on 5/12/05 10:42 PM, Joe Zitzelberger at [EMAIL PROTECTED] wrote:
--------------SNIP-------------------->> z/OS Core Technology Design > > I would be quite happy if you expanded the current PARM keyword to > allow for 65535 chars. This could have some issues for older programs, > but the benefits when running parameter intensive job steps (like IMS > or a compiler) are worth the trouble. > > There are two other issues with JCL that I wish you would consider > addressing at some point. Allowing instream datasets in JCL procedures > and symbolic replacement in parameter datasets. > > Possibly because of the historically short parameter length, many > programs use files instead of JCL parms for simple control statements. > This leads to the creation of two very common user written utilities -- > a JCL-parm-to-dataset utility and a replace-symbolics-in-a-dataset > utility. I believe there are several versions of both of these > available from the CBT tape and other sources. > > Getting a client to install unsupported freeware or convincing one's > own management to allow unsupported freeware is often a problem. Thus > it seems that every shop has a completely different way of doing these > two simple but very necessary tasks. Most of the flavors of these > utilities are quite easy to use, but they all consume one of the very > precious 255 job steps. > -----SNIP--------------- Well we *KNOW* that IBM has dug in their heeels on the &date issue (and other symbolic issues). I wish they would revisit that decision and put all the disclaimers they wish on it. There is a need for it and we see the question come up on here almost every other month or so. Installations need it. Ed ---------------------------------------------------------------------- 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
