[Default] On 18 Jul 2019 09:42:13 -0700, in bit.listserv.ibm-main
0000000433f07816-dmarc-requ...@listserv.ua.edu (Paul Gilmartin) wrote:
>On Tue, 16 Jul 2019 19:22:51 +0000, Seymour J Metz wrote:
>>
>>The cardinal sin in language design is to make the compiler simpler at the
>>expense of the user. ...
>>
>I see a notable example of this in Rexx's not supporting expressions in
>compound
>symbol tails which some have justified as making recognition of assignments
>easier.
>
>OTOH, Pascal declined to provide an exponentiation operator in order to expose
>to programmers the underlying costly implementation:
> X ** Y is implemented in other languages as exp( log( X ) * Y )
>
>Physics graduate students complained to me when the FORTRAN runtime
>threw an exception on X ** 2.0 when X was negative. "Why can't I square
>a negative number?"
Why won't FORTRAN square a negative number? Is this behavior peculiar
to FORTRAN?
Clark Morris
>
>(Some BASIC interpreters take special paths when a number has an integral
>value.)
>
>And they complained when (FORTRAN, but easier in C pseudocode):
> for ( X = 0; X!=1.0; X+=0.1 ) { ...; }
>... never terminated. "Roundoff error!? Why doesn't it just use the exact
>value
>of one tenth?"
>
>DWIM?
>
>-- gil
>
>----------------------------------------------------------------------
>For IBM-MAIN subscribe / signoff / archive access instructions,
>send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
