A tragicomic excerpt from Private Eye magazine (a scurrilous UK satirical bi-weekly) neatly shows how computers are not social, and only people are. (The jury’s out on the replicants, though.)
The tsunami disaster seems to have gone unnoticed in the Daily Telegraph’s advertising department.
Three days after the waves retreated, with the death toll standing at 60,000 and rising and most of the Telegraph’s news section devoted to the devastation left behind, the paper’s front page was inviting readers to “Have your DAMP PROBLEM solved the clean, green way! Although you rarely see this hidden enemy approaching, winning the war against condensation, rising damp, and mould can prove troublesome and costly, but not with the Schrijver Systeem®.”
Over on the paper’s website things were not much better. Their “targeted advertising” system, which selects products according to key words in news stories, decided that a suitable accompaniment to the tsunami coverage was an enticement to “relax with a soothing Ocean Waves CD”.
Offensive? Not really. Tasteless? A bit. Insensitive? Definitely. Obvious to you and I, but not a computer.
So, do you really plan on trusting your online personal reputation to these same machines? Scared yet?
The stupid network doesn’t end at a smart PC, phone or server. These are stupid too. You need a cerebral cortex to be smart. In the original end-to-end paper, the authors almost get there:
Identifying the ends
Using the end-to-end argument sometimes requires subtlety of analyis of application requirements. For example, consider a computer communication network that carries some packet voice connections, conversations between digital telephone instruments. For those connections that carry voice packets, an unusually strong version of the end-to-end argument applies: if low levels of the communication system try to accomplish bit-perfect communication, they will probably introduce uncontrolled delays in packet delivery […] The natural redundancy of voice, together with the high-level error correction procedure in which one participant says “excuse me, someone dropped a glass. Would you please say that again?” will handle such dropouts, if they are relatively infrequent.
They then contrast this with a system that forwards voice messages, which would use TCP/IP since you can’t ask the sender to repeat themselves when listening to a voicemail. So humans do make a guest appearance in the argument. But we don’t really get into the hard stuff of defining what the “ends” really are — just allusions to the issues.
However, in a 2002 clarification of the paper, David Reed writes:
It’s easy to resolve some of the confusions. Two are: what about servers that are owned by the network provider, and what about multipoint protocols. Well, the end-to-end principle never talked about ownership. So “economic bundling” has nothing to do with a service being “provided by the network” - the server is not “in the network”. […] One can discuss the ensemble of endpoints (even heterogeneous classes like SMTP relays [and email clients]) as the set of endpoints treated by an end-to-end argument used in the design process.
It would be nice if the technical design world were “closed” and isolated from the rest of the value system. But it’s not. People build networks and endpoints because of economic and social need. Fail to model these, and your model is incomplete.
If the end-to-end principle is such a powerful design Occam’s razor, then you have to ask yourself why so much of the communications world is not arranged according to its precepts. Why do we have 3GPP committees re-inventing smart network models? Why do we have the Bluetooth debacle? Why is SIP such a complex tangle of functionality across multiple layers? The answer is that you can’t separate the technical from the human, economic and political spheres, as the Daily Telegraph found out the hard way.
The flexibility to accommodate future requirements and change that the end-to-end principle enshrines begs an important question. Flexibility for whom? Whose changes? What the media companies and telcos are discovering is that there are many ways of anulling the user’s flexibility of a dumb pipe. DRM, network-owned retail stores and distribution, closed portals and unchangeable home pages, undocumented iPod hardware, DVD player patent licensing, etc. There’s a long and familiar list.
The result is that economic bundling is in effect inseparable from technical integration. Just like giving away Internet Explorer with Windows was all Microsoft needed to do; the faux technical integration wasn’t needed. They controlled the distribution. What if your mobile phone’s media player can only understand vodafone.com URLs? How does that materially differ from technical integration with the network? It doesn’t.
The end-to-end paper talks about applications defining new functionality. They don’t. Only people do this. The original paper implicitly assumes the goal is to maximise the system’s lifetime value to the end user. This is achieved by placing flexibility where the user has the greatest control over it.
This made sense when the Internet was a user-owned network. The universities and military weren’t trying to turn a profit. It makes sense for a municipal network of the future. What it doesn’t do is explain how commercial networks today are designed. For that, you need to look outside the technology box and see the whole picture: economic, social and political.
Posted by Martin Geddes at 11:54 AMTrackBack URL for this entry:
http://www.telepocalypse.net/cgi-sys/cgiwrap/mgeddes/MT/mt-tb.cgi/370.
Martin,
Your commentary runs counter to the stereotypical British humor. I read the ad bots choice as fodder worthy of mention in hitchhikers guide to the universe.
Your point about Bluetooth and SIP being hideously complex is true of most technologies when they come off the hype curve and reach the trough of disillusionment. CORBA is another perfect example. This of course right before they reach the plane of usability. Unfortunately the distance between these points in time is longer than we would like.
While as usual I agree with your commentary, it is worth noting that at least here in the colonies, the military and universities are trying to "turn a profit" in their own ways.
My point is that the "civilianization" of the military and higher education may help to narrow the temporal gap between hype, disillusionment and usability.
Posted by: at January 9, 2005 12:42 AM