[IWAR] Y2K embedded systems

From: 7Pillars Partners (partnersat_private)
Date: Mon Jan 19 1998 - 10:23:09 PST

  • Next message: 7Pillars Partners: "[IWAR] CHINA military center"

    New Y2K Threat: Embedded Systems
       by Scott Kirsner 
       2:00am  19.Jan.98.PST
       With less than two years left until the world's computers enter the new
       millennium, Peter de Jager is losing sleep.
       
       The world's leading consciousness-raiser for the year-2000 problem is
       worried about a less-visible aspect of the date-processing errors
       expected to plague computers as they tick past 31 December 1999. While
       much attention has been focused on fixing older mainframe computers and
       even PCs, de Jager says that embedded systems - the built-in,
       preprogrammed chips that govern everything from microwave ovens to
       nuclear reactors - have been overlooked.
       
       De Jager, a Canadian computer consultant and author who is a celebrity
       on the year 2000 speaking circuit, is not one to sit idle. On 31
       December, he announced Project Damocles, an effort to get embedded
       systems manufacturers to own up to Y2K bugs that might affect their
       products, from electric razors to heart-lung machines.
       
       "Why did I take charge?" de Jager asks. "No one was doing anything. The
       governments of the world don't take this problem seriously," he
       complains - while others complain about his solution.
       
       Project Damocles borrows its name from a Greek legend about a courtier
       given the king's seat for a banquet - with a sword hanging over it from
       a single hair, demonstrating the perilous side of power. The project
       asks that anyone who knows of products that might experience
       date-related failures to report them to de Jager or his Web site.
       
       Then de Jager will delete the sender's name and forward the report to
       the manufacturer's legal department via registered mail. If the
       manufacturer doesn't solve the problem and failures do occur, Project
       Damocles will make its copy of the report available to attorneys,
       helping them to prove negligence on the part of the manufacturer.
       
       "My objective is to force manufacturers to do what is right by hanging
    a
       double-edged sword over their heads," wrote de Jager, in an email that
       went out to the 30,000 subscribers on his year 2000 mailing list. "A
       sword which swings into action after they allow a failure to occur."
       
       The euphemistic "failures" that could occur might better be termed
       catastrophes, according to many in the emerging year 2000 software and
       services industry. Embedded systems can be found in all types of power
       plants, water and sewage systems, many of the devices used in
    hospitals,
       military equipment, aircraft, oil tankers, alarm systems, and
    elevators.
       
       "This is potentially the most destructive part of the year 2000
       problem," says Dr. Leon Kappelman, a professor at the University of
       North Texas and co-chair of the Society for Information Management's
       Year 2000 Working Group. "This isn't the inconvenience part where your
       paycheck comes a few days late. This is the blood-in-the-streets part."
       
       And compared to Y2K bugs on mainframe computers, embedded systems are
       tougher to fix. Patches or workarounds aren't effective; in most cases,
       the systems must be completely replaced.
       
       "Year 2000 has been perceived incorrectly as primarily a mainframe
       thing," says Capers Jones, chairman of Software Productivity Research
       and another leading year 2000 expert. He has estimated the worldwide
       costs of the computer crisis could be as high as US$3.6 trillion. "But
       the embedded stuff is going to be a mess, and it hasn't received enough
       attention. It wouldn't hurt to expose some of the manufacturers,
       especially those whose products can kill us, like medical equipment
       companies," he added, saying Project Damocles looks like a step in the
       right direction.
       
       Kappelman adds that only about 10 percent of the four billion chips
       manufactured in 1996 went into PCs and other devices typically thought
       of as computers. The rest reside in embedded systems. "But because so
       many of the embedded systems problems put lives at risk, there's a
    great
       reluctance to publicly share information about them," he adds.
    "Whatever
       can be disclosed through forums like Peter's is good, because the
    public
       needs to know how much is really at stake."
       
       But others question de Jager's methodology.
       
       "I'm not sure why one individual having any proprietary information is
       necessary to solve the problem," says Harris Miller, president of the
       Washington, DC-based Information Technology Association of America, an
       industry trade group. "It won't get solved by keeping secrets - it's
       about being as open and collaborative as possible." The ITAA has had a
       product questionnaire on its Web site for several months aimed at
       helping vendors and customers jointly resolve embedded systems issues.
       
       Jim Lott, a product manager at Dallas Semiconductor, acknowledged the
       value of efforts like Project Damocles to raise awareness, but still
       voiced reservations. "Rather than blowing the whistle on people, we
    need
       to be working together to find solutions," said Lott, who adds that his
       company is already shipping products that will work in a post-1999
       world. "This opens a real door for fear-mongering."
       
       Year 2000 legal expert Steven Hock, president of Triaxsys Research,
    adds
       that while de Jager's efforts are well-intentioned, they may cause more
       problems than they solve - both for de Jager and the companies he's
       targeting. "De Jager faces legal risks in soliciting information about
       potentially serious problems, and then sharing that information only
       with the legal department of the company," Hock says. "Why wouldn't he
       pass it on to regulatory agencies who should know about such problems?"
       
       Then there's the issue of bug reports without merit.
       
       "The detriment of Project Damocles is that they're going to get emails
       from many, many people who perceive problems that don't exist," says
       Hock. "Companies will spend untold hours and dollars dealing with
       whistle-blowers whose complaints aren't really legitimate. It'll lead
    to
       more wasted efforts than real results, I think."
       
       But Xavier Roy, the vice president at Litton Enterprise Solutions' year
       2000 business unit, says some effort focused on embedded systems is
       better than nothing. He should know: His clients include hospitals and
       power companies that run nuclear plants.
       
       "Embedded systems haven't gotten the attention they deserve," says Roy.
       "Just the name - Damocles - indicates there is a threat. This is the
       only way we can make people realize the severity of the problem."
    
    Copyright  1993-97 Wired Ventures Inc. and affiliated companies.
       All rights reserved.
    



    This archive was generated by hypermail 2b30 : Fri Apr 13 2001 - 13:01:17 PDT