From: slocombe@vex.net Subject: Re: Phototypesetting history Date: February 19, 2009 11:33:23 AM PST (CA) To: jmax@sfu.ca Reply-To: slocombe@vex.net

Do read this PDF page:

http://books.google.ca/books?id=QYu_v2R6fIQC&pg=PA926&lpg=PA926&dq=graphic+systems+C/A/T+phototypesetter&source=bl&ots=yMD1i0jzwT&sig=iEbwFiQ5c30HMCFsG68HoYv3DTg&hl=en&ei=p7CdSdr0MozaNNOOsZ0F&sa=X&oi=book_result&resnum=1&ct=result

(Which is a page I found with Google. It should be entitled "A Bit of Unix Typesetting History".)

That page quotes from:

http://www.kohala.com/start/troff/cstr97.ps (Note: Postscript, not PDF) which is Kernighan's paper on modifying troff to create ditroff. One short para in this paper has served me very well in the 30 years since, in everything I've done:

"Note that at this stage the program has not been changed in any fundamental way — it still generates CAT output, so a bit-for-bit regression test against the original version can be per- formed after each change. This proved to be very important for maintaining sanity in both pro- gram and programmer."

I (think I'm) still sane after all this time because I've always followed Kernighan's pattern of small-change/compare-with-previous/small-change/... pretty religiously in my projects. You don't go down the garden path very far that way before you know you've goofed.

Search this wiki: