A VoIP call must not be treated worse than an FTP transfer or web traffic with the same or higher bandwidth requirements.
It also means that a high-bandwidth HDTV IPTV stream can not be treated "better" (i.e. lower latency, reserved bandwidth) than an FTP transfer or web traffic. This means 3rd-party IPTV over existing broadband connections will never happen, because who wants television that cuts out every time your kid gets online?
Without net neutrality, the "seller" (the content provider) is the one getting the bill.
In the general case of your average web content provider, I agree with you that billing the content providers for their content makes little sense.
However, consider services such as IPTV. Over the public Internet, with no measures to preserve quality of service, high-bandwidth IPTV streams (such as for high-definition broadcast-quality programming) are in direct competition with bulk file transfers, e-mail, etc., when links become congested. (Your home DSL connection is most likely the location of this congestion any time you start downloading a large file over HTTP or BitTorrent.)
So if you decide to subscribe to HBO over your public Internet connection, and you get your programming streamed to you over DSL, you're basically guaranteed to have your TV program cut out every time you start up BitTorrent or fire up your favorite bandwidth-intensive application.
Of course, we already have a solution to this problem: QoS. It's specifically designed to allow for things like prioritization of packets, bandwidth reservation, etc. Unfortunately, you can't do it over the public Internet, because if we were to trust every random person's QoS requests, it would be immediately abused.
So in order to do QoS properly, you need to work out an agreement with the content provider to set up a dedicated QoS-aware network connection between them and your own network. With end-to-end control of this network, you can now reliably do QoS and allow intermediate routers (right up to your "DSL doorstep") to make prioritization decisions when congestion occurs to guarantee that your IPTV streams aren't degraded every time there's contention.
Who should be required to pay for this new infrastructure? Or do you advocate banning the very thing necessary for 3rd-party IPTV competition over broadband?
(AT&T can do IPTV over their own broadband connections because they already own the network end-to-end and can make QoS decisions that logically separate the IPTV traffic from the rest of your Internet traffic. 3rd parties can't do this, especially with the popular definition of a "neutral" Internet.)
It IS possible to provide CATV-quality service over the public internet IF networks are built up for it. Let's say I'm an ISP and want to do that for my customers. I provide symmetrical 100Mbps connections for them. I oversell 2:1 rather than 100:1 so I advertise truthfully (being that kinda company) 50Mbps burstable to 100.
Like I said in my earlier post, it doesn't matter how fat you make the connections. There will be a bottleneck somewhere along that path. So I have symmetrical 100Mbits to play with. I flip on HBO-over-IPTV and settle down to watch a movie. Back to the Future is on, and I think, "Hey, I'd like to see the sequels to that." So I fire up BitTorrent and start downloading these off of some random Internet site. This doesn't take very long at 100Mbits, but for the sake of argument let's say that it takes a full minute to download. During this entire time, BitTorrent is attempting to saturate my network connection with packets. It's doing what it was designed to do: give me the best possible download time. So after a few seconds, it's ramped up to nearly 100Mbits, at which point my network connection starts to become congested.
Pop quiz: what happens next?
Without QoS, all services are degraded equally, and I lose HBO for a while. This is not "CATV-quality" IPTV service. When I fire up BitTorrent on my cable modem today, I don't lose my TV signal. It doesn't matter how fast you make your connections or your uplinks. So long as you have applications that are geared toward saturating that connection, and no ability to prioritize, it simply can't be done. You might be able to get your end user connections up fast enough that most of your users simply won't be able to find enough data online to download to keep their connection saturated for more than a fraction of a second. You'd side-step the problem because your IPTV stream wouldn't be interrupted for long enough to matter, but this is an arm's race. With all of this bandwidth available to them, content providers will find ways of utilizing it, and then you'll be back where you started.
Otherwise, I agree with most everything else that you're saying.
That's why client side QoS is limited. It cannot throttle UDP traffic, but it can "wag the dog" for TCP connections by closing the window down to a packet or two and delaying ACKs to approximate the desired incoming packet rate.
Until that problem is solved for all types of bandwidth-unconstrained traffic users are likely to see, it's not solved. So great, we can sort of do some hand waving with TCP and hopefully get it to not kill my IPTV stream, but what about other types of traffic? Users aren't going to say, "Ho-hum, my HBO is cutting out because I started up application XYZ. I should have known that would happen." They're going to say, "Dammit, my IPTV service from provider ABC is crap! Every time I start up XYZ it cuts out. My AT&T service doesn't do that."
Of course, many ISPs are currently claiming to provide a service adequate for CATV quality even though they not only won't but can't
I suspect something is getting lost in translation here. ISPs can do this if they are the ones providing the service. Since they own the network, they can do all sorts of QoS magic to guarantee bandwidth and latency for their IPTV services right to the customer's premises. There's absolutely no way they can guarantee CATV-quality IPTV for 3rd-party IPTV streams over the public Internet. I'd be interested in seeing the exact wording of this.
Since large carriers certainly seem to be willing to spend for new routers to let them double dip...
I think this is the issue though, unless I'm grossly misunderstanding things. Major ISPs are willing to build out the infrastructure for a QoS-enabled path from content provider to end user. They just expect the content providers to cough up the money to do it. The ISPs aren't going to build out new infrastructure/lines for a trusted QoS path unless they're getting paid for it.
There is no Qos or congestion control that can make 80Kbps adequate even with multi-casting once phone service moves to VoIP (consider a 2 teen family) especially when people will inevitably want a webcam feed to go with it soon enough.
VoIP and webcam-style videoconferencing could work fine in bandwidth-constrained environments. Here it's the prioritization features of QoS, not bandwidth reservation, that makes the big difference. You delay BitTorrent packets to ensure that your VoIP packets arrive in a timely fashion. If you've reached the point where queuing is insufficient and you have to drop packets, it might be OK to drop VoIP and video packets alongside BitTorrent packets. VoIP and video recover with some degraded quality, and other services retransmit.
Large scale content-neutral caching at the ISP level would help a lot for other sorts of content (and will STILL be useful even with $9/Mb upstream) but of course, you can't cache a real time videophone conversation.
