channel announcement: (N)eutral (N)etworks

Chairman Powell will receive another letter that he will rather not have received. This one is from Congressman Kind and Boucher, calling on the Chairman to preserve neutrality on the Internet. It’s not quite 700,000 letters (yet, at least), but this campaign does have the support of a large number of interests, including a few large companies (Microsoft, Amazon, and Disney, if it would only have the courage to stand up to the cable companies).

Tim Wu and I filed an Ex Parte with the FCC on this. Stay tuned for more.

This entry was posted in good code. Bookmark the permalink.

31 Responses to channel announcement: (N)eutral (N)etworks

  1. So this campaign is to urge the FCC to enact a rule prohibiting a behavior that nobody has engaged in, and is probably a technical impossibility as well? The FCC already has enough on its plate without pre-emptive rule making.

  2. lessig says:

    (A) Lots have engaged in it, and therefore (B) it is technically possible. Read first; write after understanding.

  3. Fuzzy says:

    This is a campaign to prevent the FCC from enacting rules that would prevent consumers from connecting ‘unapproved’ services or devices to the broadband (DSL,Cable) networks.

    This is definitely necessary if you have been following the FCC, since the topic is already on the FCC plate, as part of FCC proceedings looking into how cable-modem and digital-subscriber-line service provided by phone companies should be regulated.

    For instance on Fri, Apr. 25, 2003, The Mercury News ran an article on this topic when “Leading technology, consumer electronics and entertainment companies … have joined together to lobby the Federal Communications Commission to adopt rules preserving the free-flowing nature of the Internet.”

    Any time I hear that ‘requiring’ an open network would ‘stifle investment, innovation and growth’ from the National Cable and Telecommunications Association, I know that we need the requirement, since what the Internet has proven is that open networks *PROMOTE* – not stifle – investment, innovation
    and growth.

  4. Fuzzy says:

    It sounds like Powell is having a bad day. The Third U.S. Circuit Court of Appeals in Philadelphia granted a stay order that prevented the new FCC media ownership consolidation rules from taking effect. Check the Reuters article on Wired News or Yahoo.

  5. Brian says:

    Comcast is doing it in Boston: I just switched from their cable-modem service to somebody else’s DSL. They had installed several pieces of software on my machine, which it told me were required for the service to work (given that was the spare Windows 486, which was never on except for them, I doubt it): one program when they were MediaOne, one when they were AT&T, and another when they were Comcast. Now that I was off their network, I wanted the software gone. They certainly didn’t provide any easy or automated way of doing this. Their programs failed to cleanly remove themselves using the Add/Remove Programs component of Windows. When I called Comcast technical support, they told me I’d have to format my hard drive and reinstall Windows — and that that might not work the first time, so I might have to do it twice.

    The programs monitor content viewed with the Explorer web browser and edit some of the content as it passes through: they set the default search functions to use their web sites, for example.

    To clarify: Comcast requires a Windows machine, forbids NAT and hardware firewalls (but does not enforce this), and requires use of software which changes the user experience. Then they’re bad about removing it.

  6. The latest Lloyd Dangle Troubletown cartoon plays with the idea of a non-neutral electrical system:

    Its Sept 4 now, not sure how long the strip will be up.

  7. Sorry Brian, but that’s a load. I have Comcast Internet access, and they didn’t install a single piece of software or hardware on my network. The installer came to my house to read the MAC address of a label on my cable modem (my own property, not owned by Comcast). He then made a phone call to get the cable modem’s address installed a database that would allow Comcast’s DHCP server to provide it with an IP address and he was on his way. My cable modem was already connected to a NAT box which was on my home LAN when he got there, and he didn’t touch it or whine in any way. So there is no technical way that Comcast could be “steering” me toward certain sites rather than others unless they were intercepting DNS queries and lying about the results or something equally brain-damaged, and they’re clearly not doing that. So let’s drop the Urban Legends.

    I can understand Comcast wanting to charge extra for people who have home networks attached to their cable modems to the extent that these networks generate lots of traffic or lots of support calls, and I believe it’s within their rights as a business operator to have such policies, just as I believe it’s within their right as a business owner to prevent me from hogging the bandwidth I share with all their other customers. Lessig and Wu’s internally inconsistent and somewhat incoherent letter to the FCC recognizes their legitimate interest in do this, when they say it’s OK for cable guys to do the following: “Prevent Broadband users from interfering with other Broadband or Internet Users� use of their Internet connections, including but not limited to neutral limits on bandwidth usage, limits on mass transmission of unsolicited email, and limits on the distribution of computer viruses, worms, and limits on denial-of-service or other attacks on others;”

    It’s not the FCC’s business to be dictating business practices to cable operators. If Comcast wants to adopt some brain-dead practice like putting their own browser on your machine, they should be free to make that mistake, and they’ll be corrected by the market as they already have. If they want to prevent me from accessing Google (good luck), they’re free to try that and let the market correct them again. It’s OK to make bad business decisions in America; we’re not claiming to be a Utopia, but we do claim to be free, and that means being free to fail.

    What’s most amusing about these “network neutrality” proposals is that they would require the replacement of every cable modem in America if taken literally. Cable modems rely on a protocol called DOCSIS that provides both a connection-oriented isochronous service and a connectionless asynchronous service. Cable companies sell local telephone service that uses the isochronous DOCSIS option, and that service gets higher-priority treatment than the asynchrounouse DOCSIS that carries VoIP.

    This feels an awful lot like the “discrimination” Lessig and Wu want to prohibit, and as I don’t want to throw my cable modem in the trash, I’m against their proposal.

    There’s no use making rules where you haven’t demonstrated that a problem exists, and when you make gratuitous rules you invite unintended consequences that are worse, by definition, than the problem the rules might have resolved.

    The FCC do do well to ignore this “neutrality” malarkey.

  8. Stewart B says:

    I must chime in here after lurking almost everywhere for about 6 months on this topic. It always amuses me that for how much experience you have, Richard, how misguided you are in some things dealing with networks.

    1- just because you don’t have the same experience with Comcast doesn’t mean that Brian is lying.

    2- You should know that network traffic shaping/choking/control has been a technical possibility since the time you ever first heard the word “modem”. Captive network connections and transparent proxies can happen without any knowledge on your part, just like anyone with a wiretap can see your packets anywhere along the line without you knowing it. DNS interference or bandwith throttling…the point is that these things can, and do, happen, based on content requested.

    what’s funny is that you always love to steer arguments about neutrality into unrelated arguments about network-level priorities, and not priorities based on content, and expect no one to notice, or maybe it’s just that you’re hoping that geeks with want to digress to a topic you know more about.

    the fact is, they don’t need “good luck” to prevent you from accessing google; they *can* do it, and whether they succeed or not is not the point at all….their attempt is all that is needed to prove a move towards less neutrality, and when such practices is commonplace and unchecked, then that’s when there is a problem. You seem to think that they can make such attempts, and that’s fine, and the market will ‘correct’ it. I don’t think that you can guarantee it will.

    “they�re clearly not doing that”

    That’s what you say….good for you that you don’t see this experience. But it doesn’t change that there is evidence to the contrary elsewhere.

    “and they�ll be corrected by the market as they already have”

    Are you suggesting that they have attempted to affect neutrality before, and then changed those practices ? What is the market “correction” that you speak of ? Be specific.

    It’s no secret that you will pretty much oppose every opinion that Lessig and Wu have on sheer principle, and not on fact…you have never agreed with any of them, and would never concede if you realized an error on your part. I see you have promised a review of the Future of Ideas in the past (saying that it “won’t be pretty”) and have yet to produce it.

    Don’t get mired in the “it’s not technically feasible to do real-time video over IP” argument, either, because that’s the fallback you have gone to in the past. Related topic, yes, but not the point here.

    The bottom line here is exactly what Prof Lessig said above: 1-it has been done, and 2-it *is* technically possible, no matter what you’re experience-based knee-jerk reactions say.


  9. Stew says:

    p.s. where is there a “definition” about ‘gratuitous rules’ that you talk about ?

  10. In the deep-dark past, At Home installed a custom version of Netscape on some computers that was tailored toward their portal services, just as AOL used to give their customers a funky browser. While the At Home browser didn’t prevent people from accessing Google and Alta Vista, it made it easier to access their portal by setting the browser’s intrinsic search function to it. Netscape had deals with other search engines in the native version that made it easier to access them. As this feature was rejected by the marketplace, along with Netscape’s entire business plan, this doesn’t happen any more. This is what I was talking about when I said the market has already corrected some bad business practices, and it did so without the FCC imposing draconian rules on broadband providers.

    Today, Comcast doesn’t “steer” its customers away from Google or away from their own home pages, although it does try to encourage its customers to use a Comcast home page. For many people, this is desirable as it has a bunch of trouble-shooting stuff that makes for more efficient customer service.

    It’s possible for Comcast to jigger up their DNS in an effort to “steer” customers away from Google, and it’s possible for them to redirect IP addresses. Doing this would involve getting ahold of some highly customized code, and in practice would be more trouble than it’s worth, so from that standpoint I’ll say it’s technically infeasible, if not theoretically impossible. In practice, I’m not bound to Comcast’s DNS even though I’m a Comcast customer, so their only route for “steering” is to mess with IP addresses, and I don’t see them doing this as a practical matter because of all the bother involved and the furor it would create in the market.

    Lessig and Wu argue that the FCC needs to write some new rules for broadband based on some things they think At Home might have done once upon a time which they think Comcast would like to do in the future on the theory that Comcast is a bunch of power-mad control freaks. This isn’t sufficient basis for rules which would, in effect, make every cable modem in the world today obsolete.

    If Comcast wants to try and charge extra for home networks, more power to them, it’s their business. If Comcast wants to block spam, more power to them it’s their business.

    The fact of the matter is that to the extent that cable Internet providers have imposed content restrictions in the past they’ve done so in an attempt to limit certain kinds of traffic that they didn’t have the means to block any other way. I don’t have a problem with them using any means they wish to load balance their networks, because I believe the market will keep them in check, but I have a big problem with people who lack an understanding of network architecture trying to dictate network management practices and using the government as the means to do that. I believe the professors have their hearts in the right place, but their heads in the sand and I see no indication that they’re open to learning about where their understanding of the Internet is off-base.

    To continue citing “End-to-End Arguments in System Design” as the Internet gospel is misleading and naive, and I will object every time they do this.

  11. stewart says:

    “I�m not bound to Comcast�s DNS even though I�m a Comcast customer”

    No, but in the future, you could be. it’s quite possible to have outbound routers allow only the assigned dns servers udp/tcp traffic, and none others. would you argue that it’s not ? Don’t we want to follow the providers into their ‘countries’ and ‘caves’ and ‘strike’ them before they ‘strike’ us ? 🙂

    Citing the well-known ‘e2e’ paper isn’t the only paper that they cite, and they wouldn’t need to cite it to prove that point. The people in the FCC who allowed non-monopoly devices to be attached to the phone network weren’t “network architects”, either, but it doesn’t make their idea less sound. (thank God)

    There are many many technical ways that providers can impose whatever they want, and to the argument that the market will police that….it’s possible, but it’s also very possible that it won’t. That’s enough.

    Take a look at it. It’s asking for neutrality. Not some insane crazy behaviour that is difficult to allow. For the FCC to evaluate and decide on this, it’s very much worth the effort.

  12. …No, but in the future, you could be…

    That’s the problem with this whole exercise: it’s predictated on what might happen in the future, what (some people think) may have happened in the past, and not on what’s happening today and any harm that’s actually being caused.

    Every rule, no matter how well intentioned, has at least some bad side-effects. I’m not willing to pay the price for these side-effects unless there is some actual benefit from the rule, and in this case there isn’t any.

    Lessig has invented a new argument for it, however, and that’s kinda interesting – that this rule is necessary to bring on the next great wave of Internet innovation. I don’t see it, frankly, and I’m not ready to give up my phone service for it either.

  13. Fuzzy says:

    There are no market forces when there is a monopoly. The town in which I live does not have have multiple cable operators and thus I cannot simply switch to a different provider to obtain service if I dislike the cable operator’s license. Saying that market forces will correct a problem in this situation is misguided.

    It is very much the Federal *Communication* Commision’s *duty* to make sure that the cable *communication* providers do not exploit their monopolies. The legal right-of-way provided to cable operators and bandwidth spectrum provided to terrestrial and satellite broadcasters are both part of the public domain and the public and government most definitely have a right to say how those public facilities are exploited.

  14. Cable companies don’t have a monopoly on broadband.

  15. Fuzzy says:

    > Cable companies don�t have a monopoly on broadband.

    You are right. I could have an independent contractor install a fiber optic cable from my home to the nearest Internet traffic exchange facility. No problem if I have a few hundred million dollars. Or I could rent a pair of dark wires and run a T-1 line for a few thousand $$$ a month

    Yes, I exaggerate. But the reality is there are two basic options affordable to consumers at this time. Either the phone company (DSL) or the cable company (digital cable modem). Yes, there are some more esoteric options like satellite and power lines but they are not as competitive or as commonly available.

    Unfortunately, my area is not close enough to the central office (CO) to have the option of DSL. Thus my only option is the cable company. A monopoly.

  16. Tim Wu says:

    If anyone has documented examples of behavior that deviates from network neutrality, Cable DSL or otherwise, please let me know at the email above; I collect them. It is true that some behavior has improved over the last year (like the end of enforced bans on VPNs), but some of the control is just better hidden.

    Also, Mr. Bennett, I was in the telecom industry before law teaching, and I don’t think our technical understanding of the internet is “off base.” For example, your suggestion that DOCSIS connection-oriented voice and a network-neutrality principle are inconsistent is incorrect. The proposed regulation doesn’t seek to ban the use connection-oriented protocols for voice or any other prioritized data; on the contrary, it is designed to promote investments to prioritize different classes of service instead of banning individual applications on an ad-hoc basis, which is damaging to competition in the application market. For cable, it is designed to encourage, but not force the adoption of DOCSIS 1.1 or 2.0, and I can’t imagine you’re opposed to that.

    Tim Wu

  17. Caching web pages is a fundamentally different problem that redirecting web surfers away from the place they tried to go, Fuzz, just as manually overiding an SOA record is a fundamentally different problem from keeping it overridden as the net tries to correct what you’ve done. Understand?

    But be sure and let me know just as soon as Comcast starts hijacking IP addresses and fudging the results of DNS lookups, I want to see that for myself.

    The phone company can put a repeater on your line if they feel like it, or you can go satellite for your broadband, and some day 802.16, so no, the cable company doesn’t have a monopoly on broadband, even for you. But it’s that kind of statement that makes Lessig think he’s on the right track, unfortunately.

    Now the question I’d like to put to you is whether you’re more likely to get some choices for your broadband from a regulatory environment guided by pre-emptive regulations like these that Lessig and Wu want, or from one that’s more free.

  18. Fuzzy says:

    Ahem… I did not say that a web cache operated by redirecting web surfers, I said that many commercial web caches have builtin support for redirection – something you said was “technically infeasible, if not theoretically impossible”.

    The following is a quote directly from the link about CacheFlow I provided above:
    “All CacheFlow products can operate as simultaneous forward, transparent, and reverse proxies. CacheFlow provides detailed, product-specific instructions for transparent redirection using L4 switches from Alteon, ArrowPoint, or Foundry, Cisco routers with WCCP v1 or v2, or Bay routers with packet filters.”

    And as long as the DNS resolver queries for customers are redirected to the Internet service providers DNS servers, a simple SOA change is all that is necessary. The network hardware and software won’t care and won’t try to correct it.

    I am much more likely to get broadband options with a good regulatory environment that encourages consumer confidence and supports comptetition, than from a profit-driven anti-competitive unregulated environment.

    I’ll stick with Professor Lessig, thanks.

  19. Regardless of how you estimate the difficulty of building a bogus DNS and of intercepting IP addresses and spoofing them, the fact remains that cable Internet providers do not, in point of fact, do these things today. I know there are lots of Urban Legends circulating about this stuff, but I have yet to see any credible evidence in support of the threat that Lessig and Wu perceive. Given that nobody engages in these practices, we don’t need to debate their legitimacy or their feasibility, and we certainly don’t need to plead with the FCC to put a stop to them.

    The only thing resembling a content preference on the part of a cable Internet company that actually does exist is the case I mentioned, where the cable telephony that uses the DOCSIS isochronous service has priority over VoIP telephony. I think cable telephony is a legitimate, valuable service, and I don’t want it stopped, as to do that would be to give an advantage to incumbent Telcos.

    You need to ask Lessig why he wants the FCC to enact a rule that would, in effect, take away the only legitimate competition Telcos have in the wireline telephony business. My bet is that he’s not even aware that his pleading would have that effect, which underscores the danger of well-meaning amateurs attempting to impose rules on the FCC.

  20. I missed Prof. Wu’s contribution until just now, so let me just comment on this part: “For cable, it is designed to encourage, but not force the adoption of DOCSIS 1.1 or 2.0, and I can�t imagine you�re opposed to that. “

    Of course I’m not opposed to that, but I don’t see any sense in which this consequence flows from your pleading. In fact, DOCSIS telephony has priority today over VoIP telephony, and a strict application of “content neutrality” would ban that practice.

