Mentioned here: UNIX: A History and a Memoir https://www.rulit.me/data/programs/resources/pdf/UNIX-A-History-and-a-Memoir_RuLit_Me_616356.pdf https://www.rulit.

Mentioned here: UNIX: A History and a Memoir https://www.rulit.me/data/programs/resources/pdf/UNIX-A-History-and-a-Memoir_RuLit_Me_616356.pdf https://www.rulit.me/data/programs/resources/pdf/UNIX-A-Hist... 5.3 Early Formatters The problem was that there was no interactive computer system like CTSS at Princeton; there weren’t any computer termi- nals either. All that was available was punch cards, which only supported upper case letters. I wrote Roff in Fortran (far from ideal, since Fortran was meant for scientific computation, not pushing characters around, but there were no other options) and I added a feature to convert everything to lower case while automatically capitalizing the first letter of each sentence. The resulting text, now upper and lower case, was printed on an IBM 1403 printer that could print both cases. Talk about bleeding edge! My thesis was three boxes of cards. Each box held 2,000 cards, was about 14 inches (35 cm) long and weighed 10 pounds (4.5 kg). The first 1,000 cards were the program and the other 5,000 were the thesis itself in Roff.