As usual, Steve Hobson has his own views, which are very different from mine.
I prefer to use 'null' for pointer values that point nowhere, either x'00000000' or x'0000000000000000' and 'nul' for the byte/character x'00'. This issue is not, however, a very important one. About his "logically appropriate data typing" I am much less enthusiastic I prefer to use macro-language facilities like |&nul setc BYTE(x'00') On 8/4/14, Steve Hobson <[email protected]> wrote: > At the risk of upsetting people again... > > I prefer not to call 0 "null" (which, btw, appears to be the preferred > spelling in PoPs) when what I mean is "zero". Null is a control character > which has particular functions (for example, marking the end of a > null-terminated string) and which happens to have the value X'00'. It does > not have the functions *because* it has the value zero. > > Consider: If I want to clear the high-order byte of a register, I don't > say "insert a null into the high byte". I say "insert binary zeros into > the high byte". If I want to set up a null-terminated string, I say "add a > null after the last character", not "add binary zeros after the last > character". > > More generally, I prefer logically appropriate data typing, for example: > > DC AL1(64) A byte containing the number 64 > or: DC FL1'64' A byte containing 64 > DC B'01000000' A byte with bit 4 on > DC C' ' A byte containing EBCDIC space > DC X'40' A byte containing X'40' for some other > reason > > > > Best regards, Steve Hobson > > Je me presse de rire de tout, de peur d'être obligé d'en pleurer > Unless stated otherwise above: > IBM United Kingdom Limited - Registered in England and Wales with number > 741598. > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU > -- John Gilmore, Ashland, MA 01721 - USA
