In a recent note, Walter Farrell said: > Date: Tue, 13 Feb 2007 12:57:00 -0500 > > You're right, though, that all the applications that are passing the > password along need to know to leave it as the user entered it. That > makes migrating to mixed-case passwords harder than it would have been > if we'd made the security product do the upper-casing of the input many > years ago. > A similar principle should have been applied to data set and member name transformation and enforcement -- this should have been done in a single common component at a low layer. If the intent of the Data Management design was to have a mixed case file system, all names should be taken as-is. If the intent was to have a single-case file system, any attempted use of the other case should result in a syntax error. If the intent was to have a case-insensitive file system, a low level component should perform the translation.
Alas, Conway's law took its pernicious toll. The design groups didn't communicate and did not form a common objective. In consequence, allocation assumes mixed-case and takes names as-is. JCL and Catalog assume single-case and treat most uses of lower case as syntax errors. And TSO et. al. assume case-insensitive and convert to upper before calling lower level layers. -- gil -- StorageTek INFORMATION made POWERFUL ---------------------------------------------------------------------- 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
