bugs

Geoff Canyon gcanyon at inspiredlogic.com
Sun Apr 9 23:24:18 EDT 2006


On Apr 9, 2006, at 1:04 PM, Rob Cozens wrote:

> Is it a bug that Parameter Setup didn't check for the existence of  
> a COA when it opened?  By my definition, "yes", even if the specs  
> didn't call for it.

The thing is, this calls for common sense, but what is common sense  
if it isn't specified?

To give an extreme example, suppose I open Revolution and type the  
following into the message box:

2+2=

and press return. Is it a bug that Revolution doesn't put 4? Common  
sense would say what should happen, but that's not what the spec says.

Obviously this is an extreme example, but the question is where does  
the line get drawn?

The only way to be certain to make sure the spec is complete, and  
code to the spec. In the example above, the spec should be amended  
before the behavior is considered a bug. The rationale for this is  
clear: what _should_ happen? I agree that an infinite loop is a bad  
thing, but I don't know what the right thing to do is.

So no -- if the behavior doesn't violate the specs, I wouldn't call  
it a bug.

But I agree completely that common sense pertains, and the  
requirements need to be updated.

regards,

geoff



More information about the use-livecode mailing list