Internet service is a free market with lots of competition -- at least right now. There are currently more than 4,000 independent ISPs in the United States -- that's 80 per state, on average.
Not that this free market could not be destroyed. If government pursues policies that kill off all but the cable and telephone companies, or does not act aggressively to stop anticompetitive behavior, there will no longer be a free market. The FCC and the courts, by failing to enforce the provisions of the Telecommunications Act of 1996 and by allowing the telephone companies especially to cut off competitors, have done a great deal of harm to competition, but it's not by any means dead yet.
Regulation that cripples independent ISPs by destroying their quality of service, raising their costs and prices, and preventing them from innovating would kill them. And that's exactly what the so-called "neutrality" regulation proposed by the lobbyists -- which isn't actually "neutral" at all -- would do.
As I mentioned in my testimony to the FCC (see http://www.brettglass.com/FCC/remarks.html), the government should stop anticompetitive behavior. If it does, and competition flourishes, no other action is necessary.
You are trolling... and intentionally misrepresenting what I said.
In my letter, I asserted that it is a best practice (and excellent security policy) to ban all activities which are not explicitly permitted. This is the reason why UNIX is secure and Windows isn't.
I also asserted that ISPs should always monitor networks for abuse (which does not mean that they should "inspect" all traffic, but that they should have automatic mechanisms that are triggered by behaviors which violate their Terms of Service or Acceptable Use Policies). They should also be able to stop abuse without blocking legitimate use of the network, so that their actions prevent abusers from denying service to others.
The requirements advocated by the meddling lobbyists and lawyers at "Free" Press amount to a requirement that ISPs meter by the bit... and they admit that.
The reason is simple. Bandwidth costs money. If you don't constrain the amount of bandwidth people use implicitly (by prohibiting bandwidth hogging behavior such as P2P) or explicitly (by throttling), the only way to ensure that a user does not cost the ISP more than he pays per month is to charge by the bit. They try to soft pedal this and instead make bogus claims that Comcast is limiting "free speech," because if consumers realize that Free Press is trying to raise their bills and put them on a meter, they'll rebel. But it's the truth.
The only factual error in his article was a minor, technical one regarding the response to the "Internet meltdown" back in the 80's. (Van Jacobsen's kludge did not prioritize traffic.) But he understands that Comcast wasn't censoring, or trying to (contrary to the lobbyists' bogus claims). He understands that network management is necessary to keep the bandwidth hogs from taking over. He recognizes how bad things can get once the nose of the camel of regulation is allowed into the Internet tent. And while he's a Republican, he shows that he's above the nasty partisanship that has infected Washington, DC by quoting Bill Clinton. Finally, he calls for cooperation rather than constant feuding and head-butting such as have been going on during the whole proceeding. This shows a level of perspective and maturity that we haven't seen from the other Commissioners. Good for him!
Actually, "network neutrality" -- as defined by the inside-the-Beltway lobbyists, such as Free Press (AKA "Save the Internet") -- involves massive regulation of the Internet. It would regulate and constrain ISPs' acceptable use policies and pricing schemes. In some cases, it even amounts to a requirement that ISPs meter service by the bit! The fact is that any definition of "network neutrality" that goes beyond prohibitions on anticompetitive behavior is not neutral at all.
As I explained in my testimony at http://www.brettglass.com/FCC/remarks.html, the agenda that is falsely being promoted as "neutrality" isn't neutral at all. In fact, it prohibits ISPs from enforcing neutrality, letting P2P take over the networks and shift costs from content providers to ISPs. (That's neither neutral nor fair.) It's being done for the benefit of a few corporations -- such as Google, Vuze, and Slingbox -- that support the lobbyists at Free Press, MAP, and the other organizations that ae pushing for regulation. Google, for example, funds Larry Lessig's empire on the Stanford campus, so Lessig (a board member of Free Press) has gotten it to pursue Google's "cause."
What's more, there's no "free speech" issue here, because Comcast was not censoring the Internet. (Comcast is even letting its users right through to the site of Free Press, which is posting outright lies and slander about it.) Note, however, that the government is censoring the Internet in schools, and is proposing to offer censored public Wi-Fi. (The proposal is now before the FCC.)
So, as you can see, the Republicans actually are for true neutrality -- which in this case means not letting the government pick winners and losers by regulating.
Actually, the law -- the Administrative Procedure Act -- says that the FCC must announce in advance that it's going to make rules and go through a formal proceeding before passing them. The agency normally issues a "Notice of Inquiry" and then a "Notice of Proposed Rulemaking" and publishes proposed rules for public comment. This is a fundamental requirement of due process. It can't just go, "Hey, we're suddenly gonna make this policy statement, which we said wasn't binding, into enforceable rules just 'cause we say so." When the FCC tried to do the same thing by fining CBS for the infamous Super Bowl Wardrobe Malfunction, the courts slapped it down.
