Popular Power

2000 - 2001
dead us
marc hedlund and nelson minar started popular power in january 2000 on a simple trade: install a java client, let it use your processor while you were not using it, get paid. the free half of the deal ran an influenza vaccine simulation and the paid half was supposed to sell those same cycles to companies. they raised about 1.6 million dollars from angels and could not close a venture round in the year the money stopped. the site was still normal on 1 march 2001 and carried a goodbye note by the first of april.

~ the obit ~

The idea of lending your computer to somebody else's math reached most people as SETI@home, which asked for nothing and gave back nothing except a screensaver and a place on a leaderboard. Popular Power wanted the same processor and offered to pay for it. The paying version is the one that did not survive.

Marc Hedlund and Nelson Minar founded it in San Francisco in January 2000. They had been working together for more than ten years by then, both out of Reed College, Minar by way of the MIT Media Lab. The product was the Popular Power Worker, a Java client you downloaded and left running, pulling jobs down from the network and sending results back while your machine sat idle. What separated it from the volunteer projects was the metaclient design: rather than one application tuned to one problem, the client accepted mobile code, so the network could be pointed at a new job without anybody reinstalling anything.

Minar's own estimate of what a member would earn was somewhere between one and five dollars a month, which he described as about half off your net bill if you were on dial-up. Members could set how much of their time went to nonprofit work and how much went to paying customers. The nonprofit half is the part that got written about: a computer model of how the human immune system reacts to candidate influenza vaccines, run across borrowed machines.

There is no personal angle to reach for here, because Popular Power lasted fourteen months and spent most of them in preview. It never got far enough out of San Francisco to be a thing you either used or missed. What it left is an argument, made early and made properly, about whether the idle half of a few million home computers was worth money to anyone.

~ the rap sheet ~

BornJanuary 2000 in San Francisco
Peaka Java client on volunteer machines running influenza vaccine models
KilledMarch 2001
Lifespanfourteen months
Made byMarc Hedlund and Nelson Minar
Killed bya venture round that did not close

~ the pitch ~

In 2000 the pitch had a name and the name was distributed computing, and real money went into it. Entropia and United Devices were selling variations of the same claim Popular Power was: the world's home computers are idle most of the time, and we will be the layer that sells that idleness. The technical half of the claim was correct. What none of them had was a set of customers who wanted compute delivered in that shape.

The problems that parallelize cleanly across unreliable volunteer machines are a narrow list: protein folding, signal analysis, key cracking, molecular simulation. The companies with those problems mostly had budget for real clusters and did not want their data sitting on strangers' hard drives. The companies without those problems had nothing to send. Popular Power's answer was to make the network general with mobile code so it could take whatever job showed up, and then nothing showed up in time.

~ the day it died ~

The archive holds the transition exactly. On 1 March 2001 popularpower.com was still the working site, with a member login and a paragraph inviting you to donate or sell your computer's idle processing time to problems previously thought insurmountable. By 1 April the whole address was one page with the logo across the top of it.

What replaced the site was signed by both founders. It said that Popular Power was no longer in business and that the nonprofit projects would keep running for as long as they were able. It thanked the people who had given time and money to the idea. It thanked the employees last and hardest. Underneath sat a link labeled original site, so you could go look at what had been there the month before.

They had raised roughly 1.6 million dollars from angel investors and could not close the venture round they needed. March 2001 was a bad month to be asking, with the Nasdaq down by more than half over twelve months and no appetite anywhere for a business whose revenue depended on reselling other people's spare capacity. Hedlund went on to found the personal finance site Wesabe and later ran engineering at Etsy and Stripe. Both men are credited on the opening chapter of O'Reilly's Peer-to-Peer, which came out in the same season the company closed.

~ leave a rose ~

an anonymous tribute. one rose per visitor per day.

@}-->-- ... roses left here

~ leave a wreath ~

a rose fades by morning. a wreath stays on this grave for good, with your name on it.

from $1, paid once through stripe. it keeps the graveyard running: the hosting and the nightly digging. adopt it and the headstone reads "maintained in memory by" you. the groundskeeper plants new wreaths at the next tending, usually within a day.

» one email a week: new graves and the anniversaries of the dead. subscribe

~ leave a note ~

visitors before you have left these graveside notes. anonymous welcome.

other software: T/Maker, iTunes Movie Trailers, Parse, Windows Mobile

also died in the 2000s: In2TV, Bildschirmtext (BTX), eMule Paradise

« previous  ·  random  ·  next »