Europe reverses course on net neutrality legislation
arstechnica.com
arstechnica.com
The problem in the states is very very different. In the states you have three corporations providing shit, slow service. Th
Traffic shaping, which stops one person on a contented link (ie ADSL) ruining everyone else's day by torrenting every last drop of bandwidth. Its been a fact of life for many years.
THe problem in the states can be characterised in two ways:
* A lack of competition
There is often only one ISP in town, and they don't have to try hard. This means that service is terrible, and so is innovation.
* A lack of bandwidth
In Europe bandwidth exchanges, or peering exchanges are normally run by a cooperative. This means that its in everyone's interest to talk to everyone else as fast as possible. Its really really cheap to peer in london, (http://www.lonap.net/fees.shtml)
In the states, exchanges are run by either AT&T, verizon or level3. Which means everything costs squillions. As there is no need to provide more bandwidth, because customers can't move between ISPs so no one peers.
For example in the EU, ISPs are more than keen to have CDN leaf nodes in their infrastructure, it lowers costs and improves customer experiences. However they can't charge Netflix et al to install them because there is too much competition.
Its not like they can limit access to netflix for a few months to make a point, because their customers would change ISPs. Something not possible in the States.
So net neutrality is really a side issue to competition.
No, net neutrality is not a side-issue to competition.
Competition would mitigate some of the harms of non-neutrality, but it does not address the full picture. Non-discrimination is essential both for consumers and for content producers[0]. Non-neutrality has an incredibly chilling effect on business, even if it only applies to a portion of the population[1].
We need both neutrality and competition; we can't sacrifice either one for the other.
[0] ie, web companies, startups
Once again, with competition that sort of practice doesn't work. If a service is popular it will cost customers to ban it.
Nokia phones came with music for the best part of 3 years, it didn't stop the rise of spotify. (back when data cost £1 a megabyte)
Spotify is bundled with many contracts here in the UK, but people still use other services.
None of those VC ideas are businesses, it won't be net neutrality that stops them it'll be common sense. Who invests in clones of already loss making services?
With these new rules governments can now decide which services are qualified as "specialized", with this they can for example censor the opposition by slowing them down to a crawl, or take bribes to declare some corporation's service "specialized" gaining an edge over the competition.
Second of all net neutrality is not about freedom of speech. It is a concerted campaign by Google, netflix, yahoo and amazon to keep costs low.
There is a very real threat that people like verison will charge real money to peer with the above companies. This is a direct threat to their business model.
The reason why its not an issue here is that all ISPs in the UK peer at coop exchanges. The concept of charging $17 million dollars to peer just doesn't apply.
ISPs over here wanted to ban torrenting, as 1% of the userbase started racking up 80% of the bandwidth costs. However as ISPs introduced caps and blocks, other popped up offering light traffic shaping instead, or bandwidth caps. The market forces won.
Europe actually does, and net neutrality violations are actually commonplace within Europe. In fact, the EU documents a list of currently on-going violations of net neutrality: http://ec.europa.eu/digital-agenda/en/net-neutrality-challen...
Mobile carriers are the worst offenders in Europe. Prepaid data plans in particular will actively block VoIP connections so as to prevent you from making Skype or other VoIP calls over their networks because they want you to pay for their own minutes.
P2P Throttling is also very commonplace. This is something that Comcast was lambasted for in the past, and as far as I know hasn't happened on a large scale in the US for some time, but it's very common in Europe.
While Europe does have a lot more providers to choose from, you'll often find that there are still monopolies just like in the US. Most of the DSL lines in the UK are owned by BT, and most of the DSL lines in Germany are owned by Telekom, for example. Add to this that cable isn't nearly as widespread in Europe, and you have a situation that's even worse than the usual Cable vs. DSL choice that you have in the US. In Europe your choice is likely limited to only DSL, and only one company provides the copper running to your residence.
P2P will always need to be throttled, and has been since pretty much the start of ADSL. If you want to be uploading gigs of data, at high speed then pay for the correct line. don't use consumer lines.
As I pointed out, if you want uncontended connections, peer at an exchange. They are cheap.
Yes BT has a monopoly, but unlike the US they are highly regulated. They can only charge a certain amount for each line(I think ~£4 a month). They cannot stop arbitrary data for business reasons, all resellers have the same SLA.
The BT network is in much better shape than the US, with a thriving reseller market. This means that ADSL is often thrown in for free with TV.
With mobile data, its only vodaphone that do that kind of thing, as they are very keen for you to burn through your data allowance and run up bills.
Also VOIP over 3g is terrible (300ms round trip) by the way :)
A lot has been accomplished with public awareness and public outcry has repeatedly averted disaster - so far. But the proponents of this come back with another attempt at least once a year, in both Europe and the US, and they have money on their side. It's amazing net neutrality has gone on as long as it did, against vast corporate and political interests.
Whereas net neutrality advocates have to beat this back every time everywhere, their opposition has to be successful only once, and only in one of these regions, for the whole thing to fall.
It is, I'm afraid, just a matter of time.
It's especially hard when so many people are spouting stuff about how the free market will solve this (if I got a dollar every time "free market" was the drooled out answer to a complicated problem...). It's like we have to convince a critical mass of our peers of the benefits of free speech every time the issue comes up.
However, it's still important we fight each battle, each time one rears its ugly head, wherever it happens, with whatever resources we have available. The internet is too powerful a tool to be left in the hands of greedy slimeballs. It's the great equalizer. Anyone can have a voice, and anyone can listen to that voice, whether you're a giant corporation or a 12 year-old kid with a good idea.
It's worth the struggle.
I'm not ideologically opposed to net neutrality regulations - I think they can be a useful tool after you fuck things up, like the US did.
What I dispute is the idea that suddenly everyone has that problem just because the US does. How about letting each country implement its own net neutrality laws if and when they're needed?
The capital expenditure required for downtown areas may not be that high, though. Government may extend the natural monopolies to rich, urban areas by requiring ISPs to wire up poor areas, thus increasing capital expenditure.
Government may make the natural monopoly worse (or better), but the problem is a natural part of the business.
Boy, this is complicated, isn't it? It's almost like "free market" isn't a magic incantation that fixes everything.
a) We don't "win". We just "don't lose". Yet. b) You are doing a great favor to them, thinking they are "idiots". The truth couldn't be further from the truth. They just have different goals than we have.
It's an issue in its infancy now, sure, but it will become a problem.
I think reasonable arrangements can be made that don't destroy some of the best things about the open internet.
And I'm not sure the issue still is in its infancy. It's becoming a problem and it's stopping further innovation. I think we really need to think about "internet 2.0".
The internet has added hosts, users and services every year since 1980. It didn't cross a threshold to need priority blah in 1981, not in 1982 either, not in any of the 33 following years. What's so special about the 36th year? Or the 40th?
Likewise, if an ISP sells me and my neighbour plans that allow anytime 4K netflix, but they aren't able to deliver on that promise, it's the ISP's fault. My neighbour and I aren't assholes just because we both watch our movies in the evening.
Freedom is important. Really important. Everything else, are just excuses.
The "911 calls should not be interrupted by torrents" is such a load of crap argument. It has nothing to do with net neutrality but is all about service level agreements.
With landlines, the main problem here is not that your neighbour could prevent you from making a 911 call by torrenting, but that his torrenting affects your bandwidth at all.
When you buy 10mbit from your ISP, you should _always_ be provided with that speed, and the ISP should take measures to ensure that. Not on a per service level, but on a per customer basis.
But instead ISPs like to sell "up to 25mbit" offers which actually means "0mbit guaranteed, the rest is lottery" because you share that "25mbit" connection with everybody on the street.
Mobile contracts are a bit trickier, because there the topology is unknown and can change anytime, but even here we can calculate a lower bound if we take into account that each lte cell can only support a few hundred [(maybe a few thousand, if you have really fancy tech and own a very wide frequency spectrum)](https://www.linkedin.com/groups/Plz-anyone-help-me-How-11807...) connections at the same time.
Even if we assumed that an lte cell can have an utopian 10.000 _active_ users using every available frequency, we would still be left with 300mbit/10.000=30kbit per user, [which is still enough for that goddamn 911 call.](http://www.erlang.com/bandwidth.html)
I think this is the important part, yes some services probably should be considered special, many people here mention VoIP as an example, but it must be clearly defined in the law, we must not give ISPs the power to decide what is special because the law would be become effectively worthless.
But hey, I'm sure the providers of the services that are already in the law (like VoIP) will be very grateful for this rent you've granted them.
Also, in my mind special would mean that the service is of great public importance, it doesn't seem likely to me for such things just turn up every other day.
We don't have net neutrality laws in Europe right now to get rid of. The alternative is doing nothing, which in my experience is working pretty well. Are you European? Are ISPs in your area trying to slow down some traffic? If not, what problem exactly are you trying to solve?
And even if you are having problems, why do they must be solved at the EU level?
Also, in my mind special would mean that the service is of great public importance, it doesn't seem likely to me for such things just turn up every other day.
Services don't appear and are suddenly recognized as of great importance. They start small and unimportant until they gain enough momentum to be recognized by the public at large. And they'll never will if you enshrine into law the privilege of existing ones.
