Let's put Net neutrality to bed: Engaging the George

Let's put Net neutrality to bed: Engaging the George

Summary: One last time, with polite feeling. Net Neutrality argued rationally.

SHARE:
TOPICS: Browser
18

George Ou, another blogger here at ZD Net, persists in insisting that the Net neutrality position is deeply flawed. I'm responding to his comments on another post in this new posting because we have, at last, come to the nut of the dispute. The infrastructure works because it is shared. The spirit of the Net neutrality position is that sharing requires equal treatment. I intended to put this to bed days ago, but back comes George, like certain Star Trek space aliens who think one thought and never deviate from it.

I am comfortable being in the same camp on this issue as Tim Berners-Lee, the architect of the World Wide Web, and Vint Cerf, one of the network scientists who initially developed the TCP/IP-based Internet. These are not people whose credentials George would dare question. I know, though, that you, George, will be back, attempting to assimilate me, again.

George refuses to acknowledge that the text of amendments submitted in the House and Senate (click the links to read the texts) are tempered by existing law. There are "Preserved Rights and Exceptions" in the amendments that limit the language he claims requires a universal ban on QoS, specifically that allow the carriers to "offer varied service plans to users at defined levels of bandwidth and different prices." The word "defined" is the key, for what would QoS be without a defined level of bandwidth? Granted, it isn't a technical definition of QoS, but it is political speak for allowing such services to exist on dedicated or "private" circuits.

In response to my having pointed out that he had abandoned one line of questioning, that Net neutrality requires the "banning of QoS (quality of service agreements, which ensure a specific throughput and reliability)" across the Net, he writes:

For you to say I'm changing my tune is a lie. You're the only one jumping from imaginary language to external sources for your arguments. I've always been against the Markey and Snowe-Dorgan amendment for the reason that they would outlaw the sale of QoS. The fact that you imagine it only outlaws the sale of QoS on the backbone is irrelevant because:

A) You can't make a reasonable argument that it is only applicable to the backbone. Citing language external to the Markey and Snowe-Dorgan bill or language you imagine wouldn't fly in court. Otherwise why in the world would Markey and Snowe-Dorgan rely on built-in disclaimers? The fact of the matter is, an amendment has to be able to stand on its own merit or else it's worthless. The Snowe-Dorgan and Markey amendments that you keep advocating were worthless.

George, I have made the argument that the amendments would not ban QoS everywhere, you simply haven't refuted it. My observation is that you changed your tune while continuing to sing the same, refuted song. Making me out to be a "liar" in response is simplistic hyperbole. We are each entitled to our opinions and I believe I am informed enough about your arguments to note when they change.

The backbone should be "neutral" because it is essential to the architecture of the network
. If you refuse to acknowledge that blocked paths and administrative latency associated with dynamic billing would substantially reduce the performance of all networked services, you're not living in reality.

As for my understanding of the Net, it is not for you to judge when you fail to acknowledge the reaities of the paragraph immediately preceeding this one.

Continues George:

B) The sale of QoS shouldn't be banned anywhere, not on the backbone and not on the last mile. QoS works best when it's enabled on every leg of the trip a packet takes. Banning the sale of something perfectly legal anywhere is just wrong.

C) It's clear you don't understand the Internet because if you did, you would know that it's impossible to define what is a backbone or what is the last mile. Plenty of companies connect directly to the "backbone" by placing their own equipment in Internet peering hotels and negotiate their own connections with multiple Tier 1 providers. The idea that you could apply different rules to the "backbone" or "last mile" is based on pure ignorance of the way the Internet works.

"Net neutrality" is quite clear on this issue, as you point out when you say "QoS works best when it's enabled on every leg of the trip a packet takes": On a dedicated circuit you can charge for QoS, but on shared circuits you have to treat all packets equally. This would mean that some of the last-mile segments that are shared, such as cable companies offer, would have no QoS, but they don't today, either. Under "Net neutrality," the cable company could, however, install a port in the home that was, for example, a "Google Port" that was dedicated to Google services and offered sustained 100 Mbps services—if they could support those speeds. They could charge Google whatever they wanted for that, if Google would pay.

The infrastructure works because it is shared. The spirit of the Net neutrality position is that sharing requires equal treatment on those shared legs.

You persist in basing your arguments in an attack on my knowledge, but you continue to demonstrate how inflexibly committed to carrier "economics" you are. I'm sincerely asking you to re-examine your notions rather than become more set in them. I have considered your arguments and responded to them, explaining when and how they are wrong. The rest of us have actually paid attention to what those "economics" have done when applied without regulation to the marketplace evolving on the Internet. Throughout this engagement, you have repeatedly used the notion I am a fool and a liar as your primary argument.

At last, you've responded to the question I put to you and this is your answer. We should be able to put this to bed—agreeing to disagree without having to accuse the other of lying or stupidity—with this posting, which you and anyone else should feel free to respond to constructively and politely. I don't have time for people who begin an argument with "liar" and "fool," nor do our readers.

Hey, it's like an Update before I posted: In fact, since I published the comment on which this posting is based, George is back in comments, calling me ignorant again and pointing to his sustained attack on our fellow ZD Net blogger, Russell Shaw, in which he says:

Russell, of course you don't think this is a good idea when you have no understanding of traffic engineering (the usual dismissal of the opponent by a propagandist).... I'm proposing exactly what the true spirit of Net neutrality should be (the typical usurptation of the opponent's position by a propagandist).  Instead of having the ISP adding infrastructure to only suit their highest paying customers, I would mandate at least half of that new infrastructure is designated to general purpose "best effort" class Internet to benefit everyone.  If you're suggesting that any new infrastructure should be completely given to the general purpose Internet, then you are naive because you're never going to get any new infrastructure without some financial incentives for the people investing in the new infrastructure.

Now, the amazing thing is, we almost agree, but George's is a position that refuses to acknowledge the political and business realities in which "traffic engineering" exists. The infrastructure today is not "completely given over to the general purpose Internet," either, and the Net neutrality amendments explicitly say that in the "Preserved Rights and Exceptions" sections. In fact, the "general use Internet" is precisely what the Net neutrality position hopes to preserve, by ensuring a "general use Internet" exists with rules that provide fair and equal access to the market—any company may still buy dedicated circuits to bring them logically close to their customers. It is the obligation of the common carriers to make sure that, if there is a shared hop in the network, all traffic travels at the same throughput and reliability.

Also, George is intent on disliking me because I am a "liberal" and because I gave him as good as he served and more. I don't blame him for not liking me, since I can be a jerk, too. Dismissing your opposition by dismissing them personally or their politics generally rather than dealing fully with the realities of the issue is intellectually bankrupt. I'm trying, George, to address your position. You aren't ignorant, but you are refusing to deal with everything involved.

Specifically, George is refusing to deal with the growing barriers to competition at several layers of the Net, from basic connectivity (what if, for example, third-party DSL carriers were effectively blocked from the local loop by carriers imposing QoS fees in the last mile for their traffic?), applications (the "tax" a carrier could place on a packet from a Web services provider that they charge extra to deliver at the best effort speed, let alone increased speed—and George, I am using the word "speed" in the loosest sense so that I don't have to define throughput and QoS here), and in content markets, where the same carrier tax could be applied or access blocked, as demonstrated here by Verizon. George has not commented on that posting, probably because it is damning of his political position, which needs to be reconciled with his technical arguments.

We're talking about a simple idea, that when you haven't paid for dedicated capacity at a given throughput and reliability (again, shorthand, George, so deal with the principles in your response) your traffic will travel at the same speed and with the same reliability as everyone else's traffic. The carriers' position is that they should be able to define that freedom of access. It is the long and successful tradition of the United States that it does not allow a monopoly or duopoly to operate free of regulations which define a minimum common good that they must fulfill in exchange for their market positions.

George is from all appearances against the tradition of freedom that made the Net a transformative economic phenomenon.

Topic: Browser

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

18 comments
Log in or register to join the discussion
  • It's about to get worse. Here comes "Cellular-Net Neutrality."

    I'm sure you've seen [url=http://news.zdnet.com/2100-1035_22-6091853.html?tag=zdfd.newsfeed]this story about a similar experience for former AT&T customers now on Cingular.[/url] The third paragraph describing the charges against Cingular just sounds too much like the crap the Internet is going through.

