When I was coming up here, I became very nervous. I was worried that I was going to end up as the token Yank speaker, expected to defend cultural Imperialism in the global electronic world as represented by the burgeoning net presence of The New York Times.
I was, therefore, quite pleased to learn that I am actually the token young geek speaker. This is a position I am far more comfortable being in. I only hope that you're comfortable with and prepared for a few minutes of hard core geek speak.
I'm grateful to several delegates for already addressing and struggling with the issues of personal responsibility, personal involvement and personal accountability as they apply to the Net. In part, because I really believe these issues are at the heart of what this conference should be discussing in terms of citizenship an Electronic Village.
It also means I don't feel compelled to address those issues right away.
Instead, I want to talk about the ways in which we relate to and value data; in short, the way we make information. I am convinced that, if we're going to take advantage of this new, interactive, global, immediate electronic media we call the Net, we must consider how it structures the way we relate to and value data and how that dynamic will consequently influence our interaction with individuals and institutions outside the electronic world.
The manner in which we are beginning to relate to data on the Net is most obvious, I believe, in looking at the changing nature of Internet search engines. For those of you who aren't familiar with them, Internet search engines are applications which you use to find data on the Net.
As the Net grew, and the wealth of data on the Net also grew, a great many people recognized the need for tools which would allow us to better navigate through all this data and find that which we need; that data which, for us, could become information.
Most of us who have been on the Net for any length of time have had the experience of being unable to find data which we know is available. It's out there, we just don't know where it is.
As a result, it should be of no surprise to anybody that the few people who have found a way to make money off the Net are those making tools to allow people to better navigate [and] access data, both in the forms of software and perhaps, more tellingly, books.
The earliest commercially available search engines were all keyword-driven. This means that you typed in a word or several words and the engine looked for these words on other sites in a variety of different ways, either by parsing and indexing entire sites, or by creating data base files.
This sort of searching resulted in some pretty amusing results. You got back some interesting documents and some interesting data based on what you were looking for, but didn't really reliably direct us to the data of most interest. It did not direct us to that data which could become information. At least not all the time.
There are still quite a few keyword-driven engines out there. There are also a bunch of second-generation search engines, which I would describe as context-based. With these, you type a few words and you indicate the contextual relationship between those words.
These engines have been far more useful in terms of delivering valued data, because they recognize that certain data is inherently more valuable than other. But they still fall short.
The reason both types of searching tools fall short, I believe, is because they rely on outmoded ways of relating to and valuing data. Both use what I would consider typographic data models.
In the case of a keyword search, the value is being placed on one word alone. This is inherently incompatible with the nature of the data structure of the Net. You can't value just one word.
In the case of a context search, the value is being placed on just the contextual relevance of words. This also is inherently incompatible with the nature of the data structure of the Net. It's inherently too limiting for a three dimensional data structure. There is too much data out there to search with such limited tools.
So, if the available search engines aren't up to the job, and the resident data structures they rely on are incompatible with the data structure of the Net, what is it we need?
The answer has already been provided by a bunch of folks in a bunch of different locations who are already hard at work developing conceptual search engines. These are applications which will search for data, based not on keywords or context but on the actual concept you are interested in. This is an incredibly significant development.
What I find so exciting about this work about the idea of a concept-based search engine is not the application itself. It's not the tool that will come. That's not true; a concept-based search engine is a pretty cool act. It's going to be killer. But what is I think of deeper interest is the value model being used by such an engine.
A concept-based search engine recognizes that the data architecture of the Net is three dimensional and dynamic. And it recognizes that working within such a data architecture, the value is not placed on the data itself, but on the linkage between data.
Of course, valuing linkage is nothing new. Linkage and depth of information have always been important. What is new is that this linkage is both dynamic and is directed by the user. This is revolutionary. We allow people to follow the links they feel are important or to follow multiple, parallel and potentially-contradictory links. We no longer get only the linkage deemed important by the author, the television producer or the film-maker.
Effectively, we create our own information.
Now, at this point some of you might be wondering what all this has to do with citizenship. Quite a lot, I think.
Remember, the Net allows not only human-to-data linkage or data-to-data linkage, it allows human-to-human linkage. And that human-to-human linkage becomes similarly valued. Working on the Net not only facilitates but actually inherently puts value on human interaction.
Given that we are valuing linkage, therefore, it is up to each of us individually to accept the responsibility for maintaining, protecting and aggressively pursuing linkage.
We must, each and every one of us, be the guardians of the Net. We must decide which linkages are not only of little value but are socially unacceptable. And then we must act to break those links whether it be at our own end, at the level of service provider or at the point of origin.
We must also demand that the media and government provide us with the sort of dense and relevant data without which linkage is pointless. And we must give government the resources and media outlets the freedom to meet that demand.
Moreover, we should remember that linkage is also important beyond the Net. Surfing is cool. But walking out onto the street and actually talking to somebody is pretty cool, too.
I want to close by briefly talking about the responsibility of the new electronic media all this.
Search engines and plain old net surfing will always turn up some linkage of high value. But it is really up to us the folks working my business and the folks working traditional broadcast and print media to be the primary providers of linkage.
All linkage is not of equal value. It is up to us to edit, critically and thoughtfully. The Net is a big, dark warehouse full of incredible stuff. Some of it scary, some of it brilliant, some of it trivial.
We may not be able to turn the lights on, but we should at least give people a flashlight and a map of the aisles.