Agreed, and a good point. But we already have content-neutral caching technology in the form of caching HTTP proxies. But nobody likes proxies because occasionally they're perceived as being slow, and occasionally perceived as delivering stale content. So everybody shift-reloads and all applications add "no-cache" headers to explicitly forbid caching of their pages. Imagine the impact if every ISP had a reliable caching HTTP proxy farm for their customers, and all of their customers used it (kept it fresh)?
The one and only reason net-neutrality advocates object to QoS and congestion control is that they are WELL aware that ISPs will use it primarily to double dip and avoid getting more upstream even if the price drops.
... There isn't a well utilized router anywhere on the net that won't have to be upgraded significantly before it will have the raw power and buffer space needed to actually do QoS.
I think the double-dipping fear is semi-founded, but consider that you can't do QoS over the (untrusted) public Internet. It will be abused without mercy. If you have some 3rd-party IPTV service that wants CATV-quality service to its customers, and you want to use QoS to make that happen, you have to contract with the individual ISPs and get dedicated connections or, at the very least, arrangements to honor QoS flags from those sources. Who's going to pay for that?
A GOOD way to deploy QoS is on the client side.
By the time those IPTV packets have arrived at the customer's premises, it's already traversed the bottleneck (e.g. their DSL connection) and whatever "damage" that could potentially be done to those packets has already been done. QoS decisions must occur upstream of the congestion in order to be of any practical use, which means the ISP needs to be involved.
I do agree, though, that QoS decisions should be made by the customer. If I want to have my IPTV packets prioritized over BitTorrent, it would be nice to be able to communicate that preference to my ISP's routers, which would then respect my wishes when the time came to choose. That way the ISP can't be accused of making these decisions secretly or without my input, or in such a way that its competitors are locked out regardless of my wishes. But this only covers that last hop. It does nothing to deal with congested uplinks to the ISP's peers. You need special arrangements with the content providers to guarantee prioritized handling. But, oops, that's not a neutral Internet anymore.
3rd-party CATV-quality IPTV over DSL can only work if the DSL provider tells its routers to prioritize your 3rd-party IPTV stream over your BitTorrent (or whatever) traffic. This will never happen with the status quo, and it will never happen in a truly neutral network. The only way it's going to work is if you give either side the thing that the other side doesn't want them to have. The ISP needs the technical capability of prioritizing packets, but they need to be prohibited from doing it arbitrarily or abusively. Allowing the user to CHOOSE what types of traffic to prioritize over what other types of traffic is a step in the right direction, but technology to do this doesn't exist yet. You can do bandwidth reservations and the like, but you can't propagate that over the public Internet because it will be abused. So you have to do it over private, dedicated network connections set up between IPTV content provider and DSL consumer, and who's going to pay for that? Re-enter the net neutrality debate...
Scaling up bandwidth isn't the final solution, though. As backbones scale up, so do end user connections. You're always going to have bottlenecks. If I'm downloading a DVD over the Internet, that bottleneck is usually my DSL connection. If a thousand people spontaneously decide to download the same DVD (assuming simple unicast), the bottleneck might be at the server's end.
Unless you know of a way to transfer data infinitely fast, these bottlenecks will always exist and will always constrain bandwidth. The trick isn't an arm's race, it's deploying technologies like QoS to allow services to work as anticipated *despite* a congested link. When your 10Gbit SuperDSL line becomes fully utilized (aka congested) with a BitTorrent transfer, you want your 10Mbit IPTV stream to proceed without any packets getting delayed too severely, right? You have to prioritize. That's what QoS is for.
Except the extreme net neutrality crowd doesn't want to allow that.
This problem only exists because some people think the idea of congestion control is evil. (Or people think the idea of providers arbitrarily *degrading* service is evil, but the proposed solution also outlaws legitimate congestion control.) Scaling up bandwidth doesn't solve the problem because certain types of bandwidth-unconstrained services (such as BitTorrent or even a simple HTTP or FTP transfer) will attempt to use as much bandwidth as it can between the hosts. You're going to have a bottleneck (congestion) somewhere. "Infinite bandwidth" is silly. "Lots of idle bandwidth" is stupid and expensive.
If an employee of a company goes out and murders someone, the company is not automatically* held "100.0% responsible" for murder. If an employee uses the company's Internet access to solicit sex from minors, the company is not automatically held "100.0% responsible". The company is normally responsible only if the employee is doing something that's part of their job. If it's not part of their job (the company didn't ask them to do it), why should the company be held responsible for it?
* Bear in mind that when you see lawsuits where employees do something near company premises, or using a company vehicle, or anything where a lawyer can make the smallest connection to a big business, the plaintiff's incentive is to go after the deep pockets, not to apply some sound legal principle that generally holds employers responsible for the acts of their employees. If you can convince a jury that the company is even 1% responsible, you can now go after the business for far more money than you'd ever hope to get from the individual that actually committed the act. But this is far from automatic and usually based on the lawyer's ability to convince a jury.
I suspect many of these edits are made by employees not acting on official instructions. Many of these people like where they work, or otherwise feel the need to defend their employer when the opportunity arises. These could just be well-intentioned but short-sighted employees acting unilaterally. If your company is large enough, you're bound to have at least one person able and willing to do something like this. You can't entirely fault the company for it.
Though I imagine there will be some Wikipedia guidelines appearing at some companies in the near future.
How hard would it be to have a message "could not send, content of message is not allowed"?
I don't know, but I bet it would cost more to present the user with that message than it would cost to simply drop the messages on the floor. I absolutely agree that this is a poor user experience, but I have no clue how easy or difficult it would be to make it better. Do you? A good manager would then compare that cost with the estimated impact on the users. How many users send legitimate messages that would be caught by this filter? Enough to matter? Do you have all of the facts here that would qualify you to make that decision?
For one thing, it also degrades the usabiltiy of the service for legitimate purposes. For another, it doesn't _actually_ stop the malware.
Sure, it degrades the usability of the service somewhat. Just like "barrier" that is airport security degrades the usability of airline travel. It does, in fact, stop malware that uses that vector for propagation. Is it a permanent solution to the malware problem? Of course not. As you say, the people creating malware will adapt, and the "fix" would be ineffective against malware that infects through some other mechanism. But the point of my post was that this could be a short-term fix while a long-term solution is under development. It would completely eliminate the ability of this malware to spread while giving the company the necessary time to fix, test and deploy a solution without operating in "emergency mode" all the time.