    I know there's a difference between the nets; The Internet is an open net owned by nobody (unless the telcos have their way), while Cingular owns their own cell-network. Still, this is too much of a coincidence for my liking.
    Mr. Roboto
    • Yep, it's a trend that the legislation should halt

      I think we're past coincidence....
      Mitch Ratcliffe
  • Engineering and tech are MEANINGLESS to the discussion.

    We could be talking about tin cans and string, that is not the issue before us. The issue is who is going to have ultimate control of the internet, the corporations or you and me. It really is that simple.

    What is the newest buzz word around the internet? Answer "Monetize". Many, myself included believe the internet is the most advanced and powerful communications tool in the history of communication. The ability to freely communicate with anyone in any corner of the globe is indeed VERY powerful, in fact it scares the pants off many governments. (maybe all of them)

    Allowing corporate (government granted monopolies) desires to carve up the internet into islands of high speed / low speed access simply to generate income is not acceptable at any level. To be as simple and straight forward as possible I make this statement. The internet may well be the most valuable tool ever created by and for mankind. We can NOT risk it to the whims of a corporate P&L report.
    No_Ax_to_Grind
    • One more thought, let them go broke...

      The telcos claim they can't buiuld infastructure unless they increase profits. Ok, then don't.


      But you can believe the competition will find a way. (Cities going wireless)
      No_Ax_to_Grind
      • Municipal networking worked in Tacoma

        after the carriers said the city would have to wait for more
        profitable cities to be networked first. Now, there are multiple fiber
        networks in the city. Real competition.
        Mitch Ratcliffe
  • Tim Berners-Lee does NOT agree with you

    Here are two quotes from Tim Berners-Lee:

    http://dig.csail.mit.edu/breadcrumbs/node/144
    "Net Neutrality is NOT saying that one shouldn?t pay more money for high quality of service. We always have, and we always will."

    http://dig.csail.mit.edu/breadcrumbs/node/132
    "We may pay for a higher or a lower quality of service. We may pay for a service which has the characteristics of being good for video, or quality audio."

    Sounds like Tim Berners-Lee is saying the same thing I'm saying, that ISPs SHOULD be able to sell QoS.
    georgeou
    • show me

      George, show me one instance of my advocacy for banning QoS
      universally and you might be right. Unfortunately, I have said
      exactly what Tim Berners-Lee is saying, that QoS is perfectly
      acceptable on dedicated circuits, where customers pay for better
      service based on the provider having gotten their content as close
      to or directly to them with the best performance. See my example
      of the "Google Port."

      But rest assured, George, that Sir Tim would not agree with you.
      Mitch Ratcliffe
  • Vint Cerf works for Google now, he was singing a different tune at WorldCom

    Vint Cerf is HARDLY an unbiased source. Vint works for Google now. The problem is that Vint was saying something completely different while he was at WorldCom that the Internet should NOT be regulated.

    Vint Cerf on Internet regulation (RFC 3271):
    http://tools.ietf.org/html/3271
    Internet is for everyone - but it won?t be if Governments restrict access to it, so we must dedicate ourselves to keeping the network unrestricted, unfettered and unregulated. We must have the freedom to speak and the freedom to hear.

    Internet is for everyone - but it won?t be if it cannot keep up with the explosive demand for its services, so we must dedicate ourselves to continuing its technological evolution and development of the technical standards the [sic] lie at the heart of the Internet revolution.

    Furthermore, the link you provided says nothing about Vint proposing to outlaw the sale of QoS.
    georgeou
    • You are arguing that Vint Cerf would break the Internet?

      That's what you are accusing Vint of doing, selling out. I've
      known Vint in a number of capacities and can say you are wrong
      and offering a grave insult to him if you insist on this position.
      The quote you offer from the RFC is pro Net neutrality in pre-
      Net neutrality language and anyone can see it. He is talking
      about limits on access being bad.

      George, I don't know how many times and ways I can say to you
      that I am not for outlawing QoS and that the Markey and Snowe-
      Dorgan Amendments would not outlaw QoS agreements on
      dedicated circuits. Please, refrain from repeating an already
      discredited argument over and again.

      You've bent my argument so far out of the reality it represents
      that you aren't even talking to me anymore, but to an idea of
      what you think I represent. Read every posting of mine on this
      topic and you will not find any instance of me arguing for the
      banning of QoS, except where you injected it through
      misreading the amendment language. I challenge you to find a
      statement on my part that I support banning QoS universally.
      Mitch Ratcliffe
      • Vint Cerf

        Vint Cerf is getting up in years, and hasn't practiced engineering for twenty years, so why don't you leave the poor bat alone.

        He's obviously gone senile.
        richardbennett
        • Speaks volumes about you, Richard

          You're not a class act, Richard.
          Mitch Ratcliffe
  • This is the silliest thing ever written about Net Neutrality

    In the first place, Mitch, if you want people to take you seriously as someone who claims to understand the Internet, you need to purge your vocabulary of the word "circuit." The Internet is all about shared links, it's a packet-switched network, not a circuit-switched one. That's how it's able to provide high performance to applications that access the network for short bursts of traffic.

    In the second place, your interpretation of Markey and Snowe-Dorgan is exactly 180 degrees out of phase with the intent of the the authors, who have explicitly claimed on several occasions that they're out to regulate ISP behavior on the last mile, not the inter-ISP behavior that the ignorant call "the backbone." So you need to get on the phone to Ben Scott at Free Press and get your marching orders corrected. Those of us who don't like the last-mile QoS ban in Markey and Snowe have pointed out that the language is so vague that it applies on all broadband segments, and they dispute that.

    What you've done here is read the Markey and Snowe amendments in a completely quirky way because you can't afford to acknowledge they're out of sync with Berners-Lee's statements. It's an admirable attempt at sycophancy, but ultimately not persuasive.

    The main error that you folks are making is to try and make the Internet into a network. In fact, it's not a network, it's a method of interconnecting networks. That's what the prefix "inter" means, you see. When the Internet was conceived 30 years ago, the networks that it interconnected were Ethernets with one level of packet delivery service, the so-called "best effort" which consisted of 16 retries after a collision. That model went obsolete with the advent of VLANs, and MPLS was added to the Internet to preserve VLAN classification. Nowadays we have WiFi networks with 802.11e and four levels of priority, so we naturally need an Internet that can honor them and interconnect WiFi networks correctly.

    This is simply the forward march of technology, and anybody who hasn't made a fetish of "end-to-end" and "best effort" can see that.

    Any well-informed and rational person, that is.
    richardbennett
    • LOL, great point Richard

      "In the second place, your interpretation of Markey and Snowe-Dorgan is exactly 180 degrees out of phase with the intent of the authors, who have explicitly claimed on several occasions that they're out to regulate ISP behavior on the last mile"

      Great point Richard! They are indeed for regulating the Internet and specifically on the last mile since they often cite the "duopoly" at the last mile as justification. I'm glad you put the nail in this coffin.
      georgeou
    • 180 degrees?

      A packet-switched network is made up of circuits, which are
      dedicated or shared, so at this point what you're going to do is
      try to argue that I "don't understand the Internet" from yet
      another angle instead of dealing with the points I made about
      George's previous arguments. You're also suggesting, again, I
      have advocated "banning QoS," which I have not.

      The last-mile is where the end-user at home has purchased
      capacity. It is where, if the carriers can impose restrictions of
      any sort they defeat access to what George most recently
      described as "the general use Internet." So, from the law's
      perspective, it is "shared." A dedicated link or circuit or
      connection?whatever you want to call it?would not be covered
      by this, as the sponsors of the bills have made clear.

      The fact that the Snowe amendment failed through a tie vote
      indicates that there is a lot of room for negotiation. It should
      also have occured to you, Richard, that the legislation is using
      QoS in a different sense than the QoS than you are when talking
      about routers and IP values 6 and 7. They are talking about
      billing issues, not technical ones. It maps perfectly to Tim
      Berners-Lee's definition of Net neutrality, which I pointed to: "If I
      pay to connect to the Net with a certain quality of service, and
      you pay to connect with that or greater quality of service, then
      we can communicate at that level."

      In other words, if I pay for a "slow" connection, everything will
      come through slow. If I pay for a "fast" connection, everything
      will come through fast. If a content provider needs more
      bandwidth at their data center, they can pay for it and their
      traffic will get to the backbone faster, where it will be treated
      equally. If the content provider wants to put a cache at a cable
      headend, connecting all the headends with dedicated 100 Mbps
      connections, they can do so and pay for it.

      Now, let's look at your suggestion about the four levels of
      priority closely. The way the law could and probably would after
      it went through conference with a clear mandate for the
      inclusion of Net neutrality language would be this: Every priority
      level expressed by the IP values 6 and 7 in a given packet would
      be honored by every router on the "backbone" (in quotes, as we
      are reduced to jargon when talking politics) the same. the
      question for you, the engineer, is how you enforce a
      standardized use of the EF, AF1, AF2, and best-effort (BE) class
      selectors without having to set up a filter that must refer to
      customer billing information that introduced significant latency
      into the network. The answer is you probably can't, so the
      backbone simply treats traffic as traffic and lets the end-point
      customer define the priority level. Carriers make their money
      offering end-point connections, just as they do today.

      Finally, I haven't made a fetish of "end-to-end" and have not
      even used the phrase "best effort" in this series of postings.
      David Isenberg, who wrote the Rise of the Stupid Network, made
      some very important points about the faults in the business
      model at AT&T, most of which have been proven by experience.
      I suggest you reread it to understand better the idea of end-to-
      end services within an open network. I don't agree with all
      David's conclusions, but that's a different debate.

      The forward march left the carriers behind and they want to be
      back in control in a way that technology obliterated. They are
      now trying to regain what they lost through legislation, which is
      why the inclusion of Net neutrality principles are critically
      important, as Tim Berners-Less said: "I hope that Congress can
      protect net neutrality, so I can continue to innovate in the
      internet space. I want to see the explosion of innovations
      happening out there on the Web, so diverse and so exciting,
      continue unabated."
      Mitch Ratcliffe
      • What a total crock

        Let's try and educate you on some basic networking concepts, Mitch.

        1. "Quality of Service" is a term of art in networking, and it doesn't mean "speed" ("Bandwidth" is the closest approximation in the networking vocabulary to that pedestrian concept.) QoS means latency, jitter, and packet error rate. The section of Markey that permits service plan differentiation by bandwidth doesn't contradict the section of Markey that forbids service differentiation by QoS. I see no reason to jump to the conclusion - as you have - that the aide who drafted this amendment is as confused about these terms as you apparently are. When you say: [i]"It should also have occured to you, Richard, that the legislation is using QoS in a different sense than the QoS than you are when talking about routers and IP values 6 and 7. They are talking about
        billing issues, not technical ones,"[/i] you're blowing smoke.

        2. There is no "neutral backbone" inside the Internet, only a series of privately-owned links that carry packets to or from subscribers. This is true whether we're at the residential last mile, the business last mile, inside an NSP network, or inside a Tier 1 link. They're all private, and they all carry packets only for customers. So when you say: [i]"the question for you, the engineer, is how you enforce a standardized use of the EF, AF1, AF2, and best-effort (BE) class selectors without having to set up a filter that must refer to customer billing information that introduced significant latency into the network. The answer is you probably can't, so the backbone simply treats traffic as traffic and lets the end-point customer define the priority level"[/i] you're blowing smoke again; subscriber and QoS routing happens today, and with no introduction of latency because it's done by hardware. That's been the case for several years. Duh.

        What Tim Berners-Lee is probably talking about is the fear that ISPs that make deals to speed up some best-effort traffic over over best-effort traffic -- by caching content inside the offices -- are somehow violating a sacred principle that exists in his mind. But in fact, the caching of content is already being done, and the very people who are advocating for these regulations are doing most of it. Double duh.

        I appreciate that your concept of "dedicated circuits" and "shared circuits" is your valiant attempt to read some coherency into the muddle of neut regulations, but you've only succeeded in making your own amendment embodying your own concept of neutrality. Everybody else is doing this too, so have a ball. Just don't think your musings have the first thing to do with any bills pending in Congress.
        richardbennett
        • You're trolling, Richard

          You've posted precisely the same language in two different places
          in responses to different comments or postings. I've answered you
          in the most recent thread. But be polite and don't troll the
          argument by popping up every bridge....
          Mitch Ratcliffe
          • Are you completely nuts?

            You posted exactly the same thing in a talkback here and in a separate blog posting. Of course my response is the same to your identical posts, you flaming nitwit.
            richardbennett
          • Name-calling....

            We're being nice, Richard. You're trolling, I am trying to bring the
            blog readers the story. You got your ideas in the postings, so try to
            stay with the group.

            The postings are actually different, not exactly the same.
            Remember, read carefully.
            Mitch Ratcliffe