The TV in the Snake of Time

There’s only one way to justify Internet data speeds as lopsided as the one to the left.

Television.

It’s an easy conclusion to draw here at our borrowed Parisian apartment, where the Ethernet cable serving the laptop comes from a TV set top box. As you see, the supplier is FreeSAS, or just http://free.fr.

I don’t know enough French to interpret that page, or the others in Free’s tree, but the pictures and pitches speak loudly enough. What Free cares about most is television. Same is true for its customers, no doubt.

Television is deeply embedded in pretty much all developed cultures by now. We — and I mean this in the worldwide sense — are not going to cease being couch potatoes. Nor will our suppliers cease couch potato farming, even as TV moves from airwaves to cable, satellite, and finally the Internet.

In the process we should expect the spirit (if not also the letter) of the Net’s protocols to be violated.

Follow the money. It’s not for nothing that Comcast wishes to be in the content business. In the old cable model there’s a cap on what Comcast can charge, and make, distributing content from others. That cap is its top cable subscription deals. Worse, they’re all delivered over old-fashioned set top boxes, all of which are — as Steve Jobs correctly puts it — lame. If you’re Comcast, here’s what ya do:

  1. Liberate the TV content distro system from the set top sphincter.
  2. Modify or re-build the plumbing to deliver content to Net-native (if not entirely -friendly) devices such as home flat screens, smartphones and iPads.
  3. Make it easy for users to pay for any or all of it on an à la carte (or at least an easy-to-pay) basis, and/or add a pile of new subscription deals.

Now you’ve got a much bigger marketplace, enlarged by many more devices and much less friction on the payment side. (Put all “content” and subscriptions on the shelves of “stores” like iTunes’ and there ya go.) Oh, and the Internet? … that World of Ends that techno-utopians (such as yours truly) liked to blab about? Oh, it’s there. You can download whatever you want on it, at higher speeds every day, overall. But it won’t be symmetrical. It will be biased for consumption. Our job as customers will be to consume — to persist, in the perfect words of Jerry Michalski, as “gullets with wallets and eyeballs.”

Future of the Internet

So, for current and future build-out, the Internet we techno-utopians know and love goes off the cliff while better rails get built for the next generations of TV — on the very same “system.” (For the bigger picture, Jonathan Zittrain’s latest is required reading.)

In other words, it will get worse before it gets better. A lot worse, in fact.

But it will get better, and I’m not saying that just because I’m still a utopian. I’m saying that because the new world really is the Net, and there’s a limit to how much of it you can pave with one-way streets. And how long the couch potato farming business will last.

More and more of us are bound to produce as well as consume, and we’ll need two things that a biased-for-TV Net can’t provide. One is speed in both directions: out as well as in. (“Upstream” calls Sisyphus to mind, so let’s drop that one.) The other is what Bob Frankston calls “ambient connectivity.” That is, connectivity we just assume.

When you go to a hotel, you don’t have to pay extra to get water from the “hydro service provider,” or electricity from the “power service provider.” It’s just there. It has a cost, but it’s just overhead.

That’s the end state. We’re still headed there. But in the meantime the Net’s going through a stage that will be The Last Days of TV. The optimistic view here is that they’ll also be the First Days of the Net.

Think of the original Net as the New World, circa 1491. Then think of TV as the Spanish invasion. Conquistators! Then read this essay by Richard Rodriguez. My point is similar. TV won’t eat the Net. It can’t. It’s not big enough. Instead, the Net will swallow TV. Ten iPad generations from now, TV as we know it will be diffused into countless genres and sub-genres, with millions of non-Hollywood production centers. And the Net will be bigger than ever.

In the meantime, however, don’t hold your breath.

12 comments

  1. Brent Ashley’s avatar

    I used to believe that the asymmetrical nature of cable ISP data rates was purely a policy decision.

    I’ve only recently come to understand the history of cable television’s transition to cable internet provision. This document gives a very good overview of this topic up to a few years ago:

    http://www.cisco.com/en/US/docs/internetworking/technology/handbook/Cable.html#wp1020564

    With DOCSIS 1.x, the maximum raw upstream data rate is about 10 Megabits per second. This increased to 30 Mbit/s with DOCSIS 20 and with DOCSIS 3.0, multiple bondable upstream channels brings the possible upstream rate to over 100 MBit/s.

