I agree with Gerhard. Control of write-access to critical production libraries is reasonable and easily achieved using RACF or an equivalent. To go further is unwise for at least three reasons.
First, different people pursue development in different ways; and to attempt to permit it to be done in only one, notionally canonical way is arrogant and, in my experienvce at least, almost always retrograde. It enshrines some manager's already obsolescent notions of how to do things 'properly'. Some of the strongest objections to mainframes that I hear from the young have to do not with mainframes themselves and not even with JCL. They have to do with the bureaucratic encrustations---Too many rules!--- that surround mainframe use. Second, as Gerhard has already pointed out, able people will defeat any control mechanisms you put in place; and they will make a game of doing so. Third, such schemes encourage user groups to keep what are really production systems under their own private control. In two very large American banks that I know of the daily B statement to the Federal Reserve is fired off by an assistant treasurer sitting at a TSO terminal, triggering substantial, crucial processing that the IT organization wots not of. In raising children it is useful to ask the question: Do I need to say no? before one says it reflexively; and the same principle is useful in IT management. John Gilmore, Ashland, MA 01721 - USA ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