    Tell me what I’m missing if you think this is wrong.

  21. Tim Wu says:

    Mr. Bennett,

    I recognize there is room for debate over network neutrality; but I think that in your comments you have may have mistaken the difference between content neutrality and application neutrality. Our filing was primarily directed to the problem of discrimination among applications.

    You keep saying that the cable operators have never banned any content. True or not, that’s not the point. The trouble has been with bans on internet applications, like VPNs, or particular uses, like home networking. As I have documented in another paper, such bans were widespread and also enforced until the publicity surrounding the Net Neutral movement this year led to their repeal.

    Second, there is nothing in our proposal (take another look at the specific regulation) designed to prevent broadband operators from giving priority to different forms of traffic. We see that as a good thing — we don’t have some kind of crazy, literalist view of content neutrality that you seem to be attributing to us. The idea is to prevent bans on individual applications, for fear that the broadband operators are externalizing their costs on the application market, instead of investing in something like DOCSIS 1.1. Does that make our position clearer?

    Tim Wu

  22. I don’t share your opinion that “bans” on VPNs and home networking have every been “widespread” in the cable Internet industry, and I certainly don’t believe any such “bans” are widespread today. DOCSIS is widespread, however, and we’re not waiting for its adoption — that’s already happened.

    I understand that some cable operators have attempted to charge per-computer hookup fees, which is what I think you mistake for the kind of home network ban that justifies the strained extension of Hush-a-Phone to Internet regulation, and I understand that some cable operators may have had problems with VPNs in the era when the Clipper chip debate was hot, but that’s not an issue that ever affected the cable networks I’ve used. In any case, these sorts of practices relate to the desire of cable operators to turn a profit while complying with the law, and most of us in America aren’t opposed to capitalism or to the rule of law.

    It appears that you and Prof. Lessig, while getting the mob all worked-up over imaginary encroachments, wish to impose architectural standards upon the cable companies’ Internet access networks which follow your understanding of some sort of “end-to-end” theory of access networks and an alternate form of that theory couched as “application neutrality”. The “end-to-end” architecture simply relates to a simple-minded theory of flow control and error detection/recovery that occurred to a particular group of computer science professors at MIT prior to the rollout of the TCP/IP Internet in 1982. That network was optimized for asynchronous, point-to-point applications that didn’t require jitter-free delivery and multicast addressing, and as more sophisticated applications have come to the fore with real-time requirements, we’ve learned that these design choices were neither neutral nor ideal. The reality is that there is no “neutral network”, either in practice or in principle, so it’s simply silly to demand the FCC or any other government body impose an unattainable goal on Internet access networks or in any other sphere. I’m sure your knowledge of common law extends far enough to realize that it’s wrong for the state to impose a burden on a citizen that he can’t carry.

    The IP layer of the Internet protocol stack today favors asynchronous applications over isochronous ones, even though the undercarriage of both telephony and cable networks is capable of serving the needs of both types of applications. Network engineering in the near future will attempt to correct IP’s shortcomings, and those of us engaged in this work need the freedom to do our jobs without meddling by regulatory bodies or well-meaning bystanders.

    I’d prefer the FCC not get any more involved in the Internet than they already are, and if they must meddle, they shouldn’t do so in the pursuit of quixotic or utopian goals. Cable networks have asymmetrical, point-to-multipoint architectures that certainly favor browsing over such applications as video-conferencing or Napster. Your proposal is in conflict with this architecture, which has been widely criticized and Lessig and his minions. Given the history of this criticism, the agenda is clear, and, in my opinion, counter-productive.

  23. Aaron Swartz says:

    To Lessig and Wu: Your filing is very nice, but I fear your exceptions are big enough to drive an evil truck through.

    For example, during the Blaster and SoBig epidemics, it appeared that Comcast blocked all ICMP PING packets for several days. Would this kind of overbroad restriction be permitted under (a)(3)? Similarly, several cable providers (including block their users from accessing port 25 on any server but their own. Could this absurd regulation be permitted because it allows them to more easily prevent mass unsolicited email?

    (a)(4) is rather unclear — is it intended to only allow removal of literal “delay” and “jitter” (if so, what does that mean?) or to allow the cable provider to ban anything that causes or contributes to it? It’s easy to make a case that video streaming, VoIP, and P2P file sharing all cause “delay”.

