On Fri, 2 Nov 2018 13:01:05 -0500, Chris Bowen wrote: > >Customers are free to use such circumventions, vendors have to assume that >their customers would prefer not to. > >Full disclosure - I have no idea if my colleague's original code was intended >to run on customer systems or is for internal use only. > IBM should, of course, assume its code is intended to run on customer systems.
IBM's customers are unduly desensitized to the adverse consequences of Conway's law because of IBM's careless design practices. Consider: From: z/OS IBM UNIX System Services User's Guide Version 2 Release 3 SA23-2279-30 ... A path name can be up to 1023 characters long, ... But from: z/OS IBM MVS JCL Reference Version 2 Release 3 SA23-1385-30 ... PATH parameter ... Has a length of 1 through 254 characters, not including the slash. ... Why? Similarly, SVC 99, GUPI as far as I know, allows syntax for PDS member names which JCL treats as syntax error. Why? If the name is invalid (as an IBM rep has asserted) why does DYNALLOC not report an error. Don't these development groups talk to each other? -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
