Cerf's call for simple pricing: Net Neutrality all over again

Cerf's call for simple pricing: Net Neutrality all over again

Summary: Vint Cerf, one of the co-inventors of the underlying technology that makes the Web work, now Google's Internet evangelist, has issued a call for throughput-based pricing by ISPs. This in response to the Comcast ruling by the FCC, which ordered the cable carrier to stop interfering with certain kinds of traffic, including P2P application data.


Vint Cerf, one of the co-inventors of the underlying technology that makes the Web work, now Google's Internet evangelist, has issued a call for throughput-based pricing by ISPs. This in response to the Comcast ruling by the FCC, which ordered the cable carrier to stop interfering with certain kinds of traffic, including P2P application data. Comcast has also floated the idea of charging more for users who transport larger volumes of data, an idea referred to as "consumption-based pricing," as another way to enforce data frugality on users.

Cerf correctly points out that consumption-based pricing is bad for everyone on the Net, except the carriers, who can profit more from their carriage services when they are taxed by so-called over-users of things like BitTorrent. A robust Internet economy needs no barriers or, better, locks, like those used to move boats, which shut off or curtails a service when it consumes too much bandwidth. He suggests:

Rather than a volume cap, I suggest the introduction of transmission rate caps, which would allow users to purchase access to the Internet at a given minimum data rate and be free to transfer data at at least up to that rate in any way they wish.

Others have suggested that users should be able to contract for a "floor" capacity and that they might possibly receive more capacity if it is available. One problem with charging for total bytes transferred (in either direction) is that users will have no reasonable way to estimate their monthly costs. Clicking on a link can take you to an unexpected streaming site or a major file transfer.

That means, in a nutshell, that I should be able to buy connectivity from my ISP at a specific level, let's say 6MBps, and use that 6MBps all day long every day, regardless of whether it is problematic to the carrier who sold it to me. The carrier's only business is to support the throughput they sell. Of course, they can and should manage their network, but it would be misrepresentation to sell a 6MBps service and deliver only 3MBps or to stop providing 6MBps throughput after the user sent more than any arbitrary amount of data, say 100GB of files. Cerf continues:

For example, a broadband provider should be able to prioritize packets that call for low latency (the period of time it takes for a packet to travel from Point A to Point B), but such prioritization should be applied across the board to all low latency traffic, not just particular application providers. Broadband carriers should not be in the business of picking winners and losers in the market under the rubric of network management.

Cerf is saying "6MBps should be 6MBps," which is the Network Neutrality position I and others have advocated along with him all along. Last year my former ZD Net colleague George Ou and I had a long, ugly argument about this issue, at which time he said Cerf did not take this view.

Once again, Vint Cerf has taken the position that bits should not be monitored for data type, network service type or volume delivered, except with respect to the amount of throughput the customer pays to receive at any given moment during the day. That is the definition of Net Neutrality: Traffic is unencumbered except with regard the guaranteed bandwidth the customer has paid for, so that low-latency traffic falls to the lowest guaranteed speed that the customer paid for at their network endpoint.

How does that affect the typical ISP customer: As media migrates to Internet Protocol, IPTV services, for example, will operate based on the throughput available to each end user rather than having all IPTV traffic or one user's IPTV traffic slowed or stopped because they've watched too much video on the Web. Rather, everyone would get their full paid throughput whether they were watching IPTV or chatting via IM or playing World of Warcraft.

Cerf's approach does allow for someone to pay less to get a lower top-speed at busy times, but it doesn't allow the carrier to arbitrarily choke off any data stream at its whim. A deal is a deal, so to speak.

CNET's Marguerite Reardon misinterprets Cerf's idea as "Internet speed limits," which implies that the carriers can impose a limit, but like the German Autobahn, Cerf is suggesting that the left lane have no speed limit as long as you can pay for the speed. If I have paid for the Ferrari that can travel at 40MBps, then I should get 40MBps, regardless of what everyone else is paying for their speeds and the type of traffic I am sending over the network.

NOTE: I have corrected the error in the original version of the article, which said the FCC fined Comcast. It did not. Rather, the Republican Chairman of the FCC went out of his way to admonish the company publicly and held out the threat of fines, saying the FCC had sufficient power to enforce its decision, if Comcast did not comply in order to head off Democratic legislation which would increase regulatory oversight of telecommunications carriers. I apologize for the error, which was one of haste and not intentional.

Topics: Mobility, Browser, Networking

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


Log in or register to join the discussion
  • Cerf's call for simple pricing: Net Neutrality all over again

    [i]Rather than a volume cap, I suggest the introduction of transmission rate caps, which would allow users to purchase access to the Internet at a given minimum data rate and be free to transfer data at at least up to that rate in any way they wish.[/i]

    ISP's are already doing this. Look at DSL providers. Sheesh, I thought Google employees were supposed to be smart. Leave it up to Google to try to reinvent the wheel. The carriers won't go for this as there will be too many problems. Someone says they can't get their 6Mbps so the ISP has to send a technician out to check the line to make sure it can and does give up the 6Mbps.

    I have a better idea, how about I continue to pay my flat fee per month to use my unlimited amount of data for my internet connection that isn't capped.
    Loverock Davidson
  • Should Google pay for...

    data I download from them or should I be charged for it. Or both.
    • Both of you already pay for the throughput

      and if you download the data at one speed, because that's
      all you paid for while they sent the data over a much faster
      connection, the marketplace is unencumbered. If, however,
      you have to pay more because it is one kind of traffic
      compared to another, the marketplace is encumbered by
      the carrier. The U.S. government has successfully
      intervened in markets throughout its history in order to
      remove this kind of encumbrance and keep the market a
      fair and ordered one. In this case, the reasoning is that you
      should get what you paid for, a guaranteed level of
      throughput without regard to the data type or the amount
      of data you've sent through that connection.
      Mitch Ratcliffe
      • Getting What You Pay for and some you don't

        Hi Mitch,
        Well with Net Neutrality, I'm sure most of you are already aware that you can all get much more than you pay for... Well if we are to believe the pledge from the Belgian actress-model-activist (Tania Derveaux), who has offered her beautiful bod up for consideration for any Virginian (The Great Common Wealth State) Virginia is after all for lovers:) who offers her their support for Net neutrality. I say get in line and pick a number quick. Belgian waffles are great but from the policy perspective this sweet treat is too appealing to pass up. Besides Although I've signed up with ATT for six megs tho I've never clocked over 3 so there is some sort of poetic justice in Ms Derveaux offer. And in the light of some of the recent commercials I would like to take this opportunity to, in addition to supporting Net Neutrality platform, we should all move to Virginia and form the Paris Hilton for President steering committee because there is another gal who looks like she can get the job done.
  • Do you understand what 1 Mbps wholesale dedicated access costs?

    First of all, you were dead wrong saying that Net Neutrality applied to the core of the Internet. Vint Cerf, and the legislation in the house, specifically pertained to the broadband leg of the Internet and not the core.

    Second, do you understand what 1 Mbps wholesale dedicated access costs? 1 Mbps dedicated access at wholesale prices is $100 per Mbps. If you want 6 Mbps guaranteed, then you better be prepared to pay a minimum of $600/month. Most of us are happy with shared 6 Mbps service at ~$40 per month.

    Third, both the legal counsel policy representatives of Google and Vuze at the Innovation 2008 forum (which I was a panelist at) advocated volume caps as a better alternative to network management. Free Press similarly advocated a preference to metered Internet service.

    Larry Lessig and Tim Wu have stated that volume caps would be a better way to manage the network and to raise the necessary revenue to pay for a faster Internet. I was the one who stood up at the Stanford FCC hearing to denounce all of them for advocating metered Internet access because a fair network management scheme was a superior way of dealing with congestion.
    • George is a legend...

      ...in his own mind. His testimony at Stanford was a joke.
  • RE: Cerf's call for simple pricing: Net Neutrality all over again

    Jesus Christ on a stick, Mitch, are you in a state of permanent warfare with reality? Look at the factual errors in this post:

    1. The FCC didn't fine Comcast, and you say they did.

    2. The FCC action against Comcast, an order not to inject TCP Resets into BitTorrent uploads, had nothing to do with metered pricing or consumption-based anything.

    Cerf's silly little plan fails to address the main component of network over-use, which is increased and unacceptable delay. Other people have addressed this issue, and done so intelligently.

    Do some research for a change instead of making stuff up. I like fiction writing as much as the next guy, but you're supposed to be doing news here, not fantasy.
    • Correction made

      And noted in the main article. What was notable about the
      ruling, besides the strident dissents by two FCC members,
      was the warning that the FCC could enforce it, which
      implied that it would fine Comcast if it didn't comply.

      Otherwise, Richard, I have learned that it isn't productive to
      argue with reality-benders like you and George.

      Interfering with data based on the application sending it is
      capping use on a basis other than what the customer paid
      for, a fixed throughput regardless of data type.

      Cerf's plan is only silly if you don't believe you should get
      what you pay for.

      Again, this is an opinion blog, not news reporting. I
      apologize for rushing through the writing and have
      corrected the error.
      Mitch Ratcliffe
      • What's silly is that you admited to an error but you say we bend reality

        What's funny is that FIXED bandwidth services already DO exist and it?s called a commercial-grade dedicated circuit with a CIR (Committed Information Rate). It's just that no consumers WANT to buy it because it's at least $100 per Mbps. That is the reality and it's interesting that you refuse to acknowledge it despite more knowledgeable people explaining it to you.

        The fact that this is a blog is not a license for you to err yet you continue to err despite being presented with factual information that contradicts your opinion.

        Furthermore, you?ve claimed that Net Neutrality is about the core of the Internet yet you lack the integrity to admit that mistake. Vint Cerf says it?s about the access layer (the broadband layer). In fact, Google has stated that Net Neutrality should have been called ?broadband neutrality? and there are no less than three separate bills in the house and senate that specifically use the word BROADBAND.

        The fact of the matter is that you?re very opinionated about Net Neutrality but you lack any credible knowledge on this subject.
  • Here's where you were wrong about claiming NN was about the backbone

    During a public debate with the Grandfather of the Internet David Farber, Farber asked Cerf (28:30 on audio) how he proposes to regulate and ban a new International protocol that they deem ?discriminatory?. Cerf responded (29:00):

    "Well I would respond on that one. What we?re concerned about is the access part of the net, not the core, not the backbone. But it?s discriminatory access to the network itself. Over that part of the channel we do have reasonable jurisdiction. The access component is here on American soil."

    You claimed http://blogs.zdnet.com/Ratcliffe/?p=156 that Net Neutrality applies to the backbone. You were dead wrong despite your childish antics calling me a Borg.

    Oh, and I've never seen a ZDNet blogger calling colleagues names that would only half-way apologize http://talkback.zdnet.com/5208-11515-0.html?forumID=1&threadID=22796&messageID=432409&start=0 and then take back the apology http://talkback.zdnet.com/5208-11515-0.html?forumID=1&threadID=22796&messageID=432537&start=-9920 despite other readers admonishing you for it http://talkback.zdnet.com/5208-11515-0.html?forumID=1&threadID=22796&messageID=432450&start=-9920.
  • Simple High Pricing

    Did you ever see the prices for CERFnet??? Holy cow.
  • RE: Cerf's argument is reasonable, but

    it means that ISPs have to significantly reduce the throughput claims. An ISP can't claim 6Mb/s. They can claim something much lower than that.

    IMO, consumers deserve several metrics: A min level of service, an average level of service and a maximum level of service.

    Altneratively, they could provide average (or mean if that's turns out to be a better metric) for different times of the day.