On Mon, 9 Sep 2013 16:16:45 -0400, John Gilmore wrote: >Peter, > >To me 'experimental' is never pejorative. An experiment queries the >environment to make date-driven decisions. > I might accommodate Peter by using "adaptive" or "facultative" in place of "experimental".
>We do, however, disagree about the proper treatment of the INTRDR or, >indeed, any other tool. If you do not want some job submitted via the >INTRDR to update a file you protect that file and you do not interdict >the use of the INTRDR. Another way to make the same point is to >recall that the INTRDR is not a new facility. It is in fact a very >old one, and a scheduling product or security scheme that is unaware >of it is 1) poorly designed and 2) should not be used. > To which I'll add 3) The scheduling product itself probably uses INTRDR. >In my now long experience shops that use scheduling products >inflexibly, while they may meet their notional workload-control goals, >in fact lose 1) control of what is really going on and 2) relevance. >Decision-support and extrernal-reporting requirements are moved >outside its purview. It is left with tight control over routine >applications that are of interest only when they falter. > I would expect that one of the functions of a scheduling product is to satisfy the external-reporting requirements; maintaining an audit trail of what job was run when, and why. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
