>>>> Unfortunately, most people don't seem to realize that a *library* with
>>>> no documentation is worse than useless.
>>>
>>> Yeah, that's very *very* true.
>>
>> It could be worse; it could have *incorrect* documentation. (!!)
> 
> Not really. Usually there's two kinds of incorrect documentation:
> 
> 1) Documentation that has bugs in it, which is no worse than a library 
> that has bugs in it.
> 
> 2) Documentation for the wrong product shipped. Which is pretty much 
> equivalent to "no documentation" when you discover that's what happened. 
> Annoying to spend a couple of days trying to figure out how to get 
> something to work as documented, finally ask on the newsgroups, and be 
> told "We never documented that. We just shipped the 1.x docs with the 
> 2.x product." Woulda been nice to know!

So you've never come across documentation that says "setting this value 
negative uses the smallest supported value", when in fact setting it 
negative causes a segfault?

Then again, would that be a documentation bug or a library bug? I guess 
the choice is arbitrary...

Post a reply to this message