Jim Wise writes:
>On Wed, 4 Jun 1997, Greg A. Woods wrote:
>> Leaving things as they are results in the least work all around.

>Sure, except for the fact that the way things are now is _broken_.  It
>has been pointed out several times that our current system in no way
>compensates for the difference in argument promotion between the two
>languages, with the result that you _cannot_ use either interchangeably.

This is not so much a fault with using K&R vs ANSI, as broken code.

>> Changing the code won't make it better -- it will only make it harder to
>> port.  The current style is an ideal compromise.

>Far from being ideal, the current solution is _broken_, and results in
>incorrect code.

I assume you are refering to code like

int
foo(char c) 
{ 
}

which IMHO is usually bogus.  If you write 

int
foo(int c) 
{
	/* frob some hardware with a char by casting if need be */
}

then

int
foo(c)
	int c;
{
}

is totally equivalent, and the passing of c as an int cf char will
require fewer instuctions on many machines.  And if you were not
frobbing hardware what is the point of using a char or short?  
To save memory? :-)

[back to trying to keep out of this...]

--sjg