SCO.com and Caldera.com dead

[James Knott]

Henry Spencer wrote:
> On Sun, 7 Sep 2003, James Knott wrote:
> 
>>...Tube computers were also extremely unreliable (I used to 
>>maintain & repair one in the Toronto Stock Exchange.) and not the kind 
>>of thing you'd want to trust someone's life to.
> 
> 
> How about 250 million lives? :-)  Tube computers could be made extremely
> reliable, as witness the tube computers that controlled North America's
> air defences for many years. 
> 
> As ENIAC showed, running the tubes at much less than their rated
> electrical limits, and leaving the system on 24 hours a day so that the
> tube filaments were never stressed by cooling down and heating up again,
> made a huge difference.  The reliability still wasn't wonderful, with mean
> times between failure measured in hours rather than days... but with
> multiply-redundant systems and automatic switchover to a hot spare, a
> failure of one computer didn't mean an interruption in service. 

I'm well aware of those issues.  At the stock exchange, one of my duties
was to slowly crank up the filament voltage every morning and then wait
a few minutes, before bringing up the +- 130V DC motor generator sets to
get it going.  One comment I read about that ancient air traffic control
system, was that the only source of tubes for it, was the communist
countries (this was before the colapse of communism).  Are vacuum tubes
(other than specialized ones) even made at all in North America these
days?  About the only vacuum tubes in common use, would be the CRT in
most TVs & monitors and the magnetrons in microwave ovens.

No matter what you do, vacuum tubes != reliability.  Also, according to
a book I read about ENIAC a while ago, they'd run each job more than
once, to make sure they got the correct results.



--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml