Archive for September, 2004

As we head into more stormy discussions about the future of internet governance, let's keep in mind the differences between hierarchies (like governments) and networks (like the internet).

In a hierarchy, “subsidiarity” is solved by having higher levels be authoritative over the ones below.  In a network, nodes may not be “equal” (networks are usually scale free and subject to power laws), but no node decides for another. 

Hierarchies provide certainty and finality, both desirable things.  Networks make dynamic decisions in a constant ebb and flow of connections, each node vying for the attention of the others.

Hierarchies can last for a long time or can crumble when attacked at a sensitive place.  Networks can heal themselves by routing around broken connections.

Networks are “better” — in terms of their potential for the creation of complexity — than hierarchies.  Because of the competition for attention and deference that happens on networks, interesting variations and combinations can happen. In a hierarchy, variation depends on the creativity of the highest levels.  Biological developments — evolution — happen in networks.

And everything is biological in the universe.

There is nothing more inspiring and humbling than the study of complexity.  Everything is revealed to be biological, variations competing contextually and endlessly to create an order to which all things tend, but an order that trembles on the edge of chaos.  Great stuff.

The problem is that humans — and particularly humans selling products and services to the rest of us — use “complexity” (the word) as an imprecation.  This has its effects, and I'm worried that too much dependence on the metaphors of biology and complexity will backfire. Or, on the other hand, we could work really hard at capturing the joy and benefits of “complexity” (the word) for ourselves.  But we have an uphill fight in front of us.

Let me illustrate.  I recently listened to a debate among Craig Mundie, Halsey Minor, and Larry Babbio at the August 2004 PFF summit on the Future of the Internet.  In answer to a question about where opportunities lay in serving consumers vs. businesses in the future, Halsey M. talked about the fact that 90% of online traffic is now coming from consumers.  He said that there is a huge opportunity to make the internet more productive for businesses, and that we'll be retooling businesses to take advantage of the internet's possibilities.

Then he began to analogize [broadly paraphrasing]:

Just as the utility grid allowed centralization of complexity — and the industrial revolution allowed people not to have to worry about producing power any more — the internet allows businesses to consume centralized complex services from others.  Google centralizes the complexity of search.  This means that companies can stick to their core competencies.

Craig Mundie agreed with him, saying that the complexities of infrastructure and identity mechanisms would be overcome by a few very large companies, and adding that “Without this scale [of services being provided by large companies], businesses can't change and grow.”

These were eminently reasonable comments.  But they point to the “complexities” of identity management (a subject focused on particularly by Mundie) being handled by a few very big players, and, generally, to “complexities” created by the internet being smoothed for the consumption of businesses and consumers.

Complexity = bad for business.  That's the lexicographical development.

Complexity involves, by its nature, lots of choices, and a good deal of confusion if you are trying to predict the future.  But if you look at a complex system from the outside, you see emergent order – like the miracle of food delivery happening every day in the city of New York.

If identity management is something that is too “complex” for individual businesses to handle — much less individuals — and the perception is that it must be given over to a few large entities in order for businesses to function, we're heading in the wrong direction.  We need to take back the language of complexity — or perhaps come up with another word.

The Copyright Office redraft [pdf] of S. 2560 got a lot of attention this week.  Essentially, the redraft suggests that if you do a single overt thing (raise an eyebrow? build a microchip?) that could (a) be “reasonably expected” to (i) cause an infringement or (ii) persuade someone to infringe, you're liable for direct infringement yourself. 

“Overt acts” aren't defined, but there are specific things listed that aren't overt acts — like providing phone service to a “distributor of dissemination technology.” Thanks.

The last time I wrote about the Induce Act, I suggested that we shouldn't jump up and down and attack it — that Congress should hold a hearing and get people to explain why the bill would or wouldn't work.  That hearing happened in late July.  Most of the people testifying, save for the Copyright Office and the RIAA, said that the bill shouldn't go further.