    Now that I understand the constraints of the infrastructure, how it has been engineered forward, and where it’s going soon, I’m a little less pessimistic than I was when I was in the dark.

  2. Doc Searls’s avatar

    Hey, Brent.

    Some asymmetry is natural and architectural and has to do with the way traffic optimally moves. (Richard Bennett explained more in a comment here a few months back. Can’t find it now. Maybe he’ll weigh in again with remedial facts.)

    Internet on cable has been optimized from the start to work in a TV-based environment (even the “6MHz channel spacing” is a TV convention). Getting Internet on TV cable has been in some ways a marvelous thing from the start. Very creative and resourceful.

    But still biased for TV.

    The lopsided data speeds above are mostly a symptom of policy. My guess is that upstream Internet data here is pushed down to a lower RF channel with less data carrying capacity.

    In towns around where I live in Massachusetts, Comcast downstream speeds hit 50Mbps in bursts, and are routinely around 30Mbps. Upstream they promise up to 10Mbps, but the best I’ve seen is about 4.

    At our house in California, Cox delivers about 30Mbps down and 4 up. They’ve told me their plans for DOCSIS 3 are to deliver 100Mbps down and 5Mbps up. Why? No demand for more. But how can they tell? I hang out more at our apartment in Mass at least to some degree because we get an upstream speed of 25Mbps (it’s symmetrical) from Verizon FiOS.

    But all that is beside my point, which is that TV is moving to the Net (and cellular data), and will dominate (or at least inform) both engineering and business decisions for the easily forseeable future. My own take is that this will still be a temporary stage after which ambient connectivity will have ample speeds in both directions, to accommodate an infinite variety of uses, rather than mostly-TV.

  3. Seth Finkelstein’s avatar

    Well, maybe there’s something I’m missing, but, seriously, I don’t really grasp your argument. As far as I can make out, the logic seems to be this (stripped-down, correct me if I’m wrong):

    1) A society where everyone was an equal peer of everyone else would be great (world-of-ends, “democracy”, blah blah blah).

    2) Because of #1, we should make Internet architecture rigorously based on that ideological idea.

    3) And then if we do #2, it’ll revolutionize politics, because by making the network architecture like that, then citizen interaction will be like that (well, encouraged to be like that).

    I’m wary of making a straw-man here, but this is what comes across. And it’s why there’s often a lot of flamage about it being nonsense. It’s voodoo networking. Worse, it diverts from very real issues such as gatekeepers and “power-laws” into fuzzy buzzwords that are very corporate-friendly (by making everything about the right technology).

    Oh, regarding:
    > When you go to a hotel, you don’t have to pay extra to get water …

    In places where water is expensive, you do have to do this if you’re going to use lots of it – it’s called “room with a bath”, and it costs extra. I agree with you that bandwidth will get cheaper in the future, but now it’s still relatively expensive, so it’s segmented out. Saying that it shouldn’t be so right now, seems unreasonable to me.

  4. Doc Searls’s avatar

    Hey Seth.

    Actually my argument is pretty much what you say in your last paragraph.

    Backing up a bit, I’m also saying:

    1) It’s hard to overstate the size of the shift of TV “content distribution” from old pipes (over-the-air, cable, satellite, DVDs) to the Net, which will increasingly include cellular systems.

    2) Because of #1, those of us who would like the Net to be neutral to the content that flows through it, are going to be disappointed for awhile.

    3) The Net will eventually return to a relatively neutral state because it will serve many purposes other than what we used to call television.

    Your point, “now it’s still relatively expensive, so it’s segmented out” summarizes that nicely. I’d just add that it’s bound to get cheaper eventually (and unevenly, and messily, as things do).

    I’m not saying “it shouldn’t be so right now.” I am saying that there will persist a relatively Utopian ideal (which many of us call “neutrality,” though that’s a word I never liked because it’s a political lightning rod) toward which the design of the Net in the first place is biased, and toward which its development will head, eventually.