Yes, but that's not what's happening.
And you know this.. how?
I don't understand why people act as though the concept of a cost-benefit analysis is some strange or evil thing Stop acting like a rabid anti-Microsoft troll for a moment and look at the problem rationally, from a business perspective. Businesses must be cost-effective in everything that they do. Pulling out all of the stops and blowing wads of cash all the time will put you out of business fast, and then where would your customers be?
Oh, and did you realize that this censoring (which really has been going on for months if not years now) can also be used as a stepping stone to censoring things that Microsoft considers harmful, even if the users would likely find them bona-fide? I've already had several of my messages blocked by the filters, and I assure you they did not in any way relate to malware. Perhaps a few cases of open-source software, though.
Personally, I'd rather they fix the vulnerabilities that make those strings dangerous in the first place: it's not like IM is the only place a URL can get on your machine.
Do you really think they're diverting resources away from fixing bugs so that they can add "censorship" features to IM? Perhaps this is just one effort among multiple efforts to correct problems AND mitigate their effects? If it's going to take X weeks to fix the bug, but Y days to implement a filter that will stop some large percentage of infections, don't you think that both avenues are worth exploration at the same time? There's more to slowing and preventing the spread of malware than fixing the defect that allows them to propagate.
This also assumes that the same organization even owns the bug in question. Not all of these defects may be Microsoft's problem to begin with. This might even be a MORE reasonable action for them to take, since they're doing "everything in their power" to fight the problem rather than just sitting on their hands waiting for a 3rd-party to correct their bug, and sitting on their hands longer waiting for the end user to update their software.
Seriously... Who benefits from a state granted monopoly to telecoms? Have you seen that work well lately?
A voluntary creation of a monopoly where none previously exists ought to benefit both the municipality and the telecom. This is little more than a contract codified as legislation. If it doesn't benefit both sides, why in the world would both sides agree to it?
How about the power of the people to decide for themselves what service they want?
Taken to an extreme, this is a Direct Democracy, and doesn't work at large scales. We rely on our governments to make decisions for the greater whole that some people may not agree with. One state with all-out competition might have broadband in a community in 3 years, at an average of $50/month. Another state might have a community that decided to contract out with a telecom to get broadband there sooner for a modest tax break in 1 year, at a cost of $60/month. Another community might think that's outrageous, so they wait 4 years. A third state might decide that they want universal coverage, so they set milestones and use tax breaks as incentives.
The point is, US states have traditionally been able to experiment with various approaches to solving problems. Other states can then see what works and what doesn't, and apply what they've learned. And then you have to deal with the fact that what works for one state may not work for another.
If you have a problem with telecom competition in your state, and you think it's a bad thing that municipalities can't run their own broadband, isn't that something you could just as easily talk with your state about? Why did your state create that ban? Why insist on going over their heads to the federal government when you don't get your way? The people of your state have spoken, and your personal opinion was deemed not to be in line with the desires of The People of Your State. Deal with it. Stop working to take their ability to self-govern away.
The state governments are presumably acting on something. Nobody thinks to themselves, "hey, let's ban municipal broadband, because, I don't know, it's not cool!" Usually you have a telecom making a case and/or an offer to the government. If the government thinks the case/offer is reasonable, they may pass legislation that looks like a "ban" here.
In this case, the federal government isn't acting on anything except an ideological conflict. People think, "bans on municipal broadband are bad mmk, so let's create federal legislation to prevent them!" No thought is given to why those governments chose to create those "bans". Maybe the state was getting something in return that more than made up for the apparent "loss" that a ban would create? All everyone can think of is "screw the evil telecoms!"
One of the reasons we have semi-autonomous states instead of just one giant federal government ruling everything is that the states can experiment and try things that their neighbors aren't doing. This lets us see what works and what doesn't work. Everyone wants to homogenize all laws across all states, so they go directly to the federal government to Get Things Done, but really this is the wrong approach. If you don't like that your state has banned municipal broadband, and don't like the reasons they chose to do that, take that up with your state. We're all supposedly pro-competition, right? So let's try some different things and let's see which state's approach works.
Except we're talking about bans on municipal broadband. These aren't federal bans, they're state (and maybe county?). These governments probably wouldn't ban municipal broadband unless they felt it would be bad for their community, or if the telco agreed to do something for the community in exchange for the ban.
I say let municipalities and states alike have the flexibility to do whatever it is that they want in this regard. The federal government has no business sticking its nose here.
The problem is that while it would work in theory, negotiating contracts with every individual municipality will kill any buildout simply through delay.
This is certainly a good and interesting point, but it seems somewhat orthogonal to my post. You seem to be going beyond simply prohibiting bans on municipal broadband, and are advocating for something more.
While some federal "standards" on broadband roll-outs would help to push broadband into areas it doesn't exist yet, there are some monumental obstacles that you'd require telcos to overcome here. Your suggestion that 40% of households have the highest tier of service within 5 years of a commercial offering might actually make it impossible for any but the incumbent provider to offer service. Shouldn't we be pushing for more competition here rather than setting high barriers to new deployments? But really, my point here is that every market is going to be different. Every community is going to be different. One Federal Standard might be completely inappropriate for a certain state or county. Why not let that state or county decide the best approach for its own citizens? If one state/county wants to ban municipal broadband, but a neighbor doesn't, it should become pretty clear pretty quickly which approach yields the best service and price for the customers.
By all means, the federal government should be there to make things easier to do across state lines. But that should never mean that the federal government should just trump existing, properly enacted, properly negotiated state law without a damn good reason. Those states and municipalities have reasons for the laws they enact, and even if you don't agree with them, it's a bit arrogant to tell them they're all wrong and they should be running their affairs differently. Maybe their legislators are all corrupt and in the pockets of the telecoms, I don't know, but even if that's the case, that's their problem to deal with, not the federal government's.
And that may be a perfectly valid thing to be upset about. But if no penalties were prescribed in the law, that was a kind of stupid thing for the government to do, yes? If you were going to sell someone your house, would you just give it to them and assume they'd pay you, or would you make payment a condition of the sale?
