IBM Mainframe Forum

Okay, you were TOLD to go to the COBOL manual, but went to the Internet instead. None of us experienced people are one bit surprised that your attempt to fix your problem failed so miserably. Anyone who takes code -- JCL, COBOL, C, or whatever -- from the Internet without understanding what they are doing deserves whatever happens.

As has been mentioned, (1) you need to follow site standards -- sites have been know to discipline employees up to and including termination for their refusal to follow site standards, and (2) we do not work at your site so we cannot tell you how to get working compile JCL since that relies so heavily upon how your site is set up and what software is installed. You must contact a site support person, your team leader, or coworkers to get the appropriate and correct JCL to compile your code. That compile JCL would then need the OUTDD option added; if you do not know how to add the option to your compile, then talk -- again -- to your site support group, team leader, or coworkers.

Responders, go a little easier on the guy, will you? There is no need to be demeaning or talk down to him. Obviously, this website is available for those seeking information online. Researching information on the internet is not an unusual task today. Even the shop I work in does not carry any hard-copy materials and suggests Google searches even for the most experienced mainframe programmers. The Internet is a valid tool and experienced, knowledgeable individuals like yourself can help contribute. But if your desire is to put down the less experienced, then don't respond. Otherwise you fail in your contributions.

Nobody said the Internet is not a valid place for research. The o/p was told, specifically, to use the COBOL manual (http://www.ibm.com is a good place to find it, by the way) and decided to use compile JCL found on the Internet instead of relying upon site personnel or the official manual. When someone is told specifically what to do, and ignores that advice, there's not much chance that the person will be able to resolve their problem without a LOT of help since the root cause of the issue is the person's inability to do what they've been told to do. Recognizing this is not "talking down" to the person.

Furthermore, do YOU know if the site the o/p works for is using DB2, or Endevor, or CA-Librarian, or CA-Panvalet, or Change Man, or Xpediter, or Strobe? Each of these products can affect the JCL used to compile a program. The statement is made, often, on this forum that the JCL for a compile must come from someone working AT THAT SITE since nobody else can possibly know the JCL or methodology to use for a compile. Not even data set names are standard across sites -- some sites use IGY.V4R1M0 for the COBOL compiler high level qualfiiers, while other sites remove the version reference and just use IGY. Unless someone is experienced enough to probably NOT be on a "Beginner's and Students" forum, they likely don't know enough to know whether some random JCL from the Internet can be useful at their site (considering how much sites vary).

Sorry, Richard, but you are flat-out wrong. An experienced developer might well benefit by researching a matter on the Web -- I've done so myself, and no doubt you have also. For the less experienced, virtually every IBM manual is available on the Web, in HTML, PDF, and/or Bookmangler formats; it is a definite shortcoming of this site that it does not have a set of links available as a "sticky" or whatnot, although every fairly senior person here has repeatedly provided such links.

But the querent -- to whom I could apply many harsh but accurate terms -- just copied JCL that heesh found somewhere without understanding it in a blind and clumsy attempt to resolve his original problem, and was then unable to resolve the further problems that that caused. It seems to be common to the last couple "waves" of "software engineers" that they are unable and/or unwilling to do anything beyond copying and pasting from their browsers to their emulators.

Akatsukami wrote:Sorry, Richard, but you are flat-out wrong. An experienced developer might well benefit by researching a matter on the Web -- I've done so myself, and no doubt you have also. For the less experienced, virtually every IBM manual is available on the Web, in HTML, PDF, and/or Bookmangler formats; it is a definite shortcoming of this site that it does not have a set of links available as a "sticky" or whatnot, although every fairly senior person here has repeatedly provided such links.

But the querent -- to whom I could apply many harsh but accurate terms -- just copied JCL that heesh found somewhere without understanding it in a blind and clumsy attempt to resolve his original problem, and was then unable to resolve the further problems that that caused. It seems to be common to the last couple "waves" of "software engineers" that they are unable and/or unwilling to do anything beyond copying and pasting from their browsers to their emulators.