If the FCC is allowed to be arbitrary and capricious, it will have a devastating effect on investment in broadband.Who would ever want to be an ISP, or invest in broadband deployment, when the industry was regulated by an agency that was a total loose cannon? I am an ISP now, and if such a ruling is allowed to stand I will likely exit the industry.
I am an ISP and I have directly observed the large numbers of connections opened by BitTorrent. In fact, we had to increase table sizes on our SPI firewalls because they were overflowing due to P2P. (And no, Robb, you cannot have the packet dumps because they contain private information about our users.)
You, Robb, have zero experience as an ISP -- as you yourself admit. You are unqualified to comment on the day-to-day practice of network management. You do, however, have experience as a shill for the rich Washington lobbying group Free Press, which is not only a corporation but two corporations. (It's trying to get around Federal laws restricting lobbying by tax exempt organizations by setting itself up that way.) It's took in several million dollars last year, and yet pretends to be a "grassroots" organization. Somehow, I don't think that the 26 contributors who gave it between $10,000 and $300,000 each to that organization were the grass roots! (Free Press, despite its claim to be interested in openness, refuses to reveal the identities of its contributors.... Imagine that.)
The lobbyists have made so many false statements -- to the media, on their Web site, to members of Congress, and directly to the FCC -- that it's hard even to know where to begin! I could spend an hour or two writing a message that goes through just the ones I've seen. But to save time, I'll refer you to a document filed by Comcast which describes and refutes some of the most egregious false statements that they made on the record to the FCC. See http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520034944
That's not a good analogy. There's only one way in which the Internet is like plumbing: the awful, smelly pile of unwashed stuff piling up in the sink. (Gotta clean out my hard drive.)
FCC Chairmen don't like to lose. So, they gather votes before they put items on the agenda. If the Chairman has put the item on the agenda, it's because he has the votes to pass it.
Comcast in fact has claimed that the FCC in fact does not have authority to regulate the Internet. See its filing with the FCC regarding this, and its followup here.
The recent decision in CBS v. FCC (the "wardrobe malfunction" case) may also bear on this decision. The court struck down the FCC's ruling against CBS, saying that the FCC couldn't just make up the rules as it went along! Normally, the FCC promulgates rules by posting a "Notice of Proposed Rulemaking," takes comments, and only then creates rules (which are set out in writing before anyone can be cited for a violation).
But in this case, the FCC published only a vague and explicitly nonbinding "policy statement," on which the public had no chance to comment at all. And it's now trying to say, "Fooled ya! You believed us when we said that it was nonbinding, but we're retroactively turning it into a set of hard and fast rules so that we can take a swipe at Comcast. Why? Because we want to, that's why."
Worse still, that policy statement had several serious problems. For example, it required that Internet users be allowed to run the "application of their choice." While politicians may not graspthe full implications of this, the readers of Slashdot, as computer geeks, know how dangerous this could be. An "application" (a computer program which is not an operating system) encodes and embodies behavior â" any behavior at all that the author wants. And anyone can write one. So, insisting that an ISP allow a user to run any application means that anyone can program his or her computer to behave any way at all â" no matter how destructively â" on the network, and the ISP is not allowed to intervene. In short, such a requirement means that no network provider can have an enforceable Acceptable Use Policy or Terms of Service. Port scanning? The ISP has to allow it, even if it's a prelude to an attack, because it's not illegal. Better turn off all of the intrusion detection systems which detect and block port scans! Exploits? If they haven't been declared to be outright illegal, they are "applications" and so you must not block them. Anyone who engages in destructive behavior, hogs bandwidth, or even takes down the network with an intentionally or unintentionally destructive program could just say, "I was running an application⦠and I have the right to run any application I want, so you canâ(TM)t stop me."
Great work, FCC.
Now, imagine yourself as the administrator of a school network, a public hotspot, an ISP, or any other network which provides service to the public. Someone is doing something disruptive. Your users are complaining; quality of service has deteriorated. But if you act, and especially if you focus on the destructive behavior by detecting the rogue application and attempting to block it and not others (so that legitimate traffic can still get through), you would be subject to FCC fines and penalties.
The above conundrum is but one example of why any proposed rules or regulations pertaining to the Internet should be presented to the public for comment as part of a formal rulemaking process. The FCC is not only regulating what Congress, in its own policy statement, said should not be regulated. (If the FCC makes a policy statement, and Congress makes one, the one made by Congress obviously trumps the FCC's.) The Congressional policy is laid out at 47 USC 230(b), and it says:
It is the policy of the United Statesâ"
(1) to promote the continued development of the Internet and other interactive computer services and other interactive media;
(2) to preserve the vibrant and competitive free market that presently exists for the Internet an
What this result shows is that the FCC, which has driven away all of its best technical people during the past eight years, is now purely a political organization. And because the law requires a 3-2 partisan split among the Commissioners, it means that most of its decisions will be influenced by partisan politics rather than what's best for the people.
If the Chairman and the two other Commissioners of the same party agree on something, it sails right through. (This is what happened with travesties such as the Sirius-XM merger.) However, if the Chairman is motivated to support an agenda to which the other party subscribes, he can expect the two Commissioners of that party to fall into lockstep due to partisanship. That's what happened here. McDowell and Tate, the Republicans, want (as McDowell put it) to "let engineers solve engineering problems." But the Democrats, beseiged by the left-leaning Democratic lobbyists of Free Press, voted to regulate the Internet both because of the Democrats' inherent desire to regulate and because they swallowed the falsehoods of their fellow partisans at Free Press uncritically. So, if the Chairman was willing to support the same result, it would happen.
The question, of course, is why Martin -- a Republican -- would be pro-regulation. I do not know Kevin Martin, but several theories have been floated on various blogs. The first is that the Chairman was feeling pressure from Congress. (He was on the hot seat less than a month ago before a Congressional subcommittee which strongly suggested that if he did not regulate, they'd take matters out of his hands.) The second is that he is "anti-cable," and -- regardless of what harm he might do to the Internet -- wanted to take a swipe at Comcast. (Some bloggers have speculated that Martin is bucking for a job as a telephone company executive or board member when he retires from the Commission, and so is giving those companies the quid pro quo for obtaining such a post. I certainly hope that this is not the case, but then, I do not know him.)
Many people have also noted that the slates of panelists at the two hearings on network management were stacked against Comcast. In Boston, the ratio was about 2:1; at Stanford, it was 6:1. Since the Chairman picks the panelists (the other Commissioners can offer advice, but he need not take it), the fact that even the first hearing was heavily stacked against Comcast suggests that the Chairman or his staff may have had a predisposition to rule against Comcast from the start.
In any event, the fact that only one witness at either hearing was actually engaged in business as an ISP strongly suggests that politics, not engineering facts, would rule the day. And they did. The lobbyists and lawyers of Free Press, an inside-the-Beltway lobbying group which spent more than $700,000 on various Internet agendas in 2007 alone, repeated statements which were simply technically false again and again until the Commissioners believed them. And little guys like my own independent ISP? We got 8 -- count them -- 8 -- minutes to talk. This is not promising for the future of the Internet. If it's dominated by politics, and especially by an agency which has lost its technical compass and rules on the basis of politics and partisanship -- the Internet is in trouble.
Funny how, just because a few self-interested Washington lobbyists have declared it so, DPI is the new Ultimate Evil. Exactly how will we block spam if we don't inspect messages traversing the Net?
It's a sign of this DPI hysteria that this article blames DPI for redirection of domain name lookup failures. The fact is, DPI is not necessary to replace NXDOMAIN answers to DNS queries with pointers to a specific server. All one needs is to do some very simple hacking of the recursive resolver. Which is easy if you are the administrator who is running it.
The Internet is "always on." P2P can cause just as much distress to users when it clogs the pipes at 4 AM than when it does so during the day. And, yes, latency slows down P2P to the point where it's faster to use FTP. But of course, that would mean that the content provider would have to pay for its bandwidth and not try to fob the cost off on the ISP. That is, if the content provider is legal. The vast majority of BitTorrent traffic is still pirated media.
No; upload capacity at the edges is not underutilized at all. In fact, both cable modem and DSL are designed so that, of the total available bandwidth on the medium, more is devoted to downstream traffic than to upstream traffic so that users receive data most quickly (which is what they want). Also, this capacity is not "already paid for." It was an investment that is still being paid back.
As for latency: yes, it is a big issue for P2P, because P2P consists of many small transactions which are latency-sensitive. (A BitTorrent node, for example, must communicate with many other machines and exchange many packets with a "tracker.") A long FTP download (which makes much more sense and also has much, much less overhead) is far less latency-sensitive.
As usual, we see in this case that P2P is a solution looking for a problem. In fact, it is really only good at solving one problem: the one it was invented to solve. That is, making it difficult to stop people from pirating music, software, and media.
