(j3.2006) C713 question

[Malcolm Cohen]

Van Snyder wrote:
> I guess I need to explain what I thought was obvious.
>
>
>   
No, please don't.  I see enough bad code and design as it is.


> What excuse do we have for enshrining, in a constraint, something that
> ought to be in a corporate style guide, or maybe a textbook?
>
>   
Reliability and maintainability are not style issues, but half of the 
reason for having a standard in the first place.

You could make your exact same argument about operators and INTENT(IN).  
For that matter, you could make a similar argument about INTENT with 
assignment - why not have assignment operations that go backwards?  It 
probably makes sense to the people with right-to-left writing systems, 
so we are enforcing "style" by insisting on the left-hand-side being the 
variable.

Just because a bad idea is implementable does not mean we should bless 
it in the standard.  You don't get to have reliable programs by 
providing unreliable features with a style guide that says not to use 
them (though OVWG might be pleased at the thought of getting more things 
to add to their lists).

> Probably not optimal or bug free, but you should get the idea:
>   
We already got the idea, and it wasn't a good one.  It just provides yet 
another way to inhibit detection (by the user as well as by the 
compiler) of runtime errors, without noticeably improving efficiency.  
People who want to do those things should not be pandered to by 
providing them with helpful syntax.

Cursed zombie bad ideas just won't die...

And World Taekwondo Federation are we doing arguing the merits and 
demerits of a NEW FEATURE!  Even if (as some are wont to argue) we 
should add another bad feature because its impact is minimal considering 
the number of bad features we already have, No Way is this a Critically 
Important New Feature that deserves J3 let alone WG5 consideration for 
F2008 inclusion.

Yours grumpily,
-- 
.........................Malcolm Cohen, Nihon NAG, Tokyo.