If a business has to pay some new tax that wasn't there before, that's a cost of doing business. Somehow, that cost is going to be passed onto the consumer. Usually you have to raise prices, or drop out of the market if you can't be competitive. Here, they got permission to make it its own separate line item. You'll pay for it one way or the other, but by making it a separate fee, they don't have to *appear* to raise their prices and can just be done with the whole mess.
If your local government entered into a contract with your local telecom to develop services like higher-speed broadband for your citizens, with tax dollars, and your telecom failed to do that, surely the contract had penalties spelled out for that eventuality? If those penalties were insufficient or nonexistent, that's your community's fault.
Something else to consider is developing municipal broadband infrastructure with tax dollars, but contracting out the maintenance and management of that to a telecom. That way they're not really "in competition" and the contract could be equally profitable for both sides.
You could even hire the local telecom to do the actual build-out, with the understanding that it would be owned by the municipality. If the telecom can't get favorable terms, they don't have to enter into the contract to do the work.
If you're worried that the telecom is going to do shitty work or sabotage the system, there are things like "inspections" and "licenses" that work well in other industries. Have an independent party verify that things were built properly, have network plans be approved by the municipality before they're implemented, etc.
I say let municipalities and states alike have the flexibility to do whatever it is that they want in this regard. The federal government has no business sticking its nose here. If a state wants to enter into some sort of agreement with a telecom headquartered there, and that agreement means that municipalities shouldn't be allowed to do their own thing, but in return, the state gets discounted broadband, universal coverage, and more jobs, that might be a perfectly reasonable thing for that state to do. If its citizens don't like it, they need to make themselves heard. But even so, it might actually be a good agreement for the state. You don't know that. We need to try to look past our anti-telecom bias here and not automatically assume that everything they do is evil.
Usually people on Slashdot are all about democracy, but they fail to realize that a strong federal government, from the state's perspective, is the anti-democracy. It's overriding what are effectively the desires of a sovereign state. If there was a valid reason for this ban on bans to be applied at the federal level, that might be OK, but just because you can justify the law under the Commerce Clause doesn't mean it's the right thing to do at the federal level.
I hate the Commerce Clause. There should be an additional requirement that you should be able to justify why a bill should exist at the Federal layer with a straight face.
If a state really prides itself on the telecom companies that live within it, and maybe those telecom companies would give special discounts if they were the ones providing service to a municipality, it might be perfectly reasonable for that state to enact legislation that is perceived as a "ban" here. If it benefits their local telecom, then it benefits their local community, because lots of people there work for that telecom.
Of course, this makes no sense for other states that may not have those interests. But why prohibit it for everyone? Why do these "bans" exist in the first place? Because of corrupt politicians in the pockets of the telecoms? Or could there perhaps be more to these pieces of legislation, that, when understood, make the "bans" seem a little more reasonable? But I guess meddling in the affairs of lower governments, under the guise of doing something that, on its face, seems like a good thing, is all we care about? "Yay, they're really Getting Things Done in Congress! I'm going to vote to re-elect these guys!"
So you admit that it's a mistake, and in the same breath effectively say that it's OK to keep it. You're advocating that one person take advantage of another. I disagree. Would it be criminal for you to keep the money? Probably not. But you don't have to commit a crime for the court to require that you return it through a civil suit.
Multiply this mistake by 10 or 100. The party taken advantage of takes you to court, and you admit that it was an error. Who do you think the court is going to side with?
This isn't something that could be confused for a tip or interest. This isn't even cash given to you by an agent of the company that would be authorized to give you "bonus cash". This is a machine that you know is only supposed to give you what you asked for. It shocks me to read that there are people that actually think it's fair to keep that money.
Not all banks are evil megacorps. Not all ATM companies are evil megacorps. I could go out and buy an ATM for a few thousand dollars, put it somewhere useful, and make money off of the fees. For all you know this is one of those ATMs operated by some grandmother somewhere as an investment.
Yes, he's saying "two wrongs make a right". As in, "I think this behavior is evil. I think some corporations exhibit this behavior. Therefore, all corporations are evil and deserve to be taken advantage of, [regardless of what the law says]." I was attacking the premise. I didn't specifically address the "two wrongs make a right" fallacy because it's obvious.
In other words, yes, you believe corporations are evil and that people should be allowed to take advantage of them. Your rationale behind this belief is, of course, dubious, since individuals can be assholes too:
"Oh, your car broke down? So effing what? I have to get home and watch Simpsons, so I won't stop for you." "Oh, you have extra bills to pay? Sorry about your luck. I'm going to leave without telling you that I dented your car door." "Oh, you say you need at least $11/hour to live? Tough crap. I'm going to shoplift from your little store anyway."
"Oh, you're already just skimming your bills? Tough crap. We want this big construction project in our state, and it's too expensive to be worth it, so we want our representative to push for the federal government to pay for it instead, spreading the cost out over everyone else."
The world is full of people taking advantage of each other. It's easy to point the finger at Big Business, but it pays to remember that Big Business is run by people. We get through life in a world full of assholes by having laws that protect us from those that take things too far. The law doesn't care whether or not you like someone. It cares that you're taking something that does not belong to you, and refusing to give it back.
...among those that are interested in this topic. You and I know who they are. But ask anyone on the street who the RIAA is, and they'll either shrug their shoulders or say, "They're the guys that sue you when you steal songs off the Internet!"
Which is exactly what the RIAA wants to hear. Details like who the RIAA represents, the fact that mere downloaders aren't sued, only those redistributing, and that infringement isn't theft, are lost on the general public. The only people changing their habits are those that have bought what the RIAA wanted them to buy, or those that have researched the issue and are avoiding giving RIAA members their business. But they're all in a very tiny minority, I suspect. Had every RIAA headline over the last few years said "Sony" instead of "RIAA", or, even better, a specific band, I rather suspect they would feel the backlash.
Who isn't the RIAA's customer? It's not like this is a business that is trying to avoid losing customers to a competitor. The general public has no clue what the RIAA does, who they represent, or who its "competitors" are.
If the individual music labels were the ones in the news, that might be an entirely different matter. The beauty of doing all of this through the RIAA is that none of the RIAA's members are in the light. Yes, some of us know who they are, but when someone reads something in the news about the RIAA suing people downloading music, they're not thinking about the labels behind it. (And even if they do, how many are going to keep up their little protest when So-And-So's latest album just came out, and it's REALLY good, like omg!)
their approach does not necessarily target the guilty, and the innocent have almost as much incentive to settle as the guilty.
I realize I'm nitpicking here, but these are civil cases where there is no guilt or innocence. You're only liable or not liable for damages. This is also why it seems odd that they'd be pursuing cases against people that are only indirectly involved. Only extreme cases of copyright infringement are crimes, and it wouldn't be the RIAA going after you, it would be your own government.
The RIAA's tactics are no different from any other Big Business or malcontent exploiting America's legal system. The fact that you can sue anyone for any reason, and literally extort settlements because it's cheaper to pay a few thousand dollars than defend even utter nonsense at a trial is heartbreaking. Larger businesses, taking every precaution to avoid being sued (look at all of the warning labels!) nevertheless have to set aside a sizable portion of their budget just for these nuisance settlements. It's become just another cost of doing business in the US. And we all pay for it in the end, because these costs will influence whether or not these businesses can price their products competitively. And if they can't, that's less competition, which means fewer jobs and higher prices from those that are left.
Slashdot Mirror
It also means that a high-bandwidth HDTV IPTV stream can not be treated "better" (i.e. lower latency, reserved bandwidth) than an FTP transfer or web traffic. This means 3rd-party IPTV over existing broadband connections will never happen, because who wants television that cuts out every time your kid gets online?
In the general case of your average web content provider, I agree with you that billing the content providers for their content makes little sense.
However, consider services such as IPTV. Over the public Internet, with no measures to preserve quality of service, high-bandwidth IPTV streams (such as for high-definition broadcast-quality programming) are in direct competition with bulk file transfers, e-mail, etc., when links become congested. (Your home DSL connection is most likely the location of this congestion any time you start downloading a large file over HTTP or BitTorrent.)
So if you decide to subscribe to HBO over your public Internet connection, and you get your programming streamed to you over DSL, you're basically guaranteed to have your TV program cut out every time you start up BitTorrent or fire up your favorite bandwidth-intensive application.
Of course, we already have a solution to this problem: QoS. It's specifically designed to allow for things like prioritization of packets, bandwidth reservation, etc. Unfortunately, you can't do it over the public Internet, because if we were to trust every random person's QoS requests, it would be immediately abused.
So in order to do QoS properly, you need to work out an agreement with the content provider to set up a dedicated QoS-aware network connection between them and your own network. With end-to-end control of this network, you can now reliably do QoS and allow intermediate routers (right up to your "DSL doorstep") to make prioritization decisions when congestion occurs to guarantee that your IPTV streams aren't degraded every time there's contention.
Who should be required to pay for this new infrastructure? Or do you advocate banning the very thing necessary for 3rd-party IPTV competition over broadband?
(AT&T can do IPTV over their own broadband connections because they already own the network end-to-end and can make QoS decisions that logically separate the IPTV traffic from the rest of your Internet traffic. 3rd parties can't do this, especially with the popular definition of a "neutral" Internet.)
You can adjust the date and time within the Sky mode.
Like I said in my earlier post, it doesn't matter how fat you make the connections. There will be a bottleneck somewhere along that path. So I have symmetrical 100Mbits to play with. I flip on HBO-over-IPTV and settle down to watch a movie. Back to the Future is on, and I think, "Hey, I'd like to see the sequels to that." So I fire up BitTorrent and start downloading these off of some random Internet site. This doesn't take very long at 100Mbits, but for the sake of argument let's say that it takes a full minute to download. During this entire time, BitTorrent is attempting to saturate my network connection with packets. It's doing what it was designed to do: give me the best possible download time. So after a few seconds, it's ramped up to nearly 100Mbits, at which point my network connection starts to become congested.
Pop quiz: what happens next?
Without QoS, all services are degraded equally, and I lose HBO for a while. This is not "CATV-quality" IPTV service. When I fire up BitTorrent on my cable modem today, I don't lose my TV signal. It doesn't matter how fast you make your connections or your uplinks. So long as you have applications that are geared toward saturating that connection, and no ability to prioritize, it simply can't be done. You might be able to get your end user connections up fast enough that most of your users simply won't be able to find enough data online to download to keep their connection saturated for more than a fraction of a second. You'd side-step the problem because your IPTV stream wouldn't be interrupted for long enough to matter, but this is an arm's race. With all of this bandwidth available to them, content providers will find ways of utilizing it, and then you'll be back where you started.
Otherwise, I agree with most everything else that you're saying.
Until that problem is solved for all types of bandwidth-unconstrained traffic users are likely to see, it's not solved. So great, we can sort of do some hand waving with TCP and hopefully get it to not kill my IPTV stream, but what about other types of traffic? Users aren't going to say, "Ho-hum, my HBO is cutting out because I started up application XYZ. I should have known that would happen." They're going to say, "Dammit, my IPTV service from provider ABC is crap! Every time I start up XYZ it cuts out. My AT&T service doesn't do that."
I suspect something is getting lost in translation here. ISPs can do this if they are the ones providing the service. Since they own the network, they can do all sorts of QoS magic to guarantee bandwidth and latency for their IPTV services right to the customer's premises. There's absolutely no way they can guarantee CATV-quality IPTV for 3rd-party IPTV streams over the public Internet. I'd be interested in seeing the exact wording of this.
I think this is the issue though, unless I'm grossly misunderstanding things. Major ISPs are willing to build out the infrastructure for a QoS-enabled path from content provider to end user. They just expect the content providers to cough up the money to do it. The ISPs aren't going to build out new infrastructure/lines for a trusted QoS path unless they're getting paid for it.
VoIP and webcam-style videoconferencing could work fine in bandwidth-constrained environments. Here it's the prioritization features of QoS, not bandwidth reservation, that makes the big difference. You delay BitTorrent packets to ensure that your VoIP packets arrive in a timely fashion. If you've reached the point where queuing is insufficient and you have to drop packets, it might be OK to drop VoIP and video packets alongside BitTorrent packets. VoIP and video recover with some degraded quality, and other services retransmit.