    You might also consider promoting the use of fair queuing, instead of less application-neutral mechanisms, for managing bandwidth. For example, some universities have decided to prioritize Web and email traffic over Kazaa as a Quality of Service mechanism (i.e. to speed up Web traffic). Other simply cut off those who are using “too much” bandwidth. I suspect residential Internet providers already or will soon do the same thing. Fair queuing would allow users to make such decisions by giving each user an equal amount of bandwidth to “spend” on whatever applications they feel are important (of course, theoretically reserved but unused bandwidth is also distributed evenly).

    I hope the Commission takes heed of your comments.

  24. stewart says:

    Richard — seriously. Again and again with your confusion of content neutrality with network application neutrality. You have made this argument so many times that it would appear that you’re under the impression that giving a ‘history lesson’ is warranted in every comment you have on the subject will prove your off-topic point.

    “I�d prefer the FCC not get any more involved in the Internet than they already are” — there are some things you enjoy, Mir Bennett, as a consumer and network consultant, regarding the Internet, if it wasn’t for the FCC ‘meddling’ thus far.

    Whether or not your experience has been with situations involving the said ‘abuses’ of neutrality does not mean that they haven’t existed, or still don’t.

    You seem to be saying that this sort of thing:

    1- doesn’t happen at all, because it’s not technically feasible
    2- ummm…if it does, though, then it’ll all get worked out by the ‘market’
    3- er….if it doesn’t, then we shouldn’t try to prevent it from happening in the future, because that’s silly, because it can’t happen, see #1.

    so which is it ? what’s you’re final answer ? you don’t know, because you haven’t been looking for it, and I’m gonna guess you haven’t used every broadband option to investigate…it’s not a monopoly you know, there’s a lot of providers out there….

    “those of us engaged in this work need the freedom to do our jobs without meddling by regulatory bodies or well-meaning bystanders.”

    If you would do more reading, then you’d see that neutrality based on content (not on network layer, not on media, not on standard) should impose no barriers whatsoever. My local pacbell guy doesn’t care who I call on the phone (as long as it’s not illegal to do so), and the same should be with the Internet, period.

    Mr. Wu — I would also love to see what you have thus far in the way of historical evidence of non-neutral network practices. Do you happen to have them for public consumption ?

  25. Stewart, there have been at least two arguments raised by Lessig et. al. against the cable companies in the course of this discussion. The first, in the Mercury News article cited by one “Fuzzy” and also contained in Lessig’s “Future of Ideas”, is the charge that cable operators “steer” web surfers toward content the operators have some interest in having them view. This argument is variously offered as an actual practice or as a potential practice that should raise alarms about cable operators. In relation to this argument, I’ve said that it’s not actually happening and it’s not technically feasible without a lot of effort, so there’s no point in getting worked-up over it in the first place, and no point in legislating against because in the unlikely event that it did happen in the future, the market would correct it.

    The second argument, one that’s more clearly stated in the Lessig/Wu letter to the FCC, is that cable operators are banning certain applications the survival of which is essential to human progress and the advance of civilization on this planet. Examples of these applications are VPNs and home networks. Lessig and Wu go to great lengths to stretch Hush-a-Phone, a court decision against an AT&T ban on rubber cups on handsets, to cover home network restrictions. The analogy doesn’t work, for several reasons.

    In the first place, in so far as some cable operators have policies about home networks, the intent of these policies is to capture revenue for the attachment of multiple computers to their systems. To the extent that the cable operator can tell that a network is attached I don’t see this as a bar to innovation and I don’t buy the notion that Microsoft needs federal regulation expressly preventing cable operators from limiting home networks in order for Bill Gates to enjoy a prosperous future. Home networks that employ insecure Wi-Fi effectively act has an Internet on-ramp for unauthorized users, depriving the cable companies of deserved revenue and violating legitimate customer agreements. When you layer VPNs over warchalking connections you also create an interesting national security scenario that I see no reason to tolerate.

    Cable Internet is a profit-driven business, as Mr. Fuzzy points out. Cable operators need not give the service away to anyone who wants it for free, and they need not tolerate insecure Wi-Fi nets that give it away for free.

    “Application neutrality” is not a concept enshrined in our Constitution, our statutes, or in our common law, and it has no standing in the sphere of Internet regulation. The attempt to find precedent for particular business models and methods of network management in obscure Computer Science papers and RFCs is nothing more than an attempt to force public policy in a direction that it need not take. It’s dishonest, it’s blind to technical realities, and I don’t like it.

