The combinatorial problem---How many unique, licit DSNAME values are there?--is easy to solve; one can even get the right answer if one remembers that x'7c', x'5b', x'7b'---They are '@', '#', '$' in the United States---are syntactic majuscules and that left-to-right index-level repetitions, as in
SYS1.SYS1. . . . are interdicted. The answers to such questions are, however, irrelevant. DSNs like ZZ9ZPFF5.SH70RDLU.Y22023A4. . . . are as a practical matter unusable in many contexts, as are those that contain embedded instances of obscenities in locally known languages, contextually inappropriate religious terminology, and the like. (SYS1.AMGOT would be offensive to Turkish readers; and an Italian-speaking CIO even objected to the index level ADREM, probably because he thought he was being baited.) Intelligibility requirements and assorted propriety constraints in fact make the usable names a small subset of the name space defined by the syntactic rules, at least when there are people in the loop. (There is a very limited role for otherwise 'barbarous' or offensive DSNs when they are generated for exclusively internal use. People unfortunately find out about them, and they must then be looked for and screened out.) John Gilmore, Ashland, MA 01721 - USA ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