Now the Copyright Office has circulated a redraft that, if anything, is likely to get people even more worried than they were about the original S.2560.  It doesn't seem like a move towards compromise.  If anything, it signals a hardening of position:  any technology that makes infringement possible (not just KaZaa or Grokster) can be reached under the draft, and the Sony/Betamax rule is dead. 

Sony is dead because the draft says that one possible “overt act” could be distributing a technology ”that, when used as intended, automatically causes the user of the technology to infringe copyrighted works without the user making a specific, informed decision, for each copyrighted work at issue, about whether to engage in such infringement.”  This boils down to:  if you build a technology that makes it possible to distribute works publicly (broadband access? mp3 players that connect to the internet? PCs?) you're liable.  Even if you don't meet current judicial standards for contributory or vicarious liability.

What's remarkable here is that the content industry feels that Grokster has given it an opening to broaden copyright liability beyond recognition.  Why haven't the comments of the many other businesses and advocates that opposed S.2560 been listened to?  Why are people stuck negotiating a special-purpose bill that doesn't seem to be special-purpose at all – but, instead, seems to take on anything that might be used by an infringer?  Why is the Copyright Office (clearly not neutral on this subject) holding the pen?

I'm sure there will be many meetings about this draft, and I'm confident that reason will prevail.  I'm sure this redraft isn't the last word.  So I'm not jumping up and down.  I'm just amazed at this sequence of events.

Today is the 35th anniversary of the day the first node in the ARPANET (at UCLA) sent a message to another computer.  Next month will be the anniversary of the day that first node talked to the second node. 

And what was the first message sent from one node to the other?  It wasn't “was hath God wrought.”  It was one computer saying LOG and the second receiving LOG and adding IN — so they'd have LOGIN.

In Leonard Kleinrock's words:

We sent an L; – did you get the L;? YEP

We sent an O; – did you get the O;? YEP

We sent a G; – did you get the G;? CRASH!

That was the first message on the internet.

Mitchell Waldrop's The Dream Machine, in a section titled “ARPA's Woodstock,” has Doug Engelbart saying the following on December 9, 1968: 

'The research program that I'm going to describe to you,' he began in that soft, strangely compelling baritone, 'is quickly characterizable by saying, “If, in your office, you as an intellectual worker were supplied with a computer display backed up by a computer that was alive for you all day, and that was instantly responsive to every action you had, how much value could you derive from that?”'

Happy many anniversaries.

Over at Terranova, Nathan Combs recently pointed to a Popular Science article suggesting that very few science fiction writers are brave enough to write about the near future:

Only a small cadre of technoprophets is attempting to extrapolate current trends and imagine what our world might look like in the next few decades. “We’re staring into a fogbank,” Stross says, “and we literally do not know where we’re going, only that we’re going there very fast.”

The Singularity — “the moment when the world is as different from today's world as this one is from the Stone Age,” as the article puts it – is approaching, and it's foreshadowed in stupendous advances in knowledge of the brain, biology, networks, and genetics.  All of these things interconnect, and if you're feeling enthusiastic about all this you can lose the attention of your readers very quickly.

But stay with me, just for one more summer evening.

How close are we to a posthuman (or more-than-human) era, in which most people recognize that the line between “living” and “inanimate” objects is no longer real?  Can we envision what's possible once minds can be uploaded and shared? What will social life be like when “living online” becomes real — and not just a turn of phrase?

Combs and the Terranova commentators look at these questions in light of the possibilities of virtual worlds, and whether they should be “games” (with heroic possibilities) or “simulations” (that are accessible to our current social physics minds).  The policy treatment of posthuman life is even more interesting.  Virtual worlds allow us to simulate these problems, and that's why State of Play II is essential.

But the imagination of writers (and law professors!) doesn't depend on what's viable in virtual worlds. How's your imagination doing?  What will your world be like in twenty years? What will “unlawful surveillance” mean then?  Will you know people differently? Will you feel allegiance to an online guild? Will you ever be alone? Will you be wiser? Will you live forever?

And what if the machine stops?