On 23 February 2013 17:08, Clark Morris <cfmpub...@ns.sympatico.ca> wrote: [quoting John Gilmore] >>The defense of the continued use of single-case ones once dual-case >>replcements had become available, notionally on economic grounds, was >>in fact an instance of the all but reflexive responses of >>bureaucratized EDP managements to new technology. > > Actually there is a more subtle and hard to deal with reason. Any > alphanumeric field comparison or sort on alphanumeric fields assumed > upper case only. If case insensitivity were to be required, all of > them would have to be rewritten. If not, other problems could arise. > This gets worse for non-English languages if possible. Name and > address matching algorithms must be interesting even in monocase.
Dual-case hardware, and to a great extent, software, became widely available many years before correct notions of collating for sorting and searching were articulated - let alone widely implemented. The notion that one can sort or search even in mixed-case English without diacritics using byte-at-a-time comparisons persists to this day in many quarters. The first published clear statement of the problem and its solution (as it happens, for French, and en passant for English), and the beginnings of a general solution for other alphabetic languages were to my knowledge found in the report by Alain LaBonté of the Quebec Government, "Règles du classement alphabétique en langue française et procédure informatisée pour le tri" in 1988. This was followed by the still very useful and readable 1990 IBM Redbook by Denis Garneau "Keys to Sort and Search for Culturally Expected Results" GG24-3516, sadly out of print, and available only ever on paper. Much of this was standardized, first for English and French in CAN-CSA-Z243.4.1 1996, and later in POSIX, the Unicode Collation Algorithm, and other places, and these organizations acknowledge their debt to LaBonté, and the people at IBM who were ahead of the game for many years. Amazingly, though, arguments over the necessity of this were still going on quite actively as late as the early 2000s, and in some ways continue even now. Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