Internet bandwidth is most expensive at the edges, and latency to other users is the longest. It's the worst place from which to serve up data; you want to do that from servers in the "middle" of the Net.
What's more, the most scarce and valuable resource of the Internet is bandwidth near the edges. Put the servers out there, and you'll raise the cost of broadband deployment and exhaust the resources that are already there.
Anyone can buy space on a fast, cheap server at a server farm for far less than it costs to serve data from the edge. So, why don't the people who are running this project just do that? There's only one possible reason: they want to get users and ISPs to give them these resources for free. Which just doesn't wash. If use of these devices became widespread it would either drive up the cost of broadband tremendously or be banned from networks outright by businesses and ISPs. And deservedly so. It's a bad idea.
The world lost all respect for Colin Powell when he knowingly and blatantly lied to the entire world. A pawn of Bush and Cheney, Powell presented to the UN false "evidence" that Saddam Hussein was producing and stockpiling chemical weapons. If either candidate was foolish enough to pick him as VP, it would indicate that he didn't think that truthfulness mattered.
There seem to be a lot of rather heated and passionate remarks condemning what Dr. Roberts is doing. While it's true that P2P was invented for the purpose of pirating intellectual property and is still overwhelmingly used for that, the appliances sold by Dr. Roberts' company don't have any way to check to see whether the content being exchanged is pirated or not. In fact, there's no way -- a priori -- to tell whether a transfer is legal or not. (One can determine this by downloading the same content oneself and looking at what it is, or make a good guess that it's not if it comes from a site whose purpose is piracy -- e.g. The Pirates' Bay. But in many cases, one can't tell for sure, and ISPs don't want to be in the position of trying to figure this out.)
In any case, what these appliances address -- and this is legitimate -- is not the legality of the material that's being exchanged but rather a machine's behavior on the Internet. Is it hogging bandwidth? Is it attempting to exploit vulnerabilities in Internet protocols to seize priority over other traffic? (This is what BitTorrent does; it exploits the fact that (a) there's no explicit congestion notification on the Internet and (b) the "fairness" in TCP is all on a "per stream" basis, so if one starts a very large number of streams one will get priority over another user with fewer streams.)
Another very serious problem for ISPs is that P2P is increasingly being used by content providers to shift the cost of Internet bandwidth from the content providers (who are already profiting handsomely) to ISPs (many of which already operate on razor thin margins as a result of price squeezing by monopoly telephone companies). Instead of running their own servers, these content providers (Vuze and Blizzard are two examples) turn their users' machines into servers, often without their knowledge and certainly without the ISP's consent. Since virtually all Internet users in the US have "flat rate" service, the cost of the bandwidth required to operate the server is shifted to the ISP. And since bandwidth at the edge of the network -- at the user's end of an ISP's home or business connection -- is far more expensive than bandwidth in an Internet co-lo, the costs are not only shifted; they're multiplied by a factor of 100 or more. Again, many independent ISPs in this country are barely getting by, and people often complain that there isn't enough competition; they simply must throttle or block P2P or there will be no choice at all.
Americans really have two choices: let their ISPs engage in bandwidth management (which is beneficial; it also stops bandwidth hogs from impacting your quality of service) or be faced with a monopoly or duopoly. See my recent testimony before the FCC at http://www.brettglass.com/FCC/remarks.html
for more on this issue. It's important. If you opt for the "freedom" to do infinite P2P (and in this case, it's as in beer rather than in speech, because P2P is mostly used to get free movies and music and virtually never used for political speech), you'll lose your freedom to choose an Internet provider. And then the cable company and the phone company will have their way with you, which you will probably not like.
If the router is intended to run open source, then of course they would release it. Also, it is not true that manufacturers are required to release all the source code that runs on a Linux-based router. They can have as many custom applications, provided without source, as they wish.
I'd really like to have a turnkey, commercially built router with the security of OpenBSD, NetBSD, or FreeBSD. Given the business-unfriendliness of the latest version of the GPL, why aren't companies like Netgear moving to BSD? Or is it just a matter of time before they do?
The majority of ISPs are not part of any vertical monopoly. There are more than 4,000 independent ISPs in this country, and they far outnumber the telephone and cable companies.
As I mentioned in my reply above, that mailing list was formed by, and is populated by, the sort of people who despise telecommunications service providers of all kinds. They can't, or don't want to, distinguish between our small, local, consumer-friendly ISP and the corporate behemoths. In fact, it's highly inconvenient for them that our small, local ISP -- like our 4,000 to 8,000 colleagues, depending upon how you count -- isn't a big, greedy monopoly like the telephone and cable companies! But while they seem to wish that "little good guys" like us would just go away (because we blow a hole in their argument that all ISPs are evil and therefore must be regulated), we are not going away at all. In fact, we're gaining speed and strength, because rational people understand that the best way to fight back against uncaring monopolies is to patronize their smaller, more consumer-friendly competitors.
