Who knew I was so avant garde? As I understand Kottke's proposal, the
next step on the way to the WebOS is to run a web server on your
desktop so that you can get access to data on your local machine by way
of your browser and effectively erase the distinction between data out
on the web and data locally.
Back in 2001about when I started this blog, Scoble helped
me become a beta user of what shortly morphed into “Radio”. One thing
that attracted me to the product was that was precisely its
architecture. Browser access to an app that was a web server and data
store running locally. I'm writing this post in that environment right
now as I ride the train home from work. Since I've been living and
working off laptops and in various modes of mass transit since the
early 1990s this is an essential requirement. Think clients work. So
did Lotus Notes. But the architecture Dave Winer
dreamed up did too, although it's not always intuitively obvious,
especially to non-technical users. Watching the problems that many
users encountered (and still encounter) with Radio should be
instructive to anyone who wants to follow this path. At least in
today's environment, it pays to understand where your data is and how
it flows from place to place. Maybe someday it won't, but we aren't
there yet.
Since then, I've pursued a strategy of using open source tools to
replicate Winer's architecture for much of my routine knowledge work
efforts. I've put together a LAMP environment on my laptop running
Apache, MySQL, PHP, and Python. I can, and do, run a variety of open
source applications on top of this environment. I run WordPress,
several wikis, dotProject, trac, textpattern,
and others all locally.
Some of these are tools and products I am evaluating. More importantly
they host the primary tools I use for much of my knowledge work and
form the nucleus of my effort to explore and understand personal knowledge management.
For now, this is a mix of learning experiment and developing new
habits. One thing that it gives me is a degree of platform independence
coupled with an ability to work both connected and disconnected. For
now, the technology is a bit of a lash-up, but it allows me to explore
the behavioral issues. And those are what will ultimately drive
adoption of the technologies as they mature.
Jason Kottke has a lengthy and detailed proposal
for the platform builders to realize that the Web is the ultimate
platform, and to get on with building for that, rather than just for
their own private silos. When that's done, he says we'll have Web 3.0