> The rules are so convoluted that it ought to be simpler to remove the 
> exceptions.

Exactly. I have learned that the key is to write the documentation FIRST. It is 
often easier to write code that corresponds to something that was written 
clearly than to try to clearly document something that was written the easy way.

Charles


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Paul Gilmartin
Sent: Wednesday, October 28, 2020 11:47 AM
To: [email protected]
Subject: Re: Generic name for PDS/PDSE

On Wed, 28 Oct 2020 10:40:04 -0700, Charles Mills wrote:

>That's the problem. Not PGM=UNIX_thing specifically but that there are so many 
>special cases. One could not simply write an introductory paragraph "When we 
>say executable we mean any of ..." because then EXEC PGM=executable would be 
>incorrect; we would have to say PGM=executable* and *except a UNIX executable. 
>You end up with documentation that may be technically correct but reads like 
>gibberish.
> 
Yes.  I just checked the JCL Ref. for STEPLIB and JOBLIB.  It says that either
[may be] a concatenation of "private libraries".  "Private library" is defined 
in
two or more (obscure) places as "a partitioned data set or partitioned data set
extended".  Verbose, but excluding UNIX directories which would be allowed
by Using Data Sets.  The rules are so convoluted that it ought to be simpler
to remove the exceptions.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to