By "telling it like it is," our company is not "carrying water" for any large corporation. Rather, we're looking out for our users, who -- if we did not exist -- would be left at the mercy of a duopoly. Again, see my comments to the FCC, at http://www.brettglass.com/remarks.html,
for more. That page also contains links to other documents which may be of interest to those who would like to get past the propaganda to the truth.
Slashdot Mirror
Not that this free market could not be destroyed. If government pursues policies that kill off all but the cable and telephone companies, or does not act aggressively to stop anticompetitive behavior, there will no longer be a free market. The FCC and the courts, by failing to enforce the provisions of the Telecommunications Act of 1996 and by allowing the telephone companies especially to cut off competitors, have done a great deal of harm to competition, but it's not by any means dead yet.
Regulation that cripples independent ISPs by destroying their quality of service, raising their costs and prices, and preventing them from innovating would kill them. And that's exactly what the so-called "neutrality" regulation proposed by the lobbyists -- which isn't actually "neutral" at all -- would do.
As I mentioned in my testimony to the FCC (see http://www.brettglass.com/FCC/remarks.html), the government should stop anticompetitive behavior. If it does, and competition flourishes, no other action is necessary.
You are trolling... and intentionally misrepresenting what I said.
In my letter, I asserted that it is a best practice (and excellent security policy) to ban all activities which are not explicitly permitted. This is the reason why UNIX is secure and Windows isn't.
I also asserted that ISPs should always monitor networks for abuse (which does not mean that they should "inspect" all traffic, but that they should have automatic mechanisms that are triggered by behaviors which violate their Terms of Service or Acceptable Use Policies). They should also be able to stop abuse without blocking legitimate use of the network, so that their actions prevent abusers from denying service to others.
The requirements advocated by the meddling lobbyists and lawyers at "Free" Press amount to a requirement that ISPs meter by the bit... and they admit that.
The reason is simple. Bandwidth costs money. If you don't constrain the amount of bandwidth people use implicitly (by prohibiting bandwidth hogging behavior such as P2P) or explicitly (by throttling), the only way to ensure that a user does not cost the ISP more than he pays per month is to charge by the bit. They try to soft pedal this and instead make bogus claims that Comcast is limiting "free speech," because if consumers realize that Free Press is trying to raise their bills and put them on a meter, they'll rebel. But it's the truth.
See the section on this, near the end of my testimony before the FCC, at http://www.brettglass.com/FCC/remarks.html.
The only factual error in his article was a minor, technical one regarding the response to the "Internet meltdown" back in the 80's. (Van Jacobsen's kludge did not prioritize traffic.) But he understands that Comcast wasn't censoring, or trying to (contrary to the lobbyists' bogus claims). He understands that network management is necessary to keep the bandwidth hogs from taking over. He recognizes how bad things can get once the nose of the camel of regulation is allowed into the Internet tent. And while he's a Republican, he shows that he's above the nasty partisanship that has infected Washington, DC by quoting Bill Clinton. Finally, he calls for cooperation rather than constant feuding and head-butting such as have been going on during the whole proceeding. This shows a level of perspective and maturity that we haven't seen from the other Commissioners. Good for him!
Actually, "network neutrality" -- as defined by the inside-the-Beltway lobbyists, such as Free Press (AKA "Save the Internet") -- involves massive regulation of the Internet. It would regulate and constrain ISPs' acceptable use policies and pricing schemes. In some cases, it even amounts to a requirement that ISPs meter service by the bit! The fact is that any definition of "network neutrality" that goes beyond prohibitions on anticompetitive behavior is not neutral at all.
Not only am I bringing facts; I'm bringing ones that you and your organization are trying to repress.
As I explained in my testimony at http://www.brettglass.com/FCC/remarks.html, the agenda that is falsely being promoted as "neutrality" isn't neutral at all. In fact, it prohibits ISPs from enforcing neutrality, letting P2P take over the networks and shift costs from content providers to ISPs. (That's neither neutral nor fair.) It's being done for the benefit of a few corporations -- such as Google, Vuze, and Slingbox -- that support the lobbyists at Free Press, MAP, and the other organizations that ae pushing for regulation. Google, for example, funds Larry Lessig's empire on the Stanford campus, so Lessig (a board member of Free Press) has gotten it to pursue Google's "cause."
What's more, there's no "free speech" issue here, because Comcast was not censoring the Internet. (Comcast is even letting its users right through to the site of Free Press, which is posting outright lies and slander about it.) Note, however, that the government is censoring the Internet in schools, and is proposing to offer censored public Wi-Fi. (The proposal is now before the FCC.)
So, as you can see, the Republicans actually are for true neutrality -- which in this case means not letting the government pick winners and losers by regulating.
Actually, the law -- the Administrative Procedure Act -- says that the FCC must announce in advance that it's going to make rules and go through a formal proceeding before passing them. The agency normally issues a "Notice of Inquiry" and then a "Notice of Proposed Rulemaking" and publishes proposed rules for public comment. This is a fundamental requirement of due process. It can't just go, "Hey, we're suddenly gonna make this policy statement, which we said wasn't binding, into enforceable rules just 'cause we say so." When the FCC tried to do the same thing by fining CBS for the infamous Super Bowl Wardrobe Malfunction, the courts slapped it down.
If the FCC is allowed to be arbitrary and capricious, it will have a devastating effect on investment in broadband.Who would ever want to be an ISP, or invest in broadband deployment, when the industry was regulated by an agency that was a total loose cannon? I am an ISP now, and if such a ruling is allowed to stand I will likely exit the industry.
I am an ISP and I have directly observed the large numbers of connections opened by BitTorrent. In fact, we had to increase table sizes on our SPI firewalls because they were overflowing due to P2P. (And no, Robb, you cannot have the packet dumps because they contain private information about our users.)
You, Robb, have zero experience as an ISP -- as you yourself admit. You are unqualified to comment on the day-to-day practice of network management. You do, however, have experience as a shill for the rich Washington lobbying group Free Press, which is not only a corporation but two corporations. (It's trying to get around Federal laws restricting lobbying by tax exempt organizations by setting itself up that way.) It's took in several million dollars last year, and yet pretends to be a "grassroots" organization. Somehow, I don't think that the 26 contributors who gave it between $10,000 and $300,000 each to that organization were the grass roots! (Free Press, despite its claim to be interested in openness, refuses to reveal the identities of its contributors.... Imagine that.)
The lobbyists just won't let it go, will they. The link in the posting just above points to yet more false statements by Robb Topolski, a Free Press lobbyist. See http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520035605 for a complete refutation.
The lobbyists have made so many false statements -- to the media, on their Web site, to members of Congress, and directly to the FCC -- that it's hard even to know where to begin! I could spend an hour or two writing a message that goes through just the ones I've seen. But to save time, I'll refer you to a document filed by Comcast which describes and refutes some of the most egregious false statements that they made on the record to the FCC. See http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520034944
That's not a good analogy. There's only one way in which the Internet is like plumbing: the awful, smelly pile of unwashed stuff piling up in the sink. (Gotta clean out my hard drive.)
FCC Chairmen don't like to lose. So, they gather votes before they put items on the agenda. If the Chairman has put the item on the agenda, it's because he has the votes to pass it.
Comcast in fact has claimed that the FCC in fact does not have authority to regulate the Internet. See its filing with the FCC regarding this, and its followup here.
The recent decision in CBS v. FCC (the "wardrobe malfunction" case) may also bear on this decision. The court struck down the FCC's ruling against CBS, saying that the FCC couldn't just make up the rules as it went along! Normally, the FCC promulgates rules by posting a "Notice of Proposed Rulemaking," takes comments, and only then creates rules (which are set out in writing before anyone can be cited for a violation).
But in this case, the FCC published only a vague and explicitly nonbinding "policy statement," on which the public had no chance to comment at all. And it's now trying to say, "Fooled ya! You believed us when we said that it was nonbinding, but we're retroactively turning it into a set of hard and fast rules so that we can take a swipe at Comcast. Why? Because we want to, that's why."
Worse still, that policy statement had several serious problems. For example, it required that Internet users be allowed to run the "application of their choice." While politicians may not graspthe full implications of this, the readers of Slashdot, as computer geeks, know how dangerous this could be. An "application" (a computer program which is not an operating system) encodes and embodies behavior â" any behavior at all that the author wants. And anyone can write one. So, insisting that an ISP allow a user to run any application means that anyone can program his or her computer to behave any way at all â" no matter how destructively â" on the network, and the ISP is not allowed to intervene. In short, such a requirement means that no network provider can have an enforceable Acceptable Use Policy or Terms of Service. Port scanning? The ISP has to allow it, even if it's a prelude to an attack, because it's not illegal. Better turn off all of the intrusion detection systems which detect and block port scans! Exploits? If they haven't been declared to be outright illegal, they are "applications" and so you must not block them. Anyone who engages in destructive behavior, hogs bandwidth, or even takes down the network with an intentionally or unintentionally destructive program could just say, "I was running an application⦠and I have the right to run any application I want, so you canâ(TM)t stop me."
