Proposed: an à la carte Internet

Proposed: an à la carte Internet

Summary: New proposal called 'ChoiceNet' would bring market discipline, enabling end-users to 'vote with their wallets' on end-to-end connections and services. Would it work?


More choice on the Internet? A dangerous radical idea, of course. But that's what a team of researchers, in a project underwritten by the National Science Foundation, is proposing. The new approach urges more of an "à la carte" approach to selecting network services. 

Keyboard Photo by Joe McKendrick

A report out of North Carolina State University (surfaced by Engadget's Alexis Santos), says the new model, dubbed "ChoiceNet," would "provide the building blocks to create different types of services and to create alternative services of the same type." End-users would be able to select the network or service that provides the best performance for their applications.

The researchers -- led by Tilman Wolf of the University of Massachusetts and Rudra Dutta and George Rouskas of NCSU -- note that with the current Internet structure, "competition exists only at the application layer, if at all."

End-users shouldn't be “stuck when the service they receive is not consistent with their expectations," the team writes in their abstract. "Rather, they must be able to choose a different service provider, to better meet their expectations."  The end result would be better Internet access across the board, since "economic rewards are explicitly represented in the architecture -- network service providers are incentivized to create innovative offerings."

They explain how ChoiceNet would ideally work:

"A user may be offered different connection services for the purpose of watching streaming video. These connections may differ in their technical implementation (e.g., quality-of-service, use of caching, etc.). The user selects a service (i.e., a complete package of end-to-end connection and related services) and pays for its use. Depending on the user’s satisfaction with the video experience, they continue to use the chosen service or switch to another (i.e., vote with their wallet)."

The model sounds similar to the unbundling of long-distance phone services that took place a couple of decades ago for landlines. End-users maintain their on-site local ISP service, as they do with local phone companies, and select who they use to complete a call. It makes sense to encourage competition to improve access and connectivity. The question is, will this make things more complicated? Will end-users buy into having to think about network selection, on the fly, every time they want to access an online service?

(Photo by Joe McKendrick.)

Topics: Networking, Tech Industry

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
  • Have fun with that.

    "End-users would be able to select the network or service that provides the best performance for their applications."

    Sounds great in theory.

    In practice, however, it's nearly impossible to unseat a local monopoly.

    And they're not gonna invest in an alternative system that kills their own monopoly.

    Have fun with that. I wish them the best, although I fear what they are proposing is unrealistic in today's totally messed up ISP system.
  • You da cold water man!

    Robert Hahn
  • Totally wrong

    This is totally wrong out of a Internet neutrally perspective. Each user of Internet pays for their own connection speed, data limit, stability etc. It's not Youtubes or Facebooks problem if you are on an expensive mobile connection. Smart users find the best deal for their demands on Internet connection and want to choose online services (streams, voip, news, games) that is NOT substidizing internet connections for their users. Companies and private persons each pay their own connections. Then it is easier to have a strong competition on the content (without lockins by your provider).
  • Internet not TV?

    It's a great idea in theory, but consumers can't even get TV a la carte. How would the internet concept be any different? How would it be more successful? I guess it could work if you could combine the two into some sort of Internet/TV entertainment package.
    James Keenan
  • Not good

    The model doesn't take into account the interests and ownership of the physical layers, and the indirect interests applied to sub-contracted "services". Imagine having to pay a separate Internet utility fee, video fee, text fee, chat fee, etc...
  • By the packet

    We could slow everything down dramatically by having the routers conduct little auctions every time a packet wants to go someplace. "I have here a packet that's trying to get to Kansas City," the router announces. The carriers have 150 milliseconds to respond. "I'll do it for $.003," says one. "I can get it as far as Cincinnati for $.0002," says another. And so on, all day long, packet by packet.

    Think of the savings! Think of the competition! Think of a method for charging users for the bandwidth occupied by the auctions!
    Robert Hahn
  • That is the DUMBEST suggestion ever made

    So ... some m0r0n thinks that paying MORE for a hell of a lot LESS is a great idea.

    Why the hell would anybody agree to downgrade the Internet? Because that is what this suggestion is ... a downgrade where you no longer have access to "the Internet" ... only the "networks" you were forced to paid ..... just like with cable.