Re: Do we need test cases?

On Sun, Sep 05, 2004 at 01:21:07PM -0500, Rod Adams wrote:

>Personally, I could go either way. But I would be more inclined to 
>volunteer as a summarizer if the test cases were in use.
>
>What do others think?

In RFC2119 terms, SHOULD seems reasonable: "there may exist valid 
reasons in particular circumstances to ignore a particular item, but 
the full implications must be understood and carefully weighed before 
choosing a different course".

A test case for the calendar problem, for example, would have required
an additional "use this date as opposed to the current date" parameter
to be reproducible.

Roger