Agreed, and a good point. But we already have content-neutral caching technology in the form of caching HTTP proxies. But nobody likes proxies because occasionally they're perceived as being slow, and occasionally perceived as delivering stale content. So everybody shift-reloads and all applications add "no-cache" headers to explicitly forbid caching of their pages. Imagine the impact if every ISP had a reliable caching HTTP proxy farm for their customers, and all of their customers used it (kept it fresh)?
I think the double-dipping fear is semi-founded, but consider that you can't do QoS over the (untrusted) public Internet. It will be abused without mercy. If you have some 3rd-party IPTV service that wants CATV-quality service to its customers, and you want to use QoS to make that happen, you have to contract with the individual ISPs and get dedicated connections or, at the very least, arrangements to honor QoS flags from those sources. Who's going to pay for that?
By the time those IPTV packets have arrived at the customer's premises, it's already traversed the bottleneck (e.g. their DSL connection) and whatever "damage" that could potentially be done to those packets has already been done. QoS decisions must occur upstream of the congestion in order to be of any practical use, which means the ISP needs to be involved.
I do agree, though, that QoS decisions should be made by the customer. If I want to have my IPTV packets prioritized over BitTorrent, it would be nice to be able to communicate that preference to my ISP's routers, which would then respect my wishes when the time came to choose. That way the ISP can't be accused of making these decisions secretly or without my input, or in such a way that its competitors are locked out regardless of my wishes. But this only covers that last hop. It does nothing to deal with congested uplinks to the ISP's peers. You need special arrangements with the content providers to guarantee prioritized handling. But, oops, that's not a neutral Internet anymore.
3rd-party CATV-quality IPTV over DSL can only work if the DSL provider tells its routers to prioritize your 3rd-party IPTV stream over your BitTorrent (or whatever) traffic. This will never happen with the status quo, and it will never happen in a truly neutral network. The only way it's going to work is if you give either side the thing that the other side doesn't want them to have. The ISP needs the technical capability of prioritizing packets, but they need to be prohibited from doing it arbitrarily or abusively. Allowing the user to CHOOSE what types of traffic to prioritize over what other types of traffic is a step in the right direction, but technology to do this doesn't exist yet. You can do bandwidth reservations and the like, but you can't propagate that over the public Internet because it will be abused. So you have to do it over private, dedicated network connections set up between IPTV content provider and DSL consumer, and who's going to pay for that? Re-enter the net neutrality debate...
Scaling up bandwidth isn't the final solution, though. As backbones scale up, so do end user connections. You're always going to have bottlenecks. If I'm downloading a DVD over the Internet, that bottleneck is usually my DSL connection. If a thousand people spontaneously decide to download the same DVD (assuming simple unicast), the bottleneck might be at the server's end.
Unless you know of a way to transfer data infinitely fast, these bottlenecks will always exist and will always constrain bandwidth. The trick isn't an arm's race, it's deploying technologies like QoS to allow services to work as anticipated *despite* a congested link. When your 10Gbit SuperDSL line becomes fully utilized (aka congested) with a BitTorrent transfer, you want your 10Mbit IPTV stream to proceed without any packets getting delayed too severely, right? You have to prioritize. That's what QoS is for.
Except the extreme net neutrality crowd doesn't want to allow that.
This problem only exists because some people think the idea of congestion control is evil. (Or people think the idea of providers arbitrarily *degrading* service is evil, but the proposed solution also outlaws legitimate congestion control.) Scaling up bandwidth doesn't solve the problem because certain types of bandwidth-unconstrained services (such as BitTorrent or even a simple HTTP or FTP transfer) will attempt to use as much bandwidth as it can between the hosts. You're going to have a bottleneck (congestion) somewhere. "Infinite bandwidth" is silly. "Lots of idle bandwidth" is stupid and expensive.
If an employee of a company goes out and murders someone, the company is not automatically* held "100.0% responsible" for murder. If an employee uses the company's Internet access to solicit sex from minors, the company is not automatically held "100.0% responsible". The company is normally responsible only if the employee is doing something that's part of their job. If it's not part of their job (the company didn't ask them to do it), why should the company be held responsible for it?
* Bear in mind that when you see lawsuits where employees do something near company premises, or using a company vehicle, or anything where a lawyer can make the smallest connection to a big business, the plaintiff's incentive is to go after the deep pockets, not to apply some sound legal principle that generally holds employers responsible for the acts of their employees. If you can convince a jury that the company is even 1% responsible, you can now go after the business for far more money than you'd ever hope to get from the individual that actually committed the act. But this is far from automatic and usually based on the lawyer's ability to convince a jury.
I suspect many of these edits are made by employees not acting on official instructions. Many of these people like where they work, or otherwise feel the need to defend their employer when the opportunity arises. These could just be well-intentioned but short-sighted employees acting unilaterally. If your company is large enough, you're bound to have at least one person able and willing to do something like this. You can't entirely fault the company for it.
Though I imagine there will be some Wikipedia guidelines appearing at some companies in the near future.
I don't know, but I bet it would cost more to present the user with that message than it would cost to simply drop the messages on the floor. I absolutely agree that this is a poor user experience, but I have no clue how easy or difficult it would be to make it better. Do you? A good manager would then compare that cost with the estimated impact on the users. How many users send legitimate messages that would be caught by this filter? Enough to matter? Do you have all of the facts here that would qualify you to make that decision?
Sure, it degrades the usability of the service somewhat. Just like "barrier" that is airport security degrades the usability of airline travel. It does, in fact, stop malware that uses that vector for propagation. Is it a permanent solution to the malware problem? Of course not. As you say, the people creating malware will adapt, and the "fix" would be ineffective against malware that infects through some other mechanism. But the point of my post was that this could be a short-term fix while a long-term solution is under development. It would completely eliminate the ability of this malware to spread while giving the company the necessary time to fix, test and deploy a solution without operating in "emergency mode" all the time.
And you know this.. how?
I don't understand why people act as though the concept of a cost-benefit analysis is some strange or evil thing Stop acting like a rabid anti-Microsoft troll for a moment and look at the problem rationally, from a business perspective. Businesses must be cost-effective in everything that they do. Pulling out all of the stops and blowing wads of cash all the time will put you out of business fast, and then where would your customers be?
Conspiracy theory.
Do you really think they're diverting resources away from fixing bugs so that they can add "censorship" features to IM? Perhaps this is just one effort among multiple efforts to correct problems AND mitigate their effects? If it's going to take X weeks to fix the bug, but Y days to implement a filter that will stop some large percentage of infections, don't you think that both avenues are worth exploration at the same time? There's more to slowing and preventing the spread of malware than fixing the defect that allows them to propagate.
This also assumes that the same organization even owns the bug in question. Not all of these defects may be Microsoft's problem to begin with. This might even be a MORE reasonable action for them to take, since they're doing "everything in their power" to fight the problem rather than just sitting on their hands waiting for a 3rd-party to correct their bug, and sitting on their hands longer waiting for the end user to update their software.
A voluntary creation of a monopoly where none previously exists ought to benefit both the municipality and the telecom. This is little more than a contract codified as legislation. If it doesn't benefit both sides, why in the world would both sides agree to it?
Taken to an extreme, this is a Direct Democracy, and doesn't work at large scales. We rely on our governments to make decisions for the greater whole that some people may not agree with. One state with all-out competition might have broadband in a community in 3 years, at an average of $50/month. Another state might have a community that decided to contract out with a telecom to get broadband there sooner for a modest tax break in 1 year, at a cost of $60/month. Another community might think that's outrageous, so they wait 4 years. A third state might decide that they want universal coverage, so they set milestones and use tax breaks as incentives.
The point is, US states have traditionally been able to experiment with various approaches to solving problems. Other states can then see what works and what doesn't, and apply what they've learned. And then you have to deal with the fact that what works for one state may not work for another.
If you have a problem with telecom competition in your state, and you think it's a bad thing that municipalities can't run their own broadband, isn't that something you could just as easily talk with your state about? Why did your state create that ban? Why insist on going over their heads to the federal government when you don't get your way? The people of your state have spoken, and your personal opinion was deemed not to be in line with the desires of The People of Your State. Deal with it. Stop working to take their ability to self-govern away.
The state governments are presumably acting on something. Nobody thinks to themselves, "hey, let's ban municipal broadband, because, I don't know, it's not cool!" Usually you have a telecom making a case and/or an offer to the government. If the government thinks the case/offer is reasonable, they may pass legislation that looks like a "ban" here.
In this case, the federal government isn't acting on anything except an ideological conflict. People think, "bans on municipal broadband are bad mmk, so let's create federal legislation to prevent them!" No thought is given to why those governments chose to create those "bans". Maybe the state was getting something in return that more than made up for the apparent "loss" that a ban would create? All everyone can think of is "screw the evil telecoms!"
One of the reasons we have semi-autonomous states instead of just one giant federal government ruling everything is that the states can experiment and try things that their neighbors aren't doing. This lets us see what works and what doesn't work. Everyone wants to homogenize all laws across all states, so they go directly to the federal government to Get Things Done, but really this is the wrong approach. If you don't like that your state has banned municipal broadband, and don't like the reasons they chose to do that, take that up with your state. We're all supposedly pro-competition, right? So let's try some different things and let's see which state's approach works.
Except we're talking about bans on municipal broadband. These aren't federal bans, they're state (and maybe county?). These governments probably wouldn't ban municipal broadband unless they felt it would be bad for their community, or if the telco agreed to do something for the community in exchange for the ban.
This is certainly a good and interesting point, but it seems somewhat orthogonal to my post. You seem to be going beyond simply prohibiting bans on municipal broadband, and are advocating for something more.
While some federal "standards" on broadband roll-outs would help to push broadband into areas it doesn't exist yet, there are some monumental obstacles that you'd require telcos to overcome here. Your suggestion that 40% of households have the highest tier of service within 5 years of a commercial offering might actually make it impossible for any but the incumbent provider to offer service. Shouldn't we be pushing for more competition here rather than setting high barriers to new deployments? But really, my point here is that every market is going to be different. Every community is going to be different. One Federal Standard might be completely inappropriate for a certain state or county. Why not let that state or county decide the best approach for its own citizens? If one state/county wants to ban municipal broadband, but a neighbor doesn't, it should become pretty clear pretty quickly which approach yields the best service and price for the customers.
By all means, the federal government should be there to make things easier to do across state lines. But that should never mean that the federal government should just trump existing, properly enacted, properly negotiated state law without a damn good reason. Those states and municipalities have reasons for the laws they enact, and even if you don't agree with them, it's a bit arrogant to tell them they're all wrong and they should be running their affairs differently. Maybe their legislators are all corrupt and in the pockets of the telecoms, I don't know, but even if that's the case, that's their problem to deal with, not the federal government's.
And that may be a perfectly valid thing to be upset about. But if no penalties were prescribed in the law, that was a kind of stupid thing for the government to do, yes? If you were going to sell someone your house, would you just give it to them and assume they'd pay you, or would you make payment a condition of the sale?
If a business has to pay some new tax that wasn't there before, that's a cost of doing business. Somehow, that cost is going to be passed onto the consumer. Usually you have to raise prices, or drop out of the market if you can't be competitive. Here, they got permission to make it its own separate line item. You'll pay for it one way or the other, but by making it a separate fee, they don't have to *appear* to raise their prices and can just be done with the whole mess.
If your local government entered into a contract with your local telecom to develop services like higher-speed broadband for your citizens, with tax dollars, and your telecom failed to do that, surely the contract had penalties spelled out for that eventuality? If those penalties were insufficient or nonexistent, that's your community's fault.
Something else to consider is developing municipal broadband infrastructure with tax dollars, but contracting out the maintenance and management of that to a telecom. That way they're not really "in competition" and the contract could be equally profitable for both sides.
You could even hire the local telecom to do the actual build-out, with the understanding that it would be owned by the municipality. If the telecom can't get favorable terms, they don't have to enter into the contract to do the work.
If you're worried that the telecom is going to do shitty work or sabotage the system, there are things like "inspections" and "licenses" that work well in other industries. Have an independent party verify that things were built properly, have network plans be approved by the municipality before they're implemented, etc.
