On Sat, Oct 9, 2010 at 12:52 AM, Earl Miles <merlin@logrus.com> wrote:
On 10/8/2010 11:47 AM, nan wich wrote:
@Gerhard: 80 lines was how long a punch card was. What a ridiculous reason to use 80 any more. Are you even old enough to have ever seen a punch card? I almost forgot, the original IBM System/3 had punch cards
Yes, Nancy, there are actually a few adults on this list. Though I doubt many of us are old enough to have actually USED a punch card, since people who did work on punch cards should be pretty close to retirement age by now.
Nope! Used them in 1985 in a course, where the norm was to write your code on a sheet and then send it to an operator who will punch it and then it would be compiled from cards. That was a training course though, terminals were available at the same time. Also saw vendor system engineers who were puzzled for a couple of days and could not boot a new system because it was the first of its kind in that region NOT to have a punched card reader. Saw at least one client in the early 90s who had working punched card readers and JCL jobs for them. Still a decade or two until I retire. 80 characters was the common width of monitors, which descended from
punch cards, but is also pretty close to the 72 character width of the common typewriter (pica, if I remember right) with standard margins.
An 80 character card had a 6 character sequence number, for some languages (e.g. COBOL) so if the card deck falls on the floor you can do a sort run and it will sort it correctly. Column 7 was for comments (an * in COBOL for example). This leaves 73 characters. Not sure if that was related (can't use the full 80 on a terminal card image?) RFC
2822 imposed the limit (as a SHOULD not MUST) because many terminals failed to wrap on their own, and terminals often had 80 CPL in order to be standard. Though many terminals also had 132 or, if you were unfortunate enough to use a VIC-20 (and maybe a PET, I forget) you could get 40 CPL.
132 was much later, and was not a standard. Mostly some DEC VT, or that is where it started.
Also, RFC2822 is still in effect; if an email message is in text/plain, it is polite to go ahead and wrap at 78 per the spec. If your message is text/html then wrapping is pointless.
Yes! -- Khalid M. Baheyeldin 2bits.com, Inc. http://2bits.com Drupal optimization, development, customization and consulting. Simplicity is prerequisite for reliability. -- Edsger W.Dijkstra Simplicity is the ultimate sophistication. -- Leonardo da Vinci