Hate me all you want, but I think poor Indians would be better off with Free Basics.
I also don't believe in net neutrality. I believe carriers should be able to discriminate data based on economic value of the data.
If we ever desire to see real usages of the Internet(that is more than just information exchange medium) which requires real time data transfer (for instance remote surgeries, remote operation of machines by a human operator, etc etc), then that would require to get rid of net neutrality.
If we ever desire to see real usages of the Internet(that is more than just information exchange medium) which requires real time data transfer (for instance remote surgeries, remote operation of machines by a human operator, etc etc), then that would require to get rid of net neutrality.
I just don't see how this can be the case. Show me your evidence that some data is causing real time data transfers to slow?
In retrospect I shouldn't have used the phrase 'QoS', I now realize that QoS means something very specific. By my argument is for differential pricing. The service provided to higher priced content is prioritization of data.
Real time data needs reliable and fast ping rate. This isn't dictated by the bandwidth, rather by the speed of the network devices. One way to achieve that is introduction of QoS, the network devices prioritize some data over the other.
If you want to observe this effect then try turning QoS on or off on your router during your wife's Netflix session while you're playing a multi-player RPG.
QoS is not the technology used for hospitals, banks, airports, stock brokers and similar industries which need fast ping and guarantied network delivery. They use dedicated bandwidth, where QoS is used to alleviate issues resulting from congestion on shared networks.
Dedicated bandwidth is also what pays for new cables being placed between nations or continents. A new cable can often promise a small improvement in latency compared to the existing ones, which for a stock broker means a major advantage over competitors who use the old ones. The highest paying customers moves to the new one, which result in space being made available in the old cables, which tickles down to lower paying customers until its no longer worth selling dedicated bandwidth (which is normally in the from of separate wavelengths), which at that point it get used by multiple customers at the same time and people start to talk about QoS to artificially increase the number of possible customers on the same available bandwidth.
QoS is about selling bandwidth so it can pass over 100% usage. Dedicated bandwidth is about reliable network with fast latency. QoS is a way to increase the revenue of old cables and increase their lifetime. Dedicated bandwidth is what pays for new cables being placed in the ground or on the bottom of the sea.
Most games I play have pretty steady ping. I am doubtful QoS would decrease ping a lot.
> If you want to observe this effect then try turning QoS on or off on your router during your wife's Netflix session while you're playing a multi-player RPG.
That's because the network is congested and Netflix keeps trying to push as many packets as possible, hence your game packets may not reach on time. But, placing a simple bandwidth limit on other devices is enough to ensure proper ping for your game.
QoS is very, very silly approach to guarantee that a real-time deadline can be met. Over a conventional Ethernet, you can only achieve soft real-time (i.e. video streaming). For hard real-time (when people can literally die if the packet was not delivered in x ms) you need a protocol that has formally proven to have certain characteristics (like EtherCAT).
No, they should (and will) pay for themselves. On the other hand they will subsidize the rest of us. Kinda like how first class subsidizes the economy class. We let them board first, but our ticket becomes cheaper because of that.
>>first class subsidizes the economy class. We let them >>board first, but our ticket becomes cheaper because of >>that.
You mean, like governments in capitalistic nations like the US running at the mercy of some say a 100 billionaires. I don't believe that is a viable model to follow in a egalitarian society of the internet.
Currently, internet is the best example to prove in a different sense that anarchists (virtual)society can survive, lead and guide people to their goals. When people try to build establishment on top of this, thing take a different route altogether.
ISTM the ratios involved are very different. Even a first class air passenger with fifty kilos of baggage only takes up slightly more space and weight than an economy air passenger. Since they might be paying five or more times as much, the service makes sense for airlines. The realtime network you envision is already available to users willing to buy e.g. dark fiber, for less of a premium (relative to bandwidth) than that. If that doesn't work for you, why not?
Yes we know that big ISPs regularly raise examples like this as a reason why they shouldn't have to provide neutral service, but that is manifestly self-serving on their part.
Nothing in net neutrality prevents you from buying a faster pipe where the provider guarantees priority for your packets using QoS or whatever they please.
Net neutrality only prevents your provider from making this decision on your behalf.
So either I upgrade my whole pipeline, or live with a slower pipeline?
I currently run Samsung's Smartthings hub at home(and I love it), I would like the video feed and other notifications be available to me at real time speeds (don't ask why), you're suggesting that I upgrade my whole line for that?
Wouldn't it be preferable that when I use Netflix, the video goes on a deprioritized network, and when I use Skype, it goes on priority network without me having to upgrade the whole service.
The point is, from the consumer's point of view, the services he uses are a mixed bag of priority vs non-priority data. On the other hand, from the content provider's POV, the services they use are either priority, or non-priority (or a mixture, for instance paid subscribers are run through a faster network, etc etc).
Net Neutrality prevents that, not because the conversations I am having in this thread with people is what happens in every NN debate.
> If we ever desire to see real usages of the Internet(that is more than just information exchange medium)
We don't, or rather we see any and all "real usages" of the Internet to be that of (agnostic) information exchange, at their core.
> which requires real time data transfer (for instance remote surgeries, remote operation of machines by a human operator, etc etc), then that would require to get rid of net neutrality.
If you need truly real time behavior like this, then the Internet is not suited for your purposes, Net Neutrality or not.
Wouldn't it be more preferable to build a neutral layer on the top of a non-neutral network, only aimed at information exchange?
I mean what's more preferable/useful, a network where moderation is made illegal on all sites or a network where various sites moderation is based on a site's values and nature, and 4chan being one such (relatively) moderation free site?
> which requires real time data transfer (for instance remote surgeries, remote operation of machines by a human operator, etc etc)
And when a backhoe destroys a fiber line...? Or copper thieves steal cables? Or when an anchor destroys connectivity to yet another country? Datacenter fire takes out the router? AC failure overheats the router? Cosmic ray flips bits? Electrician doing a backup generator / UPS battery upgrade flips the wrong breaker? China posts a bad BGP route? Michael Jackson dies again? Amazon S3 goes down again? Landslides take out cables after earthquake? Windstorm?
How many of these scenarios is traffic shaping really going to fix? Not enough to be the camel-breaking straw, IMO. Even the U.S. power grid isn't reliable enough for this kind of thing - e.g. hospitals roll their own generators as a result.
As for remote operation - it's already happening with mining trucks, via failsafe robotics (which don't require real time data transfer) and dedicated networks (which can be properly secured against single failure points, and be kept small and separate enough to control and mitigate some of those concerns.)
Great, just letting you know it is a fight you cannot win in real life.
I understand that people on HN (and those who support Net Neutrality) have good intentions, unfortunately what they are trying to do is step between two private entities (network providers and data providers) whose business dealings are just (as much as UPS's priority shipping to those who want to pay for it is just).
The fact is, people need to stop and think. Things aren't what they seem to be. The whole Net Neutrality movement has become an echo chamber (to which you contribute by down voting). All this does is piss people off in real life when they see that they are on a losing side of things.
Right now the market thinks that profit can be made by getting rid of Net Neutrality. Once it starts to think that current Internet infrastructure cannot get rid of NN(Say some sort of law is passed), then that means more profitable venture would be a new network which can avoid the regulatory restriction of net neutrality.
I also don't believe in net neutrality. I believe carriers should be able to discriminate data based on economic value of the data.
If we ever desire to see real usages of the Internet(that is more than just information exchange medium) which requires real time data transfer (for instance remote surgeries, remote operation of machines by a human operator, etc etc), then that would require to get rid of net neutrality.