Most users don't see abends anymore because for the most part the system works. 
None of my end-users even know about IBM's manuals.
When an error occurs and it come to my application programmers or help desk 
support, they will either know the answer because it is a common occurrence and 
fix it.
If it is not a known problem, they will come to me. I may know the answer, or 
not.

But I AM the frequently the first person to look it up in the manual. And 
again, see myself just a funny bit of illogic. Yes, I know how to proceed from 
there, but the non sequitur bit of explaination is not really helpful.

As for LE, I know it does a lot and is much improved from the early days, but 
hiding an S0C7 as a U4xxx error is not helpful.

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Peter Relson
> Sent: Friday, October 31, 2014 5:44 AM
> To: [email protected]
> Subject: Re: Abend s0077
> 
> Really? I'm puzzled.
> 
> You don't like that the explanation is "what does it mean" and that the action
> is "what to do"?
> 
> The intent (and I do not claim that that intent is realized in enough
> cases) is surely
> - whoever sees the abend information (this will usually be a user, might be an
> operator) reads the explanation
> - they read the action and take the action (which might be to contact
> someone else)
> - repeat until you've reached the right party.
> 
> It must be relatively infrequent that the person who gets the abend (and thus
> is the first to see the abend) is the system programmer. It might well be that
> the user who did get the abend doesn't do what they're supposed to do (as
> evidenced by the 0C7 example), but they did see it first.
> 
> I actually think in the S0077 case that  the system programmer responses for
> 003C and 003D belong instead within programmer response. If a programmer
> did something wrong that they need to fix (whether that is an authorized or
> unauthorized programmer), then that belongs within programmer response.
> 
> As to the point about ISV's and abend codes: ISV's do not write what is 
> thought
> of as "the system". They are not in general supposed to issue system abend
> codes (although some codes are reserved for use by owners of non-system
> SVCs). It seems that a good part of the hang-up is thinking that "user" in 
> this
> case necessarily means, for example, some unauthorized TSO user running a
> program. It doesn't. It more closely means not-the-IBM-provided-system.
> Could it have been defined differently 50 years ago? Sure. I'd guess that LE
> came in similarly as "not the system".
> It is rare that unauthorized code (as LE originally was) would use system
> completion codes.
> 
> Peter Relson
> z/OS Core Technology Design
> 
> ----------------------------------------------------------------------
> 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

Reply via email to