On Fri, 24 Oct 2025 17:57:55 -0500, Jay Maynard wrote:
> ...
>Once upon a time, I stated on Usenet that it seemed to me that Unix
>filenames were case sensitive because the folks who wrote Unix originally
>didn't want to expend the few instructions needed to make case comparisons
>insensitive. Dennis Ritchie corrected me, and said that no, they thought
>being case sensitive was the Right Thing.
>
BLDL and STOW are case sensitive. There I strongly believe that
"folks ... originally didn't want to expend the few instructions", or
the several for Windows-like case preserving. Further motivated by
the wish to use the SEARCH KEY CCWs.
Then ISPF and TSO implement a pseudo case insensitivity, which
CLIST inherits while making command snames case sensitive.
Many users will feel that filenames should be case insensitive but
data should be case sensitive. How should a sort utility treat:
Schmidt is a German surname
smith means "metalworker" in English
(It may depend on Collate option.)
ISPF makes RE case sensitivity optional in regex.
DFSORT makes INCLUDE RE unconditionally case insensitive.
When choosing case insensitivity designers must carefully
consider what its scope should be.
--
gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