Look at what happened in the US, net neutrality was not an issue, just like it isn't really in the EU at this time, then services like Hulu/Netflix/Youtube appeared. In the US like in Europe most ISPs are either Cable companies or Telecoms which offer their own IPTV. Now you as a ISP have a competing service using your network, why would you just tolerate that if you don't have to?
Well they didn't, Netflix is now paying off AT&T, Comcast and Verizon so they won't throttle its service. Now in order for your service to compete with Netflix you'll have to pay up too.
How does that further competition?
>And even if you are having problems, why do they must be solved at the EU level?
Why not? How is the EU's common market supposed to work if we just keep implementing things like this on a country by country basis.
> Services don't appear and are suddenly recognized as of great importance. They start small and unimportant until they gain enough momentum to be recognized by the public at large. And they'll never will if you enshrine into law the privilege of existing ones.
Again I don't think this will be an issue, even if you allow for special cases, Netflix should never be special, and neither should probably 99% of services available today. VoIP is special (and I guess is already treated as such by ISPs) because it is a replacement for PSTN which was special and you can't have voice communication be degraded because people are watching the World Cup final or whatever.
Let's put it this way, what would be the harm of having a common net neutrality law in Europe?
Wouldn't that just mean that if you want to do business inside the EU you won't have to worry about being blackmailed by ISPs in 28 different countries?
And yes I live in Europe.
For the same reason I'm paying much less for higher speeds than the average US household. Unlike most places in the US, there's competition in the ISP market in my country. I live in a city of less than 100K, and I have four ISPs offering me fiber.
Why not? How is the EU's common market supposed to work if we just keep implementing things like this on a country by country basis.
How is the common market hampered by this particular regulation being implemented at a national level?
Again I don't think this will be an issue, even if you allow for special cases, Netflix should never be special, and neither should probably 99% of services available today. VoIP is special (and I guess is already treated as such by ISPs) because it is a replacement for PSTN which was special and you can't have voice communication be degraded because people are watching the World Cup final or whatever.
What about real time video communication? What about real time 3D environments, when they appear? Say real time VR becomes feasible, and a school wants to implement remote education for the underclass?
And that's just an example I came up with in 30s. Thinking that all important services that can be invented have been invented already is just silly.
Let's put it this way, what would be the harm of having a common net neutrality law in Europe? Wouldn't that just mean that if you want to do business inside the EU you won't have to worry about being blackmailed by ISPs in 28 different countries?
But do you already?
Laws are important, but they impose costs. Financial, administrative, and potentially worse. It's fine for us to say "No discrimination for traffic except for VoIP!", but a law will be a 30 page document that will still leave plenty for well paid lawyers to discuss and find loopholes over.
Trying to start a small ISP? Better make sure your QoS tables are exactly like the law says, or you might get shutdown if your competitors find out - but when was the last time you saw a law that could be objectively translated into a technical spec?
And then there's the fact that a EU-wide net legislation is a great opportunity for the lobbyists of the media industry to stick their clauses into.
And finally, EU officials have made some worrying claims about censorship "for the children!" for me to trust them with such things.
No, thanks. Let's talk again if and when it becomes a problem, not before.
The future holds more use cases. Remote surgery, autonomous vehicles etc etc.
I can't see how autonomous vehicles necessarily require any kind of priority access to the network. Remote surgery is a pipe dream - it has been discussed 20 years ago and nobody cares. It's certainly not worth destroying the network just for that.
There will certainly exist more and more services in need of low latency. Why stop innovation in this area by just giving up and saying that no such traffic might exist?
If you read the article there is still strong wording to protect the network as such. It is just that certain services needs lower latency and/or higher reliability. This is nothing new. But with the network nowadays full with 70% video you need to handle it.
Do you really think it would be a good idea to perform something as critical as remote surgery over the internet, where every script kiddy with a botnet might literally be able to kill you by DoSing the hospital?
We shouldn't be afraid of new technologies because they have a risk associated with them. We just need to place that risk in context against the risks of existing solutions that we already live with.
In places where it does exist, this wont be the case for long. Telephone lines are facing obsolescence and at first you're going to see them stop being used, and then you're likely to stop seeing structures being built with them.
And in places where it does not already exist, such as lesser developed parts of the world, the only connection available would be a wireless one, where neither a dedicated connection nor transport to a qualified hospital is an option.
I agree with you. I just don't understand why we let technology develop on this curious path tangential to our actual needs.
This is the main argument Marc Cuban uses to justify his opposition to net neutrality and I think it's nonsense: any critical or life-threatening systems should run on dedicated lines just like they do now.
There is nothing about a 'technical level' that implies this. It's all political. Furthermore, it's certainly not up to the service providers: if I find my work is being slowed down by a garrulous teenager on their phone, I'm gonna find a better service that will provide me what I actually pay for.