Monday, November 30th, 2009

How would a person "test" and "accept" software for automated aircraft landing ?

The software might work for a suite of "static" test cases but be unsuitable for "real-world" input data. (The real-world data might fluctuate around average values).

How would you educate your CEOs etc on the necessity to "ruggedize" the software so as to make it work for real-life data ?

Mandate that CEOs take an airplane ride once a week.

This entry was posted on Monday, November 30th, 2009 at 11:31 am and is filed under software automated. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

3 Responses to “How would a person "test" and "accept" software for automated aircraft landing ?”

giginotgigi Says:
November 30th, 2009 at 5:20 pm

All depends on the matching of the static test data to the real-world data.

However, such tests would never complete, i.e. it cannot cover every situation. It is similar to doing penetration test in information systems, which is a never ending story.

Most software are tested according to certain standards, e.g. those in IEEE. Those are aiming on the perfection of the software itself, i.e. minimum glitch and vulnerability to zero deficit (zero means tolerable)
References :

who is #1? Says:
November 30th, 2009 at 5:25 pm

Mandate that CEOs take an airplane ride once a week.
References :

Froggies angel Says:
November 30th, 2009 at 5:33 pm

You can’t, no human nor computer can see what problems that might arise along the way, unless they are foreseer’s and thats not probable, possible yes, there are sages through time have seen things but logically thinking you don’t really stand a chance, look at the latest ie8 program, new downloads are suppose to protect you from sites that are dangerous, yet flaws in the program are doing exactly the opposite to what the programmer and designer intended, need I say more :)
References :
Froggie

Leave a Reply