Some such qualification is clearly necessary. The intent of such statements is to make it clear that the aggregate size of files maintained for and processed on mainframes is [still] very large in relation to those maintained in other, non-mainframe formats for processing elsewhere.
Any such announced percentage P is of course open to attack on the grounds that something has been improperly included in or excluded from the numerator or denominator used to calculate it. The value of P can only be known very approximately, but this is usually the case with such numbers: McDonalds may no longer know at all exactly just how many hamburgers it has sold, and taking refuge in "billions amd billions sold" enables it both to avoid delusive exactitude and to save on signage-updating costs. In this particular case I am all but certain that IBM has an MBA-wrought argument for P = 85 that has passed muster with its lawyers as non-specious, defensible if attacked. 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
