Mike Arrington's new stream

Mike Arrington's new stream.

Most of you know Mike as the editor of TechCrunch,
a phenomenal new weblog that tells the story of new web services,
written from the heart of SiliconValley. I met him for the first time
in a meeting in NY, in May, when we got together to talk about the sale
of weblogs.com.

Today Mike starts a new blog, called CrunchNotes
by telling the story of how TechCrunch got started. He credits me with
being an inspiration, and if it's really true, he's taken it so far and
done it so well, all the credit belongs to him. Really.

What
he says is so true. Too many software developers wander into the market
without knowing what's been tried before, what worked, what didn't.
Often the users know more about the history of the category than the
designer of the software. What Mike does, by writing up every product
and service that he sees, is the beginning of a process that we must
develop; but is itself a revisit of something that used to be done
thoroughly and systematically, but because of the quick pace of boom
and bust in the tech business, is an art that now needs to be
reinvented, a bootup that's actually a reboot.

Mike is a
lawyer. Laws have precedents. When it was thought that Harriet Miers
believed in a constitutional right to privacy, many inferred that she
was a choice advocate. One position implies another because legal
decisions are based on previous decisions. Sometimes a higher court
changes direction and overturns a precedent, this is necessary because
the context changes over time. It's true in technology too. In 1985 we
designed software to run in 640K; today my machine had more than 500
times that amount of memory, but in order to make it run adequately, I
had to double it. Laws that made sense in 1985 clearly need another
look in 2005.

In software, I call this system of precedents design by prior art.
To really make it work, we need to go back and scour the past for lost
art. It's not enough to just chronicle what's coming online now
(although I'm glad we're doing that). Maybe the next step is for Mike
and I to visit Michael Miller, who, as editor-in-chief of InfoWorld and
then PC Mag (where he still is, I believe) put in place a system for
looking at software over time. That was a system of prior art in
software.

There's another reason it's a good idea to
study the past in software — anything that was designed or implemented
before software patents forms a prior art defense against the patent
system of the 21st century. Mike, being both a lawyer and a student of
technology, is in a great position to lead us here.

In any
case, congratulations to my friend Mike, for doing this work so well.
It's great when someone so talented and motivated finds something that
suits him so well. How lucky for him, but then we're lucky too, because
we get the full benefit of his brilliance, without having to do the
work! “;->  [Scripting News]

Leave a comment