comp.lang.ada
 help / color / mirror / Atom feed
From: barrington@logica.com (Jardine Barrington-Cook)
Subject: Re: Bug in code for Priority Queue
Date: Thu, 5 Jan 1995 12:23:20 GMT
Date: 1995-01-05T12:23:20+00:00	[thread overview]
Message-ID: <barrington.39.000C63EE@logica.com> (raw)
In-Reply-To: D1us76.Gp@inmet.camb.inmet.com

In article <D1us76.Gp@inmet.camb.inmet.com> stt@dsd.camb.inmet.com (Tucker Taft) writes:
>Newsgroups: comp.lang.ada
>Path: logica.co.uk!pipex!howland.reston.ans.net!gatech!bloom-beacon.mit.edu!uhog.mit.edu!news.mathworks.com!noc.near.net!inmet!dsd!stt
>From: stt@dsd.camb.inmet.com (Tucker Taft)
>Subject: Bug in code for Priority Queue
>Message-ID: <D1us76.Gp@inmet.camb.inmet.com>
>Sender: news@inmet.camb.inmet.com
>Organization: Intermetrics, Inc.
>Date: Tue, 3 Jan 1995 23:35:29 GMT
>Lines: 246

>Brian Hanson of Cray (brh@cray.com) kindly pointed out a bug
>in the code for a priority queue I posted several weeks ago,
>which arose during my effort to create a stand-alone
>priority queue package from a version that was embedded
>in something else.  So much for reuse by transcription ;-).

>In any case, here is a description of the bug and the fix...
 [.. real code cut out...]

There is a logical problem with the use of priority queues for warning 
messages, which I would like to publicise. I have seen this problem twice, and 
once was in an aircraft audio alarm system.

When putting an element of priority "n" onto a prioritorised warning queue, it 
is essential to search the lower priority queues for the same warning at a 
lower severity - if you do not do this the resulting warnings to the operator 
can be very misleading.

In a UK Television broadcast of the 1983 election results this caused 
two succesive headline messages to appear in the form:

"4th recount at Molesly North" followed by
"3rd recount at Molesly North".

In the aircraft audio warning case the results were of the form:

"Warning: aircraft below 30 meters"
"Warning: aircraft below 40 meters" 

(the aircraft was still descending, and below 30 meters)

I for one would be very happy never to see this problem again.


Jardine Barrington-Cook    ! "...See worlds on worlds compose one universe
Space Division - Logica    ! observe how system into system runs..."
...personal opinions,      !     
subject to change...       ! A. Pope - from "An Essay on Man", 1773



      reply	other threads:[~1995-01-05 12:23 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1995-01-03 23:35 Bug in code for Priority Queue Tucker Taft
1995-01-05 12:23 ` Jardine Barrington-Cook [this message]
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox