bugs

[David Burgun]

On 9 Apr 2006, at 21:04, Rob Cozens wrote:

> Geoff, et al:
>
>> Well, my international (from an american viewpoint) friend, how  
>> about this:
>>
>> on mouseUp -- display the date
>>   answer the system date with "OK"
>> end mouseUp
>>
>> The requirements are stated clearly in the comment, so the long  
>> date is out. The system date should work properly for you, correct?
>>
>
> Yes, the system date meets my requirement.
>
> But I would point out that this is not an international issue per se:
>
> Any user of Windows and MacOSes has the capability of specifying to  
> the operating system the format in which dates, times, and numbers  
> are displayed.  In my view, any software that ignores the user's  
> preferences is flawed, and therefore buggy.  It may be a design  
> bug, not a bug in the code; but it's a bug to moi.
>
> Since my buddy Bernie isn't on the List, here's a further example  
> at his expense:
>
> Flexware's A/R Parameter Setup screen worked flawlessly...so long  
> as the G/L module was not installed OR a Chart of Accounts was set  
> up.  If the user indicated in the A/R Params that G/L was  
> installed, the setup would loop endlessly trying to verify the  
> first G/L account # in the parameters record if there were no COA  
> set up.
>
> 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.
>

Yes, but what's the bug? The fact that it loops endlessly or the fact  
that the COA was non existent? This is where robustness comes in,  
nearly every instance of a bug (with the exception of typeo's) is  
because something was assumed to be set to something that wasn't/ 
isn't set up!

All the Best
Dave