this from the bbc.

http://news.bbc.co.uk/hi/english/sci/tech/newsid_363000/363358.stm

By BBC News Online Science Editor Dr David Whitehouse

A tremendous explosion took place on the surface of the Sun last Tuesday and for a few very nervous hours astronomers did not know whether it was heading for Earth.

The blast threw a jet of superheated plasma carrying magnetic energy into space at speeds of 1,000 kilometres per second (600 miles per second).

However, using the speed of the Internet, astronomers around the world rapidly compared images and decided that a worldwide alert was unnecessary.

"Planet-buster"

The Solar and Heliospheric Observer (SOHO) satellite observed the solar explosion, which astronomers call a coronal mass ejection (CME).

The explosive event was "a real planet-buster", according to Dr Richard Fisher of Nasa's Goddard Space Flight Centre.

If the magnetic energy within the cloud of superhot gas had interacted with the Earth's magnetic field it would have sparked spectacular aurora at polar latitudes.

But more worryingly it could also induce power blackouts, block radio communications and trigger phantom commands capable of sending satellites spinning out of their proper orbits.

Cellular phones, global positioning signals and space-walking astronauts were all at risk. Hit or miss?

"When the coronal mass ejection was observed we were not sure whether the mass ejection was moving toward the Earth or directly away from the Earth" said Paal Brekke, SOHO Deputy Project Scientist.

Astronomers were particularly concerned that the event was followed by an increase in the flux of sub-atomic particles from the Sun.

So the scientists quickly downloaded Internet images of the Sun taken by observatories in the USA, Austria, Australia, and Japan. They then compared images the taken before and after the event.

"Because the data are so distributed and so accessible we were able to identify and track this event," said one astronomer. "Even just a few years ago, this kind of instant international collaboration would have been impossible."

Fortunately, it was soon established that the CME was headed directly away from the Earth - this time.

Preliminary analysis by Dr Simon Plunkett, of the Naval Research Laboratory in the United States, shows that if the CME were travelling towards the Earth, it would have arrived in just two and a half days.

The other Y2K problem

Solar activity waxes and wanes in an 11-year cycle, which is expected to peak sometime early next year.

Astronomers point out that the solar menace comes at the same time as computers around the world could struggle to cope with problems caused by the Millennium or Year 2000 (Y2K) bug.

Some solar physicists have called the effects from the Sun "the other Y2K problem".

"The SOHO satellite will play a key role in early detection of solar storms, which is important for issuing warnings," added Dr Brekke.

-- Anonymous, June 11, 1999

Answers

A beuatifull example of Murphy's law

-- Anonymous, June 11, 1999

CME -> CMF

Common Mode Failure

:-0

-- Anonymous, June 11, 1999

An excerpt from: Common-Mode-Failure in Information Systems, Part 1 by Christopher Davis, Pennsylvania State University http://www.macscitech.org/oswars/oswars13.html

The practice of establishing standards has been part of the human endeavor, no doubt, since the start of civilization itself. Throughout history, various standards have played key roles in the progression of society. As times and situations changed, so did the standards by which society lived. In the case of Information Systems (IS), many standards have come and gone in the relatively brief history of the industry. As the industry is continuously considering new and current standards, how it will continue to develop future standards? Are the people that shape and direct this process considering all of the aspects of their work? This is the first of three articles to deal with the development of sound and productive IS standards with respect to technical computing.

Common-Mode-Failure

Common-Mode-Failure (CMF), what is it and why should we be concerned about it when developing IS standards? Let's start with the following definition, which was issued with respect to the design of advanced power plant control systems.

Common-mode failures are casually related failures of redundant or separate equipment. CMF embraces all casual relationships, including severe environments, design errors, calibration and maintenance errors, hardware and software, operating systems, supporting systems, and consequential failures.

However, a better to way to define common-mode-failure is by example.

A few years back, Intel was chastised for selling chips that contained a certain math bug. Fortunately for the company, the chip was not in popular use at the time. When this bug was announced, the company I was working for immediately announced that all engineering safety and analysis work being done on Pentium-based machines was to stop until the chips could be checked for the bug. For those machines with the bug, some analyses had to be re-worked on UNIX-based machines, using an older version of the code. It seems that the UNIX code had not been upgraded, as the analysis group wanted to standardize on a single platform, the Pentium-Wintel platform. But this code still worked, and good thing it was still available!

Another example comes from a Win95 bug that was reported to automatically erase communications software from various online services (e.g. AOL, Compuserve, etc.). Suppose that your company standardized on Win95, but relied on your America Online (AOL) or Compuserve software to handle normal business e-mail, file transfers, or even information searches as part of its customer support efforts. How would you support your customers? What costs would you incur as a result of losing this software? More importantly, what would happen if you've standardized on Win95 and Microsoft allowed some other type of bug to ship with your next upgrade?

With multiple platforms available in a business environment, it is unlikely that even the smartest of hackers could develop one single bug that could wipe out every platform type (except Y2K --aj). Even if a company has problems with their OS (e.g. Microsoft), other machines running different OSs would be immune to such problems. Hardware common-mode-failures are less like to propagate across platforms. So the question I have for proponents of single-platform IS strategies is, "Are you willing to gamble your company's entire information systems structure on one platform?"

--- end quote ---

It's to bad we didn't have an enforced sensible international standard for time and date formats thirty years ago.

Y2K is the biggest Common-Mode-Failure the modern world has ever experienced, all because of human nature's inability to plan more than ten years ahead.

It will be interesting to read some of the books that will be written about it after the fact.

--aj

-- Anonymous, June 11, 1999