I say let municipalities and states alike have the flexibility to do whatever it is that they want in this regard. The federal government has no business sticking its nose here. If a state wants to enter into some sort of agreement with a telecom headquartered there, and that agreement means that municipalities shouldn't be allowed to do their own thing, but in return, the state gets discounted broadband, universal coverage, and more jobs, that might be a perfectly reasonable thing for that state to do. If its citizens don't like it, they need to make themselves heard. But even so, it might actually be a good agreement for the state. You don't know that. We need to try to look past our anti-telecom bias here and not automatically assume that everything they do is evil.
Usually people on Slashdot are all about democracy, but they fail to realize that a strong federal government, from the state's perspective, is the anti-democracy. It's overriding what are effectively the desires of a sovereign state. If there was a valid reason for this ban on bans to be applied at the federal level, that might be OK, but just because you can justify the law under the Commerce Clause doesn't mean it's the right thing to do at the federal level.
I hate the Commerce Clause. There should be an additional requirement that you should be able to justify why a bill should exist at the Federal layer with a straight face.
If a state really prides itself on the telecom companies that live within it, and maybe those telecom companies would give special discounts if they were the ones providing service to a municipality, it might be perfectly reasonable for that state to enact legislation that is perceived as a "ban" here. If it benefits their local telecom, then it benefits their local community, because lots of people there work for that telecom.
Of course, this makes no sense for other states that may not have those interests. But why prohibit it for everyone? Why do these "bans" exist in the first place? Because of corrupt politicians in the pockets of the telecoms? Or could there perhaps be more to these pieces of legislation, that, when understood, make the "bans" seem a little more reasonable? But I guess meddling in the affairs of lower governments, under the guise of doing something that, on its face, seems like a good thing, is all we care about? "Yay, they're really Getting Things Done in Congress! I'm going to vote to re-elect these guys!"
So you admit that it's a mistake, and in the same breath effectively say that it's OK to keep it. You're advocating that one person take advantage of another. I disagree. Would it be criminal for you to keep the money? Probably not. But you don't have to commit a crime for the court to require that you return it through a civil suit.
Multiply this mistake by 10 or 100. The party taken advantage of takes you to court, and you admit that it was an error. Who do you think the court is going to side with?
This isn't something that could be confused for a tip or interest. This isn't even cash given to you by an agent of the company that would be authorized to give you "bonus cash". This is a machine that you know is only supposed to give you what you asked for. It shocks me to read that there are people that actually think it's fair to keep that money.
Not all banks are evil megacorps. Not all ATM companies are evil megacorps. I could go out and buy an ATM for a few thousand dollars, put it somewhere useful, and make money off of the fees. For all you know this is one of those ATMs operated by some grandmother somewhere as an investment.
Yes, he's saying "two wrongs make a right". As in, "I think this behavior is evil. I think some corporations exhibit this behavior. Therefore, all corporations are evil and deserve to be taken advantage of, [regardless of what the law says]." I was attacking the premise. I didn't specifically address the "two wrongs make a right" fallacy because it's obvious.
In other words, yes, you believe corporations are evil and that people should be allowed to take advantage of them. Your rationale behind this belief is, of course, dubious, since individuals can be assholes too:
"Oh, your car broke down? So effing what? I have to get home and watch Simpsons, so I won't stop for you."
"Oh, you have extra bills to pay? Sorry about your luck. I'm going to leave without telling you that I dented your car door."
"Oh, you say you need at least $11/hour to live? Tough crap. I'm going to shoplift from your little store anyway."
"Oh, you're already just skimming your bills? Tough crap. We want this big construction project in our state, and it's too expensive to be worth it, so we want our representative to push for the federal government to pay for it instead, spreading the cost out over everyone else."
The world is full of people taking advantage of each other. It's easy to point the finger at Big Business, but it pays to remember that Big Business is run by people. We get through life in a world full of assholes by having laws that protect us from those that take things too far. The law doesn't care whether or not you like someone. It cares that you're taking something that does not belong to you, and refusing to give it back.
...among those that are interested in this topic. You and I know who they are. But ask anyone on the street who the RIAA is, and they'll either shrug their shoulders or say, "They're the guys that sue you when you steal songs off the Internet!"
Which is exactly what the RIAA wants to hear. Details like who the RIAA represents, the fact that mere downloaders aren't sued, only those redistributing, and that infringement isn't theft, are lost on the general public. The only people changing their habits are those that have bought what the RIAA wanted them to buy, or those that have researched the issue and are avoiding giving RIAA members their business. But they're all in a very tiny minority, I suspect. Had every RIAA headline over the last few years said "Sony" instead of "RIAA", or, even better, a specific band, I rather suspect they would feel the backlash.
Who isn't the RIAA's customer? It's not like this is a business that is trying to avoid losing customers to a competitor. The general public has no clue what the RIAA does, who they represent, or who its "competitors" are.
If the individual music labels were the ones in the news, that might be an entirely different matter. The beauty of doing all of this through the RIAA is that none of the RIAA's members are in the light. Yes, some of us know who they are, but when someone reads something in the news about the RIAA suing people downloading music, they're not thinking about the labels behind it. (And even if they do, how many are going to keep up their little protest when So-And-So's latest album just came out, and it's REALLY good, like omg!)
I realize I'm nitpicking here, but these are civil cases where there is no guilt or innocence. You're only liable or not liable for damages. This is also why it seems odd that they'd be pursuing cases against people that are only indirectly involved. Only extreme cases of copyright infringement are crimes, and it wouldn't be the RIAA going after you, it would be your own government.
The RIAA's tactics are no different from any other Big Business or malcontent exploiting America's legal system. The fact that you can sue anyone for any reason, and literally extort settlements because it's cheaper to pay a few thousand dollars than defend even utter nonsense at a trial is heartbreaking. Larger businesses, taking every precaution to avoid being sued (look at all of the warning labels!) nevertheless have to set aside a sizable portion of their budget just for these nuisance settlements. It's become just another cost of doing business in the US. And we all pay for it in the end, because these costs will influence whether or not these businesses can price their products competitively. And if they can't, that's less competition, which means fewer jobs and higher prices from those that are left.