Wireless

by Carlo Longino




Google Wants To Auction Spectrum Now, Too

from the i'll-take-10k-impressions-and-2-GHz dept

Google's been talking about wireless spectrum for a while now, both as part of a coalition asking for the white space around TV channels to be opened up, but it's also been mentioned as a possible bidder in the upcoming 700 MHz spectrum auctions later this year. While Google's not commenting on whether or not it will bid on the spectrum, it has made a proposal to the FCC to allow for a "real-time airwaves auction model" for it. Google wants to set up a system for the spectrum that would work much like AdWords does, only instead of advertisers bidding for impressions, service providers would be bidding for spectrum. While we're supportive of ways to make spectrum more flexible and useful, and support the idea of allowing spectrum to be leased and sold in general, it's hard to see a lot of benefit in Google's plan. It says the auction would allow for spectrum to be fully utilized by letting companies easily auction off excess in real-time. That may be true, but it seems unlikely that there will be a lot of interest from potential bidders. Short-term leases that carry few guarantees aren't exactly attractive if you're trying to build a sustainable business -- for instance, it's not a great idea to start up a mobile broadband service one month, then tell your customers that you're sorry, but you got outbid this month, so you've gone dark, but they can be sure and check back next month to see if the service is up again. Perhaps the bottom line here is that spectrum owners should have the flexibility to do something like this, or lease their spectrum another way, should they see fit. But simply having that ability doesn't automatically make it a good idea to do so.

Reader Comments

Subscribe: RSS

View by: Time | Thread


  • identicon
    Bubba, 22 May 2007 @ 12:12pm

    You're thinking a step too slow

    What you're not imagining is an abstraction layer between the physical frequency and the "channel" (for lack of a better term), sort of like DNS. If there was a way for content providers to switch around frequencies, buying more or less bandwidth here or there, it could work if there was a server that pointed clients to the correct frequency.

    Now who could do such a thing? I bet Google would be willing...for a small fee from the content provider.

    The idea is this: the internet sucks (comparatively) for broadcasting because of redundancy - it's not much easier to send packets to a million people than it is 1000. But broadcasting does scale nicely that way. So what if you had an appliance that was hooked up to the internet (say an 802.11 card) and also had over-the-air reception? The device would ask Google where to look for the content (minimizing internet bandwidth) while pulling the content over the air.

    The more I think about it, the more I'm sure this has to be what Google has in mind. It would give them a fantastic role as middleman.

    reply to this | link to this | view in chronology ]

    • identicon
      Josh E, 22 May 2007 @ 12:43pm

      Re: You're thinking a step too slow

      That works well though for applications that don't require/need error checking of the packets being sent. In other words broadcasting really only works for applications that it doesn't matter if the final product is distorted, which is why radio/broadcast TV work so well in this area. People can fill in the gaps of a distorted sound or picture. Computers and the data they push around on the internet on the other hand, don't far as well. Imagine trying to read Techdirt on such a system that is so easily interfered with. You'd maybe get something meaningful through, but more likely you would get a bunch of gobbledygook.

      reply to this | link to this | view in chronology ]

      • icon
        Griff (profile), 22 May 2007 @ 1:41pm

        Re: Re: You're thinking a step too slow

        Josh E says that broadcasting really only works for applications where distortion of the data is acceptable.
        I would say that if you have more bandwidth than you need and good forward error correction this can be made to work pretty well. And if you have a broadcast system as Bubba described where you can get low bandwidth feedback on reception quality back via the net in real time, you could lease more bandwidth in real time if your user base started suffering - or use the conventional net for the (occasional) error corrections. Just like the original intent of internet, it could re-route traffic dynamically, but rather than dynamically changing the path taken it could dynamically change the amount of spectrum used to cope with reception / traffic.

        Remember the news stories surrounding Google's big datacenter in North Carolina ? If they can cache commonly downloaded data right out near the end users and update it via a (variable bandwidth) broadcast, it doesn't matter if they lose a bit from time to time as they just go and fetch the missing bits via the conventional net. This allows them to offer apparent high bandwidth access in quite remote areas where the cost of "conventional" net backbone is high. Yes a few people are using bittorrent but on the whole most folk download far more than they ever upload.

        reply to this | link to this | view in chronology ]

  • identicon
    Andrew W, 22 May 2007 @ 12:31pm

    Not quite an accurate reading of the Blair Levin q

    Carlo wrote, "it's not a great idea to start up a mobile broadband service one month, then tell your customers that you're sorry, but you got outbid this month." That's not something that could happen given the model. If you successfully bid for another licensee's excess spectrum, it's yours. You can renew it with the FCC as any other licensee does, so long as your using it in the public interest, broadly defined.

    What Carlo might have been referencing is the Blair Levin quote that "a potential problem could arise if a bidder invests significantly in devices for the consumer market that would use spectrum but were then outbid in the auction." Levin is just acknowledging a reality for all potential licensees--you have to use your capital to lease a swatch of spectrum first before you invest heavily in your physical product. It's a big risk, but it's the same risk every cell phone company, satellite radio provider, and FM conglomerate faced and profited from.

    reply to this | link to this | view in chronology ]

  • identicon
    anonymous coward, 22 May 2007 @ 1:11pm

    one word:

    backhaul

    look it up.

    reply to this | link to this | view in chronology ]

  • icon
    Steve R. (profile), 22 May 2007 @ 1:35pm

    Spectrum cannot be sold

    "support the idea of allowing spectrum to be leased and sold in general" (emphasis added)

    Big mistake. I have no problem with the FCC leasing spectrum. However, the spectrum is a public resource. If spectrum is actually sold we would end up with a private FCC that would have the look and feel of the RIAA and the MPAA. Do we really need another private entity that has no regards for consumer rights?

    In theory the FCC manages the spectrum for the public good. If privatized we would have all sorts of slippery slope arguments over who owns what and what one can do. This would be a nightmare.

    reply to this | link to this | view in chronology ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Use markdown for basic formatting. HTML is no longer supported.
  Save me a cookie
Follow Techdirt
Insider Shop - Show Your Support!

Advertisement
Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Advertisement
Report this ad  |  Hide Techdirt ads
Recent Stories
Advertisement
Report this ad  |  Hide Techdirt ads

Close

Email This

This feature is only available to registered users. Register or sign in to use it.