Great work, FCC.
Now, imagine yourself as the administrator of a school network, a public hotspot, an ISP, or any other network which provides service to the public. Someone is doing something disruptive. Your users are complaining; quality of service has deteriorated. But if you act, and especially if you focus on the destructive behavior by detecting the rogue application and attempting to block it and not others (so that legitimate traffic can still get through), you would be subject to FCC fines and penalties.
The above conundrum is but one example of why any proposed rules or regulations pertaining to the Internet should be presented to the public for comment as part of a formal rulemaking process. The FCC is not only regulating what Congress, in its own policy statement, said should not be regulated. (If the FCC makes a policy statement, and Congress makes one, the one made by Congress obviously trumps the FCC's.) The Congressional policy is laid out at 47 USC 230(b), and it says:
It is the policy of the United Statesâ"
(1) to promote the continued development of the Internet and other interactive computer services and other interactive media;
(2) to preserve the vibrant and competitive free market that presently exists for the Internet an
What this result shows is that the FCC, which has driven away all of its best technical people during the past eight years, is now purely a political organization. And because the law requires a 3-2 partisan split among the Commissioners, it means that most of its decisions will be influenced by partisan politics rather than what's best for the people.
If the Chairman and the two other Commissioners of the same party agree on something, it sails right through. (This is what happened with travesties such as the Sirius-XM merger.) However, if the Chairman is motivated to support an agenda to which the other party subscribes, he can expect the two Commissioners of that party to fall into lockstep due to partisanship. That's what happened here. McDowell and Tate, the Republicans, want (as McDowell put it) to "let engineers solve engineering problems." But the Democrats, beseiged by the left-leaning Democratic lobbyists of Free Press, voted to regulate the Internet both because of the Democrats' inherent desire to regulate and because they swallowed the falsehoods of their fellow partisans at Free Press uncritically. So, if the Chairman was willing to support the same result, it would happen.
The question, of course, is why Martin -- a Republican -- would be pro-regulation. I do not know Kevin Martin, but several theories have been floated on various blogs. The first is that the Chairman was feeling pressure from Congress. (He was on the hot seat less than a month ago before a Congressional subcommittee which strongly suggested that if he did not regulate, they'd take matters out of his hands.) The second is that he is "anti-cable," and -- regardless of what harm he might do to the Internet -- wanted to take a swipe at Comcast. (Some bloggers have speculated that Martin is bucking for a job as a telephone company executive or board member when he retires from the Commission, and so is giving those companies the quid pro quo for obtaining such a post. I certainly hope that this is not the case, but then, I do not know him.)
Many people have also noted that the slates of panelists at the two hearings on network management were stacked against Comcast. In Boston, the ratio was about 2:1; at Stanford, it was 6:1. Since the Chairman picks the panelists (the other Commissioners can offer advice, but he need not take it), the fact that even the first hearing was heavily stacked against Comcast suggests that the Chairman or his staff may have had a predisposition to rule against Comcast from the start.
In any event, the fact that only one witness at either hearing was actually engaged in business as an ISP strongly suggests that politics, not engineering facts, would rule the day. And they did. The lobbyists and lawyers of Free Press, an inside-the-Beltway lobbying group which spent more than $700,000 on various Internet agendas in 2007 alone, repeated statements which were simply technically false again and again until the Commissioners believed them. And little guys like my own independent ISP? We got 8 -- count them -- 8 -- minutes to talk. This is not promising for the future of the Internet. If it's dominated by politics, and especially by an agency which has lost its technical compass and rules on the basis of politics and partisanship -- the Internet is in trouble.
It's a sign of this DPI hysteria that this article blames DPI for redirection of domain name lookup failures. The fact is, DPI is not necessary to replace NXDOMAIN answers to DNS queries with pointers to a specific server. All one needs is to do some very simple hacking of the recursive resolver. Which is easy if you are the administrator who is running it.
The Internet is "always on." P2P can cause just as much distress to users when it clogs the pipes at 4 AM than when it does so during the day. And, yes, latency slows down P2P to the point where it's faster to use FTP. But of course, that would mean that the content provider would have to pay for its bandwidth and not try to fob the cost off on the ISP. That is, if the content provider is legal. The vast majority of BitTorrent traffic is still pirated media.
