Interval Research Corporation

~ the obituary ~

Distance from Interval Research is not a geographic problem. The lab was so deliberately quiet about what it worked on that even people a few miles away in Palo Alto were reduced to guessing, and when Salon wrote its closing piece the reporter had to say outright that we can only guess what the research lab produced in eight years of existence. No outsider got close to this one, and that was the design.

Paul Allen and David Liddle founded it in 1992. Allen put up the money, Liddle ran it as CEO, and the brief was a laboratory and technology incubator aimed at consumer products and services with an eye on the internet. The company's own 1997 web page described itself as a research setting seeking to define the issues, map out the concepts and create the technology that would be important in the future, which tells you the shape of the ambition and almost nothing about the work.

The ambition had a name attached to it from the start: a Xerox PARC for the twenty-first century. That is a very precise thing to want, and it carries its own hazard, because the original PARC is mostly remembered for inventions that made other companies rich. Interval drew serious talent and ran for eight years against the ten Allen had originally committed to.

It closed in April 2000. By Salon's account the announcement arrived as a press release on the front door of the building on a Friday, which puts it at 21 April. A small group stayed on as Interval Media to finish a few projects; that was wound up in June 2006.

~ the case file ~

~ the brief ~

A 1999 Wired piece, working from an Allen memo, reported that the funder wanted less R and more D. That sentence does a lot of work. A research lab with a ten-year horizon is being asked, seven years in, to start producing things that ship. Those two instructions describe different institutions, and the lab was being asked to be both at once.

Liddle left in the fall of 1999, which Salon connected to Allen's growing focus on the cable industry, citing a New York Times piece from that September. Those are reported explanations rather than confirmed ones and they should be read that way. The sequence itself is not disputed: the CEO went in the autumn, the lab went the following spring, and about thirty of the remaining employees were slated to move to a new broadband group serving Allen's cable ventures. Liddle went on to US Venture Partners.

~ eight of the ten ~

The clearest output was Purple Moon, the company Brenda Laurel built out of Interval to make computer games for girls, on the back of four years of interviews run before anyone designed a character. Purple Moon has its own page here. It went bankrupt, which is the part people cite when they want to score Interval's record. It is equally the case that the research worked there and the business did not.

Around a hundred million dollars went in. What came out is genuinely hard to total, partly because secrecy was policy and partly because the ideas that escaped did so through people rather than products. Judging a lab on its shipped units is the standard the funder eventually applied, and by that standard it failed; judging it on whether it changed what anyone later built is harder and nobody has done it properly. Eight years is not long enough to settle the question, which may be the only thing everyone involved would agree on.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »