Andrew Odlyzko's "Content is not King" finally reminded me that the critical difference in revenue is the change from broadcast (publication) to interaction. Not just interaction between a broadcaster and the audience (talk radio, letters to the editor) but a wholesale change from that dichotomy to everyone having the opportunity to be both or either. His example:

The primacy of connectivity over content explains phenomena that have baffled wireless industry observers, such as the enthusiastic embrace of SMS (Short Message System) and the tepid reception of WAP (Wireless Application Protocol). Combined with statistics showing low cell phone usage, this also suggests that the 3G systems that are about to be introduced will serve primarily to stimulate more voice usage, not to provide Internet access.

I disagree with the second conclusion, data connectivity will be an important part of the result.

There's a lot of optimism in the closed proprietary mind, this discussion overlooks the obvious and plumps for the conclusion:

Femtocells may look unlikely, but there's a possibility they may win out.

Yeah, right.

While I'm no longer naive enough to believe it will be one or the other, the idea that a femtocell, which is the tool of the operator, so you'll need one for each proprietary service operator (or change it when you change providers), which you probably therefore aren't willing to pay for, that will use your connectivity (and be prone to its weaknesses) is going to win a major portion of the market where the alternative gives you much greater freedom seems unlikely.

A Wi-Fi access point gives you infinitely greater choice of client devices, services, service providers, that could well outweigh the small performance advantage of limited handsets, limited services, single provider proprietary solutions. And remember, either solution uses the same backhaul.

I don't know the details, but there'd be a lovely irony if your femtocell used VoIP to backhaul the traffic to where you can be charged for it... If the argument that the cellular technology works more reliably from the handset to the femtocell, then why use it instead of DECT or any of the other cordless technologies that we know? There's already a Skype handset that does this.

Makes for a simpler phone? Perhaps, but sales figures for dual-mode handsets (Wi-Fi/Cellular) don't indicate there's any barrier to their uptake.

"Vodafone has no desire to subsidise a Wi-Fi handset"

What vodafone, and the rest of the comfortable incumbents will learn it's not what they desire, but what the customer wants, that occurs in non-monopoly environments. When you read the list of advantages of femtocells, note how many are advantages to the operator, rather than the customer.

The dead giveaway:

But there's an even more powerful business reason why mobile operators want to sell femtocells: they hate Wi-Fi, because users own it and can use it at will.

Not to disparage those operators who are going with UMA (AKA GAN) services like T-Mobile and Orange.

Even in duopolies and oligopolies, there tends to be one less powerful who will seize the chance to change the rules.

Most emperors have a few tailors turn up to sell them new clothes, few as transparently a bad idea as this one.

My basic tenet is that decentralisation and collaboration of small pieces loosely joined is a model that exceeds in many ways the hierarchical monolith of centralisation with its architectural tension between the center and the edge.

The fragmentation of the monolith into smaller pieces is a trend which continues and the latest example I have is the widget. Once standardised, this will allow you to assemble any number of functionalities in interesting ways. Standardisation is critical to extracting the full value of this technique, until then it remains a scattering of smaller or larger ponds in which every boat has to be rebuilt to float in another. We need the ocean.

This is the bite-sizing of the web, extending the power RSS gave us, and beefing up the IFRAME capability of HTML (standardised) and turning web top construction into a drag and drop task.

But that's not what I set out to write about. Currently I'm visiting a few new sites that I might want to use further, Geni.com for example. And every time I do, its the registration, profile population and configuration rigmarole all over again.

Then I read this. Could 2007 be the year of social network fatigue?

It sure is for me,

The number of sites I've been to recently that offer me one service I want, then lard themselves up with a whole lot "Web 2.0 Community" functionality that I already get somewhere else.

Building a community can happen external to almost any service, and if you do want a little more by way of profile from me, let me give it to you from another site. Now that other site is not likely to be MySpace, there have been some hints recently that they are contemplating isolation and freezing out some services that make them more desirable, duh! But from a site I control.

Now this must be sounding OpenID-esque, (not without critics) about which I know very little despite guru Doc Searls being heavily involved. I tried Sxipper, I hated it, which probably means I don't understand it well enough, but it was intrusive, gigantic and not much help.

When I go to a new site, if I register, I want to give them a password, a URL, and have them suck my favourite movies, books, etc. etc. into their application (and sync it if I change anything). Then in the small pieces loosely joined widgety world, I can hook together the various parts of my presence without repeating myself to every web site I meet from Amazon to Weebly.

Something at the data/information rather than functionality, operating in an X11-like "I am the server" mode and you are my client. Resonates with Doc's VRM musings well.

Well, enough quotes from speakers on "What Do We know?"

The answer appears akin to the Hitchhikers Guide To The Galaxies explanation of the population of the Universe.

But Nissam Taleb's book, stumbled over on The Long Tail blog, has raised a lot of thoughts.

The coincidental stumbling on "The Luck Factor" by Dr Richard Wiseman via a podcast Radio New Zealand interview couldn't be more complimentary.

Taleb's book brings together many of the usual suspects, and his observations about the inevitability of emotion, and the suggestion that its purpose is to short circuit rationality in order to reduce the burden of thinking about everything, and thus its an essential, resonates with Marvin Minsky's new book (very little of which I have read) but whose title mentions emotion as part of the discussion of AI.

Coincidentally I'm still reading, albeit very slowly, despite skipping over the self-dialogues of Kurzweil with various characters, "The Singularity Is Near." Its a discussion, one way or another of AI, particularly the "Strong AI" variant which engenders so much resistance there is an entire section, in which I am mired, that discusses the criticisms and seeks to refute them.

Taleb's premise is that the human mind has evolved with very simple purposes in mind, to be achieved in a very simple environment, one in which we no longer find ourselves. Indeed, perhaps for our evolved legacy skills, the Singularity has already arrived, and while we romantically expect certain competencies and behaviours, we find ourselves, collectively and individually, failing to live up to those expectations.

Another day, another list of atrocities committed singly or as collectives.

As for luck, to be alive is to have beaten extraordinary odds, not just the uncertainty of the environment with all its perils, but to have eluded our own species destructiveness. That the human family tree is shallow and that within relatively few generations we can find links between all of us living isn't as odd as it appears (a survivorship bias?) given our prediliction for extermination...

The other book currently being read, "Dark Age Ahead" by Jane Jacobs, fills me with dread.

Artificial Intelligence may be one that is better able to cope with randomness, more mechanical and deterministic, while aware as the billiard ball example in "Randomness" illustrates, determining anything of even moderate complexity requires collossal mega-exponential knowledge to determine.