JON POSTEL: The net's loss

by Peter Martin, Financial Times
copyright 1998, Financial Times

When Jon Postel died last week at 55, an internet era died with him. 
Postel was an important figure in the creation of the net.  But his
death implies more than just the loss of his talents. 

It also symbolises the transition from an internet managed by computer
scientists with an academic set of priorities to one reflecting
commercial and governmental pressures. 

Postel himself had been actively preparing for this transition.  He
knew his health was not good - the heart condition that eventually
killed him had been a source of concern for years - and was keen to
hand his own role in the internet's machinery to a solidly based new
body. 

What was that role? It is worth exploring in detail, because its very
obscurity reveals some important truths about the way the internet
works. 

Postel's full-time job was running the Computer Networks Division of
the Information Sciences Institute at the University of Southern
California in Los Angeles.  In other words, he was a computer science
middle manager - one of seven at the institute - with a staff of 70. 

His power over the internet came from another source entirely: his
role as operator of the Internet Assigned Numbers Authority (IANA). 

Any computer connected to the internet needs a unique identifying
address.  Although web sites, for example, are identified by names
such as www.ft.com, these are merely the human-readable proxies for
the underlying address, which takes the form of a 12-figure number. 
Some years ago, when the US government still provided the funding for
the internet, it authorised IANA to allocate those numbers, making it
the central node in the web of inter connections that bind the
internet together. 

Postel had another role, however, also shrouded in obscurity.  He was
the editor of the electronic documents known as RFCs or Requests for
Comments, which enshrine the internet's rules, or protocols.  In these
rules lie the network's power.  They set out the standards for
communication which allow every computer everywhere in the world to
communicate.  These hundreds of documents are the constitution of the
internet.  As RFC editor, Postel shaped, selected and maintained them. 
He also wrote some of the most important documents himself. 

The public image of the internet's founders is one of otherworldly
computer scientists: greying ponytails, arcane acronyms, convoluted
in-jokes. 

Much of this is true, as a glance at Postel's picture will confirm. 
But the rush to replace the structure the founders created with one
more attuned to commercial realities risks missing the point.  The
internet's ad-hoc, flexible, consensual structure offers powerful
lessons to everyone in business.  It is, in some ways, a prototype of
the way companies will have to operate in future. 

The internet works because computer scientists all over the world are
prepared to reach agreement on the best standards to adopt.  The
process of reaching that agreement is managed by a relatively small
number of people, of whom Postel was one.  Their power stems not from
official status or governmental nomination but from their ability to
create consensus. 

The consensus, in turn, stems from a shared purpose: the creation of a
network that allows easy communication.  This is in part a technical
vision, and it requires deep knowledge of computer science.  But it is
also an ideological one, requiring a humanistic commitment to freedom
of expression and to a medium of communication that rises above the
interests of government or commerce. 

The era of unfettered academic control is drawing to a close, as the
internet becomes part of the global central nervous system.  The
ad-hoc management process that has served it so well in the past needs
some form of reinforcement - if only to cope with the sad fact of
human mortality. 

The haggling over how the transition should proceed has been, at
times, acrimonious.  The debate pits the internet's founders -
including Postel - against a clutch of squabbling rivals including the
US government, the owners of big consumer brands, overseas
governments, companies already making profits from the internet, the
International Telecommunications Union and others.  Few of these
bodies see eye to eye.  None the less, in Postel's last weeks of life
a compromise seemed likely, based on his proposals for an
international, independent successor to IANA. 

Whether or not this compromise works, there is a broader lesson to be
drawn from Postel's achievements, and from the history of the
internet.  An ethic of collaboration and open discussion around a
common purpose is an extraordinarily powerful and creative force. 

In this instance, it has created a global communications system where
once there was Babel.  But there is a wider relevance.  This is a
style of working - within companies and between them, for example -
which is particularly suited to the modern era.  Knowledge workers,
spread around the world, cannot easily be herded.  But they can be
encouraged and inspired to work together within a common framework, a
process that can be both speedy and efficient. 

That is the lesson of the internet's early years.  The Postel era is
ending; but the principles of consensus, debate and agreement deserve
to live on. 

Peter Martin: peter.martin@FT.com