Framing
The debates over the future of the internet should begin (although they hardly ever do) by answering the question What Is The Internet?
It turns out that how you answer that question correlates strongly with your view of the future. And it also turns out that using the “layers analysis” may play into a telco/cableco view of the future.
If you talk to a carrier (or a former carrier — now calling himself a “network provider”), he'll probably say that “the Internet” is made up of three chunks: the backbone, the last mile between a carrier and an enduser, and the connection between content providers and the backbone.
If you talk to one of the founding fathers of the internet, he'll probably say that “the Internet” is a collection of standards that allow the networking of computers.
If you talk to a current user/producer of internet “content,” he or she will probably say that “the Internet” is the collection of interactions and relationships that happen online. (I remember the first time I used the term “online content” in a lunchroom in 1993 and someone I was talking to laughed hugely – “Content?” he said. “What on earth are you talking about?”. Now we use “content” without shame, and maybe we shouldn't.)
Who's right, and what do these frames lead to? Well, it appears that if you're a carrier, and the internet is those three chunks of wires, and you invested in rolling out a lot of fiber, then — therefore — you have property rights in some part of “the Internet” and you need to be paid for their use. And if you're not paid for their use broadband penetration will remain low. For you, “transport” and “the Internet” are the same thing.
If you're a founding father, and the internet is the standards, and you're looking to be a founding father once again, you may suggest that “the Internet” is irretrievably broken and needs to be rearchitected. (Bob Kahn and David Clark may fit in this category.) For you, “protocols” and “the Internet” are the same thing.
If you're someone who goes online and is neither a carrier or a founding father, you may have expectations that “the Internet” will continue to be a free and ordered place whose value comes from interactions — not from the access valves used to get there. You're probably aware of internet “standards,” as well. So maybe you're worried about the effect on “the Internet” of the carriers' depredations. For you, “communication” and “the Internet” are the same thing.
It all depends what you think “the Internet” is. I think “the Internet” is a combination of standards and interactions/relationships. I'm with the founding fathers on this one, but I think their view can sometimes be a little narrow. These online interactions/relationships are persistent in a way no other network (and no mere “language”) has made possible. It's a new informational construct that can be separated from the substrate used to store/forward its elements. And so, because I'm neither a carrier nor a founding father, I'm worried about the future of “the Internet.” Our internet.
Thanks to Michael Froomkin for talking to me about a recent paper and suggesting that “standards and relationships” are what make up the internet.
Comments
8 Responses to “Framing”
Got something to say?
I agree :>
Maybe its not what the Internet is, but how it gets used.
I've been watching the 'tiered' internet calls from various telcos. Honestly, when is comes down to actually making it work the way the AT&Ts, the TimeWarners, and the MCIs want it to work so they can charge for enhancing network service XYZ (voice, video, etc.), here's one way to look at it: Lets say for example RoadRunner wants to charge extra so my VOIP or Internet streaming (or is that steaming) video works better. To accomplish this, they set the QOS - Quality of Service - settings on my VOIP and streaming video traffic frames so those packets get higher priority than web traffic packets, for example. QOS allows for a number of settings on a packet to determine a number of levels of prioritization. IP also supports TOS - Type of Service - that can be used with QOS to get certain types of traffic to take precedence of other type of traffic.
The problem is that because of how the Internet actually works, the QOS/TOS settings on those packets would have to be honored on each leg of their journey in order for the QOS settings to actually result in better service. That means if my VOIP traffic travels at times over AT&T or UUNET(MCI), which is more than likely, then AT&T and MCI will have to give priority to those packets or frames the same priority as RoadRunner originally gave them.
But, what if AT&T has also sold 'enhanced' VOIP or streaming video to some of their own customers? Whose packets get priority then? I would guess that the answer is that AT&T's internal QOS tagged packets will get priority over RoadRunner's.
I don't see the tiered Internet failing at a the home consumer level, since home consumers are accustomed to paying for something that doesn't end being as good in use as it looked at purchase time. Where the tiered Internet will run into some serious problems is with business customers. If TimeWarner, for example, were to sell some kind of 'priority video' or 'priority vioce' service over the wild Internet with an SLA that specified a minimum amount of bandwidth or response time, they would find themselves paying out refunds a significant amount of the time.