No; upload capacity at the edges is not underutilized at all. In fact, both cable modem and DSL are designed so that, of the total available bandwidth on the medium, more is devoted to downstream traffic than to upstream traffic so that users receive data most quickly (which is what they want). Also, this capacity is not "already paid for." It was an investment that is still being paid back. As for latency: yes, it is a big issue for P2P, because P2P consists of many small transactions which are latency-sensitive. (A BitTorrent node, for example, must communicate with many other machines and exchange many packets with a "tracker.") A long FTP download (which makes much more sense and also has much, much less overhead) is far less latency-sensitive. As usual, we see in this case that P2P is a solution looking for a problem. In fact, it is really only good at solving one problem: the one it was invented to solve. That is, making it difficult to stop people from pirating music, software, and media.
Internet bandwidth is most expensive at the edges, and latency to other users is the longest. It's the worst place from which to serve up data; you want to do that from servers in the "middle" of the Net. What's more, the most scarce and valuable resource of the Internet is bandwidth near the edges. Put the servers out there, and you'll raise the cost of broadband deployment and exhaust the resources that are already there. Anyone can buy space on a fast, cheap server at a server farm for far less than it costs to serve data from the edge. So, why don't the people who are running this project just do that? There's only one possible reason: they want to get users and ISPs to give them these resources for free. Which just doesn't wash. If use of these devices became widespread it would either drive up the cost of broadband tremendously or be banned from networks outright by businesses and ISPs. And deservedly so. It's a bad idea.
The world lost all respect for Colin Powell when he knowingly and blatantly lied to the entire world. A pawn of Bush and Cheney, Powell presented to the UN false "evidence" that Saddam Hussein was producing and stockpiling chemical weapons. If either candidate was foolish enough to pick him as VP, it would indicate that he didn't think that truthfulness mattered.
In any case, what these appliances address -- and this is legitimate -- is not the legality of the material that's being exchanged but rather a machine's behavior on the Internet. Is it hogging bandwidth? Is it attempting to exploit vulnerabilities in Internet protocols to seize priority over other traffic? (This is what BitTorrent does; it exploits the fact that (a) there's no explicit congestion notification on the Internet and (b) the "fairness" in TCP is all on a "per stream" basis, so if one starts a very large number of streams one will get priority over another user with fewer streams.)
Another very serious problem for ISPs is that P2P is increasingly being used by content providers to shift the cost of Internet bandwidth from the content providers (who are already profiting handsomely) to ISPs (many of which already operate on razor thin margins as a result of price squeezing by monopoly telephone companies). Instead of running their own servers, these content providers (Vuze and Blizzard are two examples) turn their users' machines into servers, often without their knowledge and certainly without the ISP's consent. Since virtually all Internet users in the US have "flat rate" service, the cost of the bandwidth required to operate the server is shifted to the ISP. And since bandwidth at the edge of the network -- at the user's end of an ISP's home or business connection -- is far more expensive than bandwidth in an Internet co-lo, the costs are not only shifted; they're multiplied by a factor of 100 or more. Again, many independent ISPs in this country are barely getting by, and people often complain that there isn't enough competition; they simply must throttle or block P2P or there will be no choice at all.
Americans really have two choices: let their ISPs engage in bandwidth management (which is beneficial; it also stops bandwidth hogs from impacting your quality of service) or be faced with a monopoly or duopoly. See my recent testimony before the FCC at http://www.brettglass.com/FCC/remarks.html for more on this issue. It's important. If you opt for the "freedom" to do infinite P2P (and in this case, it's as in beer rather than in speech, because P2P is mostly used to get free movies and music and virtually never used for political speech), you'll lose your freedom to choose an Internet provider. And then the cable company and the phone company will have their way with you, which you will probably not like.
If the router is intended to run open source, then of course they would release it. Also, it is not true that manufacturers are required to release all the source code that runs on a Linux-based router. They can have as many custom applications, provided without source, as they wish.
I'd really like to have a turnkey, commercially built router with the security of OpenBSD, NetBSD, or FreeBSD. Given the business-unfriendliness of the latest version of the GPL, why aren't companies like Netgear moving to BSD? Or is it just a matter of time before they do?
The majority of ISPs are not part of any vertical monopoly. There are more than 4,000 independent ISPs in this country, and they far outnumber the telephone and cable companies.
By "telling it like it is," our company is not "carrying water" for any large corporation. Rather, we're looking out for our users, who -- if we did not exist -- would be left at the mercy of a duopoly. Again, see my comments to the FCC, at http://www.brettglass.com/remarks.html, for more. That page also contains links to other documents which may be of interest to those who would like to get past the propaganda to the truth.