Subject: Re: UNIX manpage history: runoff From: Doug McIlroy Date: 21/10/2011 22:09 To: Kristaps Dzonsons > > First, do you know the whereabouts of your own MAD sources to the > > CTSS runoff utility for, if I'm correct, the GE-645 Multics? Actually I never saw any source from the runoff trunk and branch of the family tree. I can't remember what language Morris programmed his verion in. I don't perfectly remember the sequence of events, when runoff was ported to the pre-Multics platform, the GE-635. I was not involved. I wanted to extend the runoff model so began over, writing in BCPL. This new roff ran in Bell Labs comp centers as long as they had GE machines. (I have the 1973 edition of the manual.) It was reimplemented for Unix by Thompson and Ritchie--in assembly language, I believe. > > the 1974 Multics sources don't mention Bob Morris at all, only > > Mabee. So which version of runoff was ported to UNIX roff: yours, > > Mabee's, or Morris'? Was Multics runoff renamed to roff before or > > after this? I believe Morris christened his port "roff" to distintuish it from runoff when both programs were running on CTSS. I borrowed the name. I do not know whether my roff was adopted by Multics. You can tell by seeing whether the extensions are in the Multics manual. After T&R, Ossanna wrote nroff, which was still more elaborate, and parlayed that in turn to troff when the CAT phototypesetter arrived. Kernighan (who had written his own formatting program when he was at Princeton) stepped in upon Ossanna's death. Roff was in the first edition Unix manual, 1971; nroff appeared in the second, 1972; troff in the fourth, 1973. Thanks to troff the fourth edition was the thinnest Unix manual ever printed. Doug