From:	lhall AT rfk DOT com (Larry Hall)
Subject:	Re: strcasecmp revisited
1 Dec 1998 08:49:50 -0800 :
Message-ID:	<3.0.5.32.19981201113638.00b2c770.cygnus.cygwin32.developers@pop.ma.ultranet.com>
References:	<19981201152147 DOT 18729 DOT rocketmail AT send103 DOT yahoomail DOT com>
Mime-Version:	1.0
To:	earnie_boyd AT yahoo DOT com, cygwin developers <cygwin32-developers AT cygnus DOT com>

At 07:21 AM 12/1/98 -0800, Earnie Boyd wrote:
>---Christopher Faylor <cgf AT cygnus DOT com> wrote:
>>
>> 
>> My only concern is that some target or other may have code which
>> relies on the current behavior even though it is non-standard.
>> 
>
>But, that means the analyst is conforming to an undocumented feature. 
>I'd rather it have a standard behavior, and since cygwin is supposed
>to be ANSI/POSIX standardized IMHO you should make the change without
>regard to what someone might currently be doing that is non-standard.
>


I'll second that.  Better to break something for the sake of a standard
than to propagate a non-standard.  In the end, the latter will be a 
continuous up-hill battle.


Larry Hall                              lhall AT rfk DOT com
RFK Partners, Inc.                      (781) 239-1053
8 Grove Street                          (781) 239-1655 - FAX
Wellesley, MA  02482-7797               http://www.rfk.com

- Raw text -