Again, the problem is how the Internet works. Once packets are launched onto the Internet, they will very likely traverse someone else's network, who won't honer your QOS the same way they will honor their own QOS.
It just seems to me that the outcome of a 'tiered' Internet as having traffic providers fighting with traffic providers in turn fighting with content providers.
The Internet is, in more ways then one, like the seas. There doesn't seem to be one provider out there that can tame it.
Bithead makes a great point, and there's a nice discussion of some of his/her points at internetweek.
However - to play devil's advocate a bit - if it's in the telcos' interests to create a channelized internet infrastructure, it stands to reason that they'll cooperatively adopt some QoS (diff serv) protocol. In his Senate testimony, Lessig admits that the idea of ISP competition protecting end-to-end network neutrality is pretty much DOA. I think he's right (though I still can't understand his support of a privileged - consumer - end of transmission, nor for that matter, a service-based privileging of transmission; but that's another story).
The real solution to these issues is increasing available bandwidth (fiber-to-the-home, wave division multiplexing, freeing-up of hoarded spectrum, etc.) so “tiering” becomes a moot issue. However, as long as the telcos are able to (continue to) convince the public, legislators and the FCC that bandwidth is a practically limited resource, their argument will seem cogent; allowing them to exact an even greater fee - for the paltry services they currently provide - by “tiering” their services. Moreover, they will have little to no incentive to invest in capicity growth with a “tiered” internet. In fact, one might argue that allowing “tiering” would actually provide a disincentive for investment in capacity, as it would only undermine the telcos' justification for lucrative “tiering” of services.
Finally, I'm at least a little concerned with the second to last sentence in the original post (I agree with everything up until that sentence :> ). The idea that the interactions/relationships made possible by the internet can be separated from the protocols for storing and forwarding elements is just a bit disingenuous. Sure, one can imagine another - different and/or improved - set of protocols for storing and forwarding transmissions. But, there are principles of network design implicit in the protocol substrate of “the Internet” that are the basis of its success; not just any old substrate will provide the basis for the interactions/relationships we value. You might say I've got a little “separation anxiety” . . . exactly how separate is a medium from its expressions? What would McLuhan say?
by substrate I meant the literal physical transport network, not the protocols.
I expanded on your point and connected it with others' in my post Internet Regeneration
I think the Internet is much more.
It is the revolution of our generation in the sense that it creates a new way of seeing the economical layers of our society.
The French Revolution did exactly that: it changed the players of the economic market.
No more few monopolistic people belonging to a rescricted economical Aristocracy, but many new people belonging to the new class: the one man company.
The Internet will allow the single individual to be a part of the Market.
The brain more than the capital will allow to make a new flourishing business.
The lower cost of hardware and the possibility to interact with the rest of the world in a very cheap way (a computer and an Internet connection)will allow many to be one man company.
We already saw that big ideas and few intelligent people could do more than big enterprises.
Monopolies will have to die, as the kings and the nobility died, to let the space to a new, vital, fresh generation of self made “interneteers”
Patrizia
http://woip.blogspot.com
Death to Videodrome - long live the new flesh.
Ah, the memories of “online content” in 1993. Most of the online content available was still in gopher and various anonymous ftp sites (remember http://ftp.wustl.edu?), as Mosaic was still catching on.
As I remember the definition of the Internet, it comes from internet. An internet is an interconnection of two disparate networks in such a way that they behave as one. The Internet is the big one, the ultimate interconnection of all networks. It is the Internet Protocol that makes the interconnection function in a useful way, yes, but that is not what defines it.
Back in 1993 when asked what I did for a living I told people I was in networking. Enough people mistaked what I meant as people networking not computer networking, so I got in the habit of adding computer as a standard so there was no confusion. Back then, the business of helping computers find one another and helping people find one another were entirely seperate things.
It is with some irony that now in 2006 we find ourselves in a place where this massive network of computers helps people find one another. Doesn't matter whether it's LinkedIn, MySpace, or just a google search, the computer network is the ultimate human network.
I guess I agree with Susan, even though I work for a carrier.