Personally I think we should call them "MVS data sets" or "MVS files", as
distinguished from and "MVS Unix files". We still have "MVS JCL", so why muddy
the waters whenever IBM marketing decides to change the name of the "operating
system" again?
Seems like a more generic name for PDS/PDSE is needed, though. I don't suppose
"MVS library" is reason able. Certainly (to me) an "MVS library member" is a
member that is in either a PDS or PDSE (library) data set. So "MVS library"
for PDS/PDSE/PDS(next) and "MVS Unix directory" for, well, and MVS Unix
directory.
To summarize, I'd be comfortable with all of the following terms:
- MVS data sets or MVS files, including:
- MVS sequential data sets or MVS sequential files
- MVS VSAM data sets or MVS VSAM files (or just VSAM files)
- MVS libraries, including:
- (MVS) PDS libraries and (MVS) PDSE libraries
- Are there other MVS non-Unix file/data set types?
- MVS Unix file system, including HFS and zFS (and others I may not know about)
and containing:
- MVS Unix directories, containing:
- MVS Unix files
On person's thoughts...
________________________________
From: IBM Mainframe Discussion List <[email protected]> on behalf of
Tony Harminc <[email protected]>
Sent: Wednesday, April 26, 2017 6:53 PM
To: [email protected]
Subject: Re: Terminology - Datasets
On 26 April 2017 at 20:27, Paul Gilmartin <
[email protected]> wrote:
> But what happens when "z/OS" goes the way of "OS/390" and "MVS 5.2"?
>
Yeah, that happens too. But it's a pretty easy context-free change to make,
and of course we've had to do it. Not that customers wouldn't continue to
understand "OS/390", but it makes it look as though the doc hasn't been
updated for years. (The first time around, we briefly had a lot of "an z/OS
dataset" and the like from a careless change all command.)
>
> > ... So that's what we say now: "UNIX
> >file", or in the rare case it's possible to be confused with a file on
> >another UNIX system, "z/OS UNIX file".
> >
> Is that intended to exclude NFS files (and possibly others) which
> don't support zFS extended attributes? (Don't know about TFS.)
>
Not explicitly. But I can't imagine a product the customer knows is going
to run as a z/OS started task with significant performance requirements is
going to decide to put a transaction-containing file on an NFS server on a
Linux box across the country. Nor, for that matter, are they going to make
it a temporary file.
What about sockets? Can SVC 99 allocate a DDNAME to a socket?
>
I'm willing to bet not. Can JCL?
Tony H.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
