Web 2.02 (bottom up)

Peter Merholz has a response up to my Web 2.0 piece. Peter is one of the sharpest commentators and observers of the internet around and par the course it well worth reading. And he brings up two points that need some clarification, so as Slick Rick would say, here we go:

First off Peter is entirely right that the early web was not a place where "anyone" could build. Not in the least. But I also was pretty careful not to make that claim, so I'm not sure where the "anyone" Peter puts in quotes comes from.

The distinction between and amateur and professional is like that between the ocean and land, very clear most of the time, but almost impossible to pin down at the border regions. And when you talk about amateur and professional skills rather than amateurs and professionals as in people, well then the distinction becomes almost impossible to sort out. But in the end I think it's pretty damn clear that it takes a lot more skills to be rolling your own up to general standards websites now then it did 10 years ago. You just can't learn how to make Ajax sites or database driven sites or write good CSS the way you used to be able to learn HTML or Flash 3. Precisely demarcating that difference is near impossible, but it's pretty clear that it exists. The somewhat arbitrary and perhaps a bit silly distinction I used, of a skill that a reasonably intelligent and motivated person could learn in a weekend was there precisely to make it clear that the amateur web was not one that "anyone" could get onto, but was one that took a very different sort of learning process then what exists today to become a creator.

Peter's second point is well taken and I'm afraid I'm a bit at fault. I never meant to imply in any way that Peter was intentionally arguing that companies should relinquish control over to his company. However whether he likes it or not I do believe that is part of what he in effect did end up arguing. I mean, the article appeared on the website of the company he founded, a company in the business of selling web consulting. And when he says companies should relinquish control he's not saying they should have a gang of monkey's generate their websites or hire 14 year old "script kiddies" to write their code or turn your whole ecommerce site into a wiki. Relinquishing control is not something that you can just do like its a Nike ad. Rather in order to do it, you need to make sure you do it right. And if you want to do it right, hiring Peter and Adaptive Path is probably one of the smartest things you could do. They are among the very best, I have a strange feeling they'll do a much better job figuring out how to relinquish control then you, or most other companies could do on their own.

There is a reason for this, "relinquishing control" is hard, really hard. And not just psychologically, there are an awful lot of ways you can do it wrong. There is a reason why Amazon lets you add comments to book pages, but not edit the author and title of the book or use the page a private bulletin board, and its not because they hate their customers. Flickr lets you upload photos, but not mp3s or java applets. Ebay lets you sell your items online but requires you register with them. None of these businesses would work if they just let customers do anything and everything. They aren't in the lose control business, they are in the business of facilitating the flow of information. Not just any information too, but specific information, quality specific information, information relevant to their particular focuses.

When Amazon opened up its pages to comments they radically increased the amount of information available about each book purchase. In the process they relinquished some control over to their customers, in a rather controlled manner of course. Flickr gives their users control over their own photos online, but the numerous interface innovations that in part drives their success stem from controlling exactly what type of files the users can post. By narrowing the channel of information down to a couple similar image file types, the Flickr team was able to open up a whole array of ways in which that particular type of information could flow.

It is important to understand that openness and control do not necessarily need to be in conflict, they are not paradoxical at all, but in fact often work together integrally. It is only in very localized circumstances, for instance in the specific decision whether to have an API to a system or not, that the two enter into a dialectical relationship. Most of the time the two coexist quite easily, often complementing each other, and sometime quite essential to each others operation. For instance the distributed network that is the entire internet, would be close to useless without the centralized DNS system, which dictates the address on the network of practically every publicly accessible object on the internet.

My favorite example is still Brian Eno's Music for Airports. On this record Eno set out to create a generative system for music, a way to create music without the rigid control proscribed by western (and most other) music tradition. But to give up control completely is to give up being music at all. Even John Cage, whose 4'33" opens the entire piece up to the audience to create, relies upon the piece being done in a controlled environment. Outside of the performance hall, absent a performer on stage to provide a focal point, the piece no longer is music, is no longer recognizable. Eno, went far beyond this, he carefully curated the sounds going into his record. He went through an elaborate and convoluted process to create longs loops of sound out of rhythmic sync with each other. He hijacked the entire studio space to make the mechanics possible. He gave up control over certain key elements of the piece, the time when any given sound would play, and opened up a vast potential space for variation in the piece, but in order to achieve that liberation, he needed to control most of the process.

From a creators point of view it might be helpful to think not of control, but of self-discipline. Mike Migusrki has a piece doing exactly that, and its quite insightful. As a creator, in order to achieve the freedom to create what you envision in your head you need to achieve a certain mastery of your discipline. Only once you have achieved a certain control over your tools are your free to create what you want. Translated into a networked environment this transforms into a slightly different discipline. Suddenly the tools are shared, in order for information to flow from site to site, system to system a shared discipline must be developed and maintained. This discipline then becomes both a potential means to achieve a freedom and a potential for control to be implemented.

I wrote most of the above last week, before Katrina and its aftereffects disrupted all thought patterns. Since then the Web 2.0 conversation has advanced a bit, most notably with danah boyd's "Why Web2.0 Matters: Preparing for Glocalization". Have a feeling there is plenty more to come too. But overall I have the feeling there isn't really much disagreement. Expansion yes, Web 2.0 is a pretty amorphous thing, but there is something there and everyone wants to finger it. But perhaps the real answer is that old stand by, "all of the above". Or perhaps not, I'm looking forward to what comes next...