8 Aug
2015
8 Aug
'15
8:12 p.m.
On 8/8/2015 5:47 PM, Chuck Guzis wrote:
On 08/08/2015 12:13 PM, Jay Jaeger wrote:
Actually, with FORTRAN II to IV: not. I spent quite a lot of time
translating FORTRAN stuff back in the day. COMMON blocks were always
handled differently, IO Formats differed, binary IO differed, etc. etc.
I have always felt that the language name is
SNOBOL, with multiple
versions, kind of like FORTRAN II (which is what the 1410 had),
FORTRAN IV, FORTRAN V, etc., but Griswold seems to think otherwise.
;)
I think the test would be "Can language x+1 run, without substantial
modification, programs written in language x?" If the answer is "no",
then languages x and x+1 are separate languages, and not compatible
dialects of the same language.
So FORTRAN IV would be a compatible dialect of FORTRAN II (mostly at any
rate, FII vendors had a nasty habit of adding their own features
willy-nilly, as did FIV). I think Codasyl was first to clamp down on
"the default is the standard as we say it is", then FORTRAN followed.
However, I'd submit that F95 is a separate
language, as it can't run
FIV, F66 or F77 programs without modification as it doesn't understand
ASSIGN-ed GOTOs as well as H-type (Hollerith) FORMAT specs.
--Chuck