Paul Gilmartin wrote:
In a recent note, Gregory, Gary G said:
Date: Mon, 27 Nov 2006 13:38:13 -0500
I believe that IBM "owns" "A - I" for their products.
Does this mean that customers and ISVs should avoid data set names
beginning with "A - I"? (Remember that the OP's concern was with
data set names, controlled by RACF, not member names, which mostly
don't concern RACF). And IBM certainly doesn't stay within the
"A - I" bounds for members in, e.g., SYS1.MACLIB.
-- gil
Yeah, well, an awful lot of IBM software got written before there
*were* packaging rules and naming prefixes. So don't expect all
our macros and module names to follow the conventions--the old
ones do not, and for compability reasons they likely never will.
For example, we will probably never rename the NUCLEUS, LPALIB,
and LINKLIB libraries even though they don't follow the naming
standard, or rename the WTO and ENQ macros because they don't
start with the "proper" CONSOLE and GRS prefixes. (And we
wouldn't even want to *think* about renaming all of JES2's parts,
for which I suspect that Ed, at least, is probably grateful.)
Naming conventions for *new* IBM libraries on the z/OS platform
use DDDEF names and low-level qualifiers in the form Spppxxxx and
Apppxxxx, where S is the target library prefix, A is the DLIB
prefix, ppp is the three-letter component (or product) prefix,
and xxxx is whatever the packager chooses. Likewise, names for
*new* elements is pppxxxxx, where ppp is the prefix and the
packager chooses xxxxx. Except, of course, for components and
products with 4-character prefixes, which are Sppppxxx and Appppxxx.
It's the values of "ppp" and "pppp" we reserve. We reserve App -
Ipp and App - Ippp, leaving Jpp-Zpp and Jppp - Zppp to other
vendors. We implicitly reserve $xxxxxxx as well, for JES2,
though I don't know whether that's actually documented anywhere.
And, as someone else mentioned, element at us.ibm.com can be
contacted to reserve a prefix.
So far as I am aware, there is no IBM naming standard for
operational data sets. That's probably OK, though, because I
think that few, if any, of those created over the past several
years have required names, and even many of the older ones either
no longer exist or can now have different names than the defaults
specified for them.
Last, we come to what I think was the original topic, that of
tape data set names. All SMP/E-installable IBM product tape data
set names should have the high-level qualifier IBM, and RFDSNPFX
is used on the ++FUNCTION statements to tell SMP/E about the
high-level qualifiers of RELFILEs.
This function was created to solve *exactly* this problem for
IBM's products, so you would not have to define new high-level
qualifiers to your security system for every new product you got
from IBM. There is no reason I know of that other vendors cannot
use it to do the same, though I have no idea whether they do or
don't.
--
John Eells
z/OS Technical Marketing
IBM Poughkeepsie
[EMAIL PROTECTED]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
