Re: [TUHS] Qed vs ed

The version of qed Leah refers to is not pure. It was created in the late 1970s by Tom Duff, Hugh Redelmeier, David Tilbrook and myself by hacking the (already hacked) v6 ed we had at the University of Toronto, to restore some of the programmability that had been removed when ed was created, and to have fun. Mostly to have fun. Tom Duff used a real QED (sic) on the GCOS at Waterloo and was a bit conflicted about all this. For me the programmability was fun, and I wrote the tutorial, but what ended up sticking with me was the ability to edit multiple files simultaneously, something no other editor I had available could do. I don't mean switching between files, I mean making things like global substitutions across *.c. I'd start a session by typing qed *.[ch] and go from there. Our qed traveled with me from Caltech to Bell Labs, where I used it to write jim, which I used to write sam, which I used to write Acme. I tried vi when starting jim, but again the one file thing was too much to bear. I also tried emacs, which could in principle handle multiple files but the interface was cumbersome - it was much too hard to open a new file in a subwindow - and without regular expressions I gave up after a day or two. Also with a 2-d screen and a 1-d input device (no mouse), vi and emacs were too remote, like giving directions to someone holding a map without being able to point at the map. Describing where you want to point rather than just pointing. Anyway that was 40 or so years ago, and it's clear from the screens on others' desks at work that my opinion on those matters is not widely shared. I had a lot of fun hacking qed, mostly side-by-side with Tilbrook, who was its biggest fan. He taught me a lot and I loved working with him. -rob On Sun, Jan 31, 2021 at 6:01 PM Lars Brinkhoff <lars(a)nocrew.org> wrote: