27-May-2005

jleyva in   Métricas Web continues his series of articles on extreme programming, In this post he talks about testing.

Catalogue of feeds from the Microsoft knowledge base
Catálogo con los feeds de la base de conocimiento Microsoft

Juan Palacio in  Navegapolis posts a link to the RSS Feed Product Index of Microsoft.

The computing error that caused the loss of the Viking satellite on its way to Venus
El error informático que hizo perderse a la sonda Viking camino de Venus

RFOG in  Diario de un Programador has been reading  Real Time Systems and Programming Languages: Ada 95, Real-Time Java and Real-Time C/POSIX (3rd Edition) by Alan Burns, Andy Wellings in Spanish translation. One of the situations it describes is the famous Fortran error that was thought to have lost the Viking satellite on its way to Venus. The Fortran line should have been like this:

• DO 20 I = 1,100

Instead of which it was:

• DO 20 I = 1.100

So rather than do an iteration 100 times, the value 1.1 was assigned to the variable DO20I. So the million dollar question that he poses is, "Should software engineers be responsible for the consequences of errors in real time systems?" He has opened a Forum thread for the community to respond ¿Reponsable de un fallo de soft en un sistema en tiempo real? por rfog

.

[I did a bit of research on the news item. This particular story dates back a long time, there is an analysis in the RISKS-LIST: RISKS-FORUM Digest Friday, 27 November 1987 Volume 5 : Issue 66. Of course the truth of the news item in no way changes the fundamental question. Researching this question brought up an article in Stanford Encyclopedia of Philosophy, Computing and Moral Responsibility. ]