Here here!!!
The lost art of reseaching and reading the fine manual has gone the way of water coolled processors and replaced with the easy 'cut and paste' mentality.

Hello,

Methinks the internet can be quite helpful - if you actually know what you are doing. I use it multiple times a day. . .

Unfortunately, junk posted on the internet can be as (or more) attractive than the "real thing" and if you do not have a solid foundation, you can get yourself in worse trouble just grabbng "stuff" and dropping it on your system. It is so dangerous, that some organizations consider this grounds for termination. . .

We repeat this rather often, but NO One should download "compile JCL" and try to use it on their system. . . All compiles on a system should use the standard compile jcl for that system. If there needs to be changes to the standard compile jcl, it should be done by those supporting the compilers.

Robert Sample wrote:The statement is made, often, on this forum that the JCL for a compile must come from someone working AT THAT SITE since nobody else can possibly know the JCL or methodology to use for a compile. Not even data set names are standard across sites -- some sites use IGY.V4R1M0 for the COBOL compiler high level qualfiiers, while other sites remove the version reference and just use IGY. Unless someone is experienced enough to probably NOT be on a "Beginner's and Students" forum, they likely don't know enough to know whether some random JCL from the Internet can be useful at their site (considering how much sites vary).

I completely agree with you that JCL samples that novices pull from the internet in no way match their own environment in which they are working. That was not my point. It simply is a fact that you are going to get novices on any website looking to resolve a problem-at-hand because they are simply in the position to solve it (whether they're experienced or not). Regardless of how experienced a responder is, we've all been beginners (or even novices) at one point. I also know, even after years of working with this stuff, that I often get tangled up in an issue that would be considered novice and can use a refresher from time-to-time.

My point is this: IF you are going to respond, respond with factual information and include in your "facts" of the matter that JCL variables differ from system to system (even if you've repeated it to others a thousand times). If you're tired of hand-holding them, then don't respond any more. If you told the person to go to the manual and they didn't, there's no need to take it any further (let them chew on the information you gave them). Otherwise, you make others timid to even ask a question in fear of being "put down" for even asking.

Akatsukami wrote:Sorry, Richard, but you are flat-out wrong. An experienced developer might well benefit by researching a matter on the Web -- I've done so myself, and no doubt you have also.

I assume that was to me (KSRichard) - I never said anything about an experienced developer not benefitting from researching the web. You may need to re-read what I said (quite the opposite). I just believe particular types of comments to others is not necessary to achieve the purpose of this forum.

In all these posts, I'm surprised no one mention the DDNAME=alternate keyword. About the only time it's commonly used is in cataloged procedures that invoke the Binder or the old Linkage Editor, as in -

//LINK PROC
//LINK EXEC PGM=IEWL
//SYSPRINT DD SYSOUT=*
//SYSLMOD DD DISP=(,PASS),UNIT=SYSDA,SPACE=(CYL,(1,1,1)),DSN=&&G(GO)
//SYSLIN DD DDNAME=SYSIN
//SYSUT1 DD UNIT=SYSDA,SPACE=(CYL,(1,1))
// PEND

and it's called like this -

//LINK EXEC LINK,PARM='MAP,XREF,LIST'
//OBJLIB DD ---
//SYSIN DD *
INCLUDE OBJLIB(member)
ENTRY NEWPGM

Now this does not change the DD name as seen by the program, but it does provide a mechanism to "change" the DD name in the JCL, which may be the real goal of the topic starter.

KSRichard,

If it was your board, you could tell people how to operate. It's not. If you want to discuss this further, look for other parts of this site with discussion-type names and start a new topic. This topic is getting even more cluttered than it already was.

Steve,

Good point. TS was never really clear. I assumed they had a problem with segregating SYSOUT from their program from SYSOUT from Language Environment in the same step, but who knows?