    Is that clear enough for you, Stewart?

  26. stewart says:

    Crystal, Richard. Thanks. I was clear on it before. My main argument is with the first point, and I’ll say it again: just because you haven’t seen it happen in your experience, or can’t think of a way at that moment that it could be done, doesn’t mean that it 1-hasn’t been done in the past, and 2-won’t happen in the future. I’ll wait on Mr Wu’s evidence, thanks.

    I *know* you said that it’s not happening. I also don’t think you have tried to look for it. I also don’t know that you have done enough research to make a valid claim that it’s not happening.

    I also don’t agree that it would be ‘corrected’ by the market. It would only be corrected in the market if (off the top of my head)

    1-people were aware that it’s happening
    2-people value having neutrality more than the non-neutral service they are getting
    3-people knowing what neutrality even means.

    and I see no guarantees in any of those, seeing how most Americans eat what they are fed.

    Let me ask you this: is there an FCC rule that says that any phone number I dial has to be free of intentional interference or noise ? Would it be illegal for ATT to inject noise or artifacts if I called Walmart’s 1800 number, versus Target’s ? Or redirect me from one to the other ? Why or why not ? (I’m betting that you won’t comment at all on that)

    What if you found out that they were ? On your phone ? Would you just say…’well, ATT just lost a customer, I’ll now try MCI’ ?

    If you don’t like regulations period, then say so. Don’t start making comments that suggest only network architects know the best way to use networks. It’s just like software engineers thinking that they are UI designers or Human Factors engineers. It’s dumb, and only hurts use in the end.

  27. ryan says:

    1. In Australia a small group of providers have a virtual monopoly. Telstra have an almost literal one.
    2. They do direct traffic. Not directly. But through providing “free” content that isn’t counted in monthly bandwidth tallies and ensuring their own pages are installed as the home page for the browser they install.

    Only the net/computer savy ever notice this and change it. Average home users don’t know that they can change these things or that there are other places to get content.

    This leads to a creeping lock in of users to “desired” content. Why go to another news page when it counts in your bandwidth tally.

    Doesn’t seem bad on the surface, but when those that control the network also end up controlling the content, it will be a very bad situation indeed. One not dissimilar to the current woeful traditional media outlets of print and tv.

  28. Stewart, I’d prefer to confine this discussion to the Lessig/Wu FCC pleading. The “steering” nonsense is certainly interesting, but’s it’s not the point of this thread.

    And yes, I do have a problem with regulation for its own sake. There are an infinite number of “what if” scenarios that might justify regulation, but regulatory and legislative bodies are generally so busy addressing actual problems they don’t have time for merely theoretical ones.

    The question of principle raised by the insistence on “application neutrality” is whether all applications are equally legitimate. I don’t personally accept that as a principle, and offer the examples of illegal applications in support of my position: spam, viruses, and file-stealing violate the law, over and above the harm they do to the network. They’re therefore not legitimate, and therefore there is no “application neutrality” principle embedded in the Internet of today that needs regulatory protection.

    It’s really pretty simple.

  29. stewart says:

    No, I don’t think that the ‘steering’ issue is beside the point at all. It’s very relevant. Btw, Spam and file-stealing are not illegal applications, they are the illegal or annoying use of “legal” and legitimate applications (smtp, and p2p or ftp-based networking).

    “legislative bodies are generally so busy addressing actual problems”

    are you suggesting that since the FCC is really busy, then the public should withold any concerns for future ?

    If you see the footnote #3 in the letter, they recognize that pure neutrality is not the expectation. It’s goal is not Utopia, so you can’t call it such.

    p.s. I knew you’d try to dodge the telephone questions I asked.

  30. Professor Wu himself says the steering issue is irrelevant, Stew: “You keep saying that the cable operators have never banned any content. True or not, that�s not the point. The trouble has been with bans on internet applications, like VPNs, or particular uses, like home networking.”

    So the topic under discussion is: “Application Neutrality: Friend, Foe, or Figment”.

    Dodging right along,


  31. Craig says:


    The offices of Kind and Boucher are refusing to release the official, final copy of the letter because, as one staffer told me, they “don’t want the cable companies calling other Congressional signatories of the letter….”

    The FCC claims they haven’t received the letter yet, but the Congressional staffer said it was sent. I need the official version with all Congressional names. Any ideas? Does anyone have access to this?



Comments are closed.