Patriot problem, more info available?

Patriot problem, more info available?

[SAHARBAUGH]

In June 6th USA Today there is an Associated Press article titled
"Army knew of problems with Patriot".
The last paragraph says,"The Patriot computer screen did not
show an incoming Scud because the computer software could
not calculate quickly enough the missile's path.  This was
due to two factors not previously encountered simultaneously by 
the Patriot systems in Saudi Arabia: continuous operation
of the computer for four days prior to the moment of
attack, and a faster-than-usual Scud missile."
--
Can anyone furnish more details? I do not read
Aviation Week, maybe they report it in more depth.
I am particularly interested in the connection
between four days of operation and a software failure.
Based on no more than the newspaper article I wonder
if a dynamic memory got fragmented and slow with time.
I believe someone said on this BB that the Patriot
system is not programmed in Ada but there may be a
lesson-learned here for real-time Ada software.
(correction: thats "Aviation Week").
sam harbaugh saharbaugh@ROO.FIT.EDU        
-----------

Re: Patriot problem, more info available?

[Jim Showalter]

>I am particularly interested in the connection
>between four days of operation and a software failure.
>Based on no more than the newspaper article I wonder
>if a dynamic memory got fragmented and slow with time.

That's certainly how it reads. I've been told by people
involved with the project that the system is programmed in a
mixture of assembly, FORTRAN, and JOVIAL--but from the way it
failed it sounds more like LISP!

>there may be a
>lesson-learned here for real-time Ada software.

Indeed there is--avoid access types as much as possible.
-- 
**************** JIM SHOWALTER, jls@netcom.com, (408) 243-0630 ****************
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *

Re: Patriot problem, more info available?

[Orville R. Weyrich]

In article <1991Jun7.080445.12661@netcom.COM> jls@netcom.COM (Jim Showalter) writes:
>>I am particularly interested in the connection
>>between four days of operation and a software failure.
>>Based on no more than the newspaper article I wonder
>>if a dynamic memory got fragmented and slow with time.
>
>That's certainly how it reads. I've been told by people
>involved with the project that the system is programmed in a
>mixture of assembly, FORTRAN, and JOVIAL--but from the way it
>failed it sounds more like LISP!

Hmmm ... reminds me of an article that I saw describing a program which
converts LISP into FORTRAN.  It could be a LISP interpreter implemented
in FORTRAN :-).

--------------------------------------           ******************************
Orville R. Weyrich, Jr., Ph.D.                   Certified Systems Professional
Internet: orville%weyrich@uunet.uu.net             Weyrich Computer Consulting
Voice:    (602) 391-0821                         POB 5782, Scottsdale, AZ 85261
Fax:      (602) 391-0023                              (Yes! I'm available)
--------------------------------------           ******************************

Re: Patriot problem, more info available?

[Bill Kinnersley]

In article <1991Jun7.080445.12661@netcom.COM> jls@netcom.COM (Jim Showalter) writes:
: 
: That's certainly how it reads. I've been told by people
: involved with the project that the system is programmed in a
: mixture of assembly, FORTRAN, and JOVIAL--but from the way it
: failed it sounds more like LISP!
: 
How old is the Patriot missile system?  If such a system were implemented
today, wouldn't it be required to be done in Ada?

-- 
--Bill Kinnersley
  billk@hawk.cs.ukans.edu
226 Transfer complete.

Re: Patriot problem, more info available?

[Jim Showalter]

>How old is the Patriot missile system?  If such a system were implemented
>today, wouldn't it be required to be done in Ada?

The system was begun before the Ada language definition was complete
and before the Ada mandate was in place. If it were started today,
it would be done in Ada.
-- 
**************** JIM SHOWALTER, jls@netcom.com, (408) 243-0630 ****************
*Proven solutions to software problems. Consulting and training on all aspects*
*of software development. Management/process/methodology. Architecture/design/*
*reuse. Quality/productivity. Risk reduction. EFFECTIVE OO usage. Ada/C++.    *

Re: Patriot problem, more info available?

[Kenneth Ng]

In article <1991Jun7.160927.1667@weyrich.UUCP>, orville@weyrich.UUCP (Orville R. Weyrich) writes:
: Hmmm ... reminds me of an article that I saw describing a program which
: converts LISP into FORTRAN.  It could be a LISP interpreter implemented
: in FORTRAN :-).

As a matter of fact I did once see a LISP interpreter implemented in FORTRAN
on a Perkin Elmer (now called Concurrent).  Once I saw that I was quite sure
that ANYTHING could be done in ANY language if you REALLY put your mind to it.

-- 
Kenneth Ng
Please reply to ken@hertz.njit.edu until this machine properly recieves mail.
"No problem, here's how you build it" -- R. Barclay, ST: TNG

Re: Patriot problem, more info available?

[Esther Lumsdon]

I believe I read this in Discover(y) magazine, or Spectrum. The clock in the
missile computer was out of calibration, so the radar thought the missile was
further away than it actually was (in time and distance). 
Didn't sound like a language problem.
------ Esther Lumsdon, employed by, not speaking for, Verdix Corporation
  esther@verdix.com  or  esther%verdix.com@uunet.uu.net
-- 
------ Esther Lumsdon  Product Support Engineer, Verdix Corporation
  esther@verdix.com  or  esther%verdix.com@uunet.uu.net
Formerly lumsdon@dtrc.dt.navy.mil; if all else fails, forwarding is reliable.