bugs

[Geoff Canyon]

On Apr 9, 2006, at 9:34 AM, Rob Cozens wrote:

> Geoff,
>
>>> The thing is, that is not a bug.  The programmer did not make any  
>>> error
>>> in his code at all.
>
> You've removed the smiley; so I'm taking back some of that slack.
>
> The original request was for bug-free SOFTWARE, not a bug-free CODE  
> snippet.
>
>>>   The code works as it was intended.
>>>

Ignoring for the moment that it was Garrett who wrote that and not  
me ;-)

The original statement I responded to was this:

On Apr 6, 2006, at 8:11 PM, Mark Wieder wrote:


> As a QA engineer, I'd love to find some bug-free software someday.
> Doesn't exist. Bug-free is code-free.

I agree that my example obviously qualifies as "software" only in the  
technical sense of the word. But I was really responding more to the  
statement after: "Bug-free is code-free." By _that_ statement, I  
think one line qualifies as a counter-example, if it is bug-free.


> So you're in the same camp as my Flexware buddy, Bernie Mulcahy:  
> "my software is bug-free if people use it correctly"?

I don't think I would ever phrase it that way. I'm reminded of a  
support issue I once dealt with. It was a hardware issue, but even so:

A user called because her cd drive wasn't working. I went to help.  
She wasn't there, so I put in a cd, and it worked. I left her a note  
and went back to my office. Later she called and said it still wasn't  
working for her. I don't remember how long it took me to do this, but  
eventually I asked her to show me what she was doing. Here's what she  
did:

She pushed the button to open the cd drive (this was on an old  
PowerMac with a tray drive). She put the cd into the tray, label side  
up as it should be. She then shoved the tray closed. I mean _shoved_.  
You could clearly hear the mechanism whining in protest as she forced  
it shut.

Now, I don't know why that should cause the Mac not to recognize the  
cd. After all, the cd was correctly in place to be read. The fact  
that she was destroying her drive's open/close mechanism shouldn't  
have any bearing on whether the Mac could read the cd. But it  
obviously and repeatably did.

So, the drive was bug-free as long as it was used correctly ;-)

<snip>

> Simply meeting a specification does not guarantee that every user  
> who installs your software will experience no problems using it.   
> The critiques posted by others list several valid deficiencies in  
> your example.

Agreed re: specifications not being correct, but I would argue that  
code that matches the specification _is_ bug-free. The _requirements_  
have problems, but it isn't the code's job to exceed the  
requirements. I think people have listed valid deficiencies in my one- 
liner as you say, but I don't think anyone has demonstrated a bug  
(apart from the possible question of international date formats, now  
fixed).

Finally, a funny coincidence: my middle name is Garrett. But I'm not  
Garrett Hylltun in disguise. ;-)

Regards,

Geoffrey Garrett Canyon