  5. Seth Finkelstein’s avatar

    One unclear aspect is the implications – or lack thereof – among:

    1) Bandwidth will get cheaper

    2) Asymmetry between upload and download ability

    3) Consumption vs. production (gatekeepers, “power laws”)

    Point #1, we agree and seems obviously true. I don’t see that it necessarily implies point #2, which I think is likely to remain true based on the fact that most people are consumers, not producers. However, as to point #3, you say:

    “But it will get better, and I’m not saying that just because I’m still a utopian. I’m saying that because the new world really is the Net, and there’s a limit to how much of it you can pave with one-way streets. And how much of the same kind of stuff we’ll tolerate.”

    How do you know? Is this falsifiable, or meant as technological determinism that will not be admitted wrong?
    As far as I can make out, the argument is more connectivity means more opportunities for production which means good winning over bad. However, that utterly ignores all the ways such an argument has and can be expected to fail – persistent asymmetry, new barriers, the entire social infrastructure which favors bread-and-circus, etc.

    I mean, I could spin some tale like “Health care reform is inevitable because of the Net, as it’ll defeat all the lies of the moneyed interests and we’ll have a public option” – and while that might sound good, it would be quite wrong (as far as the real world demonstrates). And one reason why it’s wrong is that powerful interests can use the Net too, for example hiring demagogues.

    Sigh. I ramble. :-(

  6. Bruno Amaral’s avatar

    Allow me to add just one small thing: just because we have access to an internet connection, it does not mean it gets used or used properly.

    Here in Portugal 51% of the population has access and uses the Internet. Most of this growth comes from artificial incentives (laptops for school children, founding for teachers etc.).

    However, there aren’t any computer literacy programs that I know about. (And that means that if we have them, they aren’t widespread)

    We do have half of the population with access to a very useful tool, not caring if we help them make the best of it and use it to its full extent.

    one last thing … Thank you all for a very interesting read :)

  7. PJ’s avatar

    I think there are reasonable technical reasons for asymmetry. People in general create/upload a lot less data than they download just because there’s a lot more people on one side of the last mile than the other. It’s totally reasonable for popular content to be forced to move to a better-connected site prepared to deal with higher demand, and better able to take advantage of CDNs and other network caches.

  8. Doc Searls’s avatar

    Where I am in France, visitors buy showers, electricity and Internet connectivity by the minute our hour, so I don’t have much time to reply. So, briefly…

    Seth, your argument makes sense, and to some degree for me various possible futures are a matter of faith, or personal biases. On some matters I’m an incurable optimist. The Net is one. For me that glass is partly full, and always will be. But I do mean to research All This Stuff more, which is part of the work I’m doing on a book right now.

    PJ, there are good technical reasons, having to do with network design, for some degree of asymmetry. But bandwidth does not respond only to demand. It also expresses policy. If you don’t want upstream activity, squeeze it with slow upstream speeds. That’s what’s happening with the example from this post above. I’m not going to upload many photos, nor am I going to do offsite backup. Both of those are activities I do at home (where I have high upstream speeds), and are businesses for my providers. Those businesses are prevented by the policies embodied in the speeds I’m seeing here. (Including the connection I’m using now, provided by France Telecom: 6Mb down and .6Mb up. (Probably ADSL.) There’s not much harm in it, but also not much good for new businesses that might depend on upstream speed.

    Wow… I just hit “submit comment” and I get redirected to a site that says …

    This Connection is Untrusted

    You have asked Firefox to connect
    securely to sc.quickspot.fr, but we can’t confirm that your connection is secure.

    Normally, when you try to connect securely,
    sites will present trusted identification to prove that you are
    going to the right place. However, this site’s identity can’t be verified.

    What Should I Do?

    If you usually connect to
    this site without problems, this error could mean that someone is
    trying to impersonate the site, and you shouldn’t continue.

    Technical Details

    sc.quickspot.fr uses an invalid security certificate.

    The certificate is only valid for the following names:
    http://www.quickspot.fr , quickspot.fr

    (Error code: ssl_error_bad_cert_domain)

    I Understand the Risks

    wtf is THAT?

    (time passes, as it does in France, only more slowly, when Procedures are involved…) I just found out. I had only received 20 minutes of connectivity. I just got 2 hours, and now I’m back on.

    BTW, when I signed on, the service wanted a big pile of personal info, and then to opt out of receiving promotional emails from it and its ‘partners’. I gave them all false crap then opted out. Pisses me off, this shit. This is at the town’s tourist office, incidentally. How to turn off the tourists…

Comments are now closed.