but "throttling" (or congestion) does occur naturally on the Internet.
I think you really don't get my point, so I'll try and explain in another way.
Today you pay a higher amount for something like a 100 mbit or 1 gigabit connection. That's for all the infrastructure and bandwidth for giving you that access (or at least hopefull something close to it). How this works differs, but generally your ISP would pay to access bigger networks.
Now imagine instead if your ISP only guaranteed you a certain amount of normal Internet traffic, either as bandwidth or data. And then it would be up to the services to pay for the infrastructure and bandwidth to deliver their content at a higher rate. It would really just be a formalization of what's already happening with transit vs. peering. If you want to see it from a technical perspective, how it is today, imagine having 10 mbit to the open internet and 1 gigabit to all the services that your ISP peers with (even in the extended network). Then imagine possibly your ISP getting paid by those services instead.
I'm not saying that it's a great idea, I'm saying that it is a winning one. Because consumers would pay less, ISPs would require less infrastructure or pay for less bandwidth (as you said these things are related) and it's actually creates incentive for people to roll out Internet access. And if you want net neutrality it's an argument like this you would have to be able to argue against.
I realized that it's pretty fruitless to discuss these things on HN, but since you responded I will give you a response.
This isn't something that is generally happening, at least not outside mobile connectivity where certain services are sometimes included. It's not really about what you, or other 'heavy' Internet users wants, but about consumers. People who want net neutrality have to have a response to normal people why it's relevant to them.
More and more people are getting access to broadband, yet it's still quite expensive. The idea, which you will have to defend against if you ultimately want net neutrality, is that companies could pay for the bandwidth. So you as a consumer would pay for connectivity, say 10 mbit. But instead of you just having access to 10 mbit, other companies would be able to pay to deliver to you at full speed. So when you pay for Netflix part of that money might go back to the ISP. This is to some extent already happening, but through various transit and peering agreements.
> I don't get your question about "why consumers are paying for both installation, the bandwidth itself and the services they access". I don't pay for "installation", but rather for access, by month. I only pay for a few services, and never see ads. For the most part, I only pay for the unmetered 100 Mbps uplink.
You do (usually) pay for installation, it's just that it's "baked in" to contract time and fees of your connection. That's why you usually can't get a really cheap connection on a wired installation. A 1 mbit connection would still usually cost you a fair bit, if it's even available. With something like electricity you do also pay for installation, but since that is largely universal it's really part of the overall cost. Point being that a lot of people lack broadband because they can't afford the installation cost and that is a big different between Internet access and other utilities which are much more universal.
[0] https://github.com/letsencrypt/boulder/blob/master/vendor/gi... [1] https://golang.org/pkg/net/url/#PathUnescape
PS. It does seem like a pretty bad idea to put data in the path instead of the query.