Skip to content

Comment on Unix: A History and a Memoir, by Brian Kernighan

Comments

Rummaging around the local mall bookstore in ~1982, I came upon 'Software Tools in Pascal', by Kernighan and Plaugher. I fell in love with the ideas, and the prose. It became the first of what is now a nearly complete collection of everything Kernighan has published (I don't have the AMPL book, or 'D for Digital'.) I can't calculate how much I know because of Dr. Kernighan, or how much my career's course has been altered by the levers he's given me, but it's a large number.

This is now on my Amazon wish list (How many of you have private lists for 'things to remember and check out later?' Mine's called 'Random Followup Stuff')

You might find this interesting, then: http://www.cs.virginia.edu/~evans/cs655/readings/bwk-on-pasc...

It's Kernighan's view of Pascal after writing "Software Tools in Pascal". It's not (as widely perceived) a hit piece on Pascal. It's Kernighan saying "I wrote the original 'Software Tools' in Ratfor. Rewriting it in Pascal should have been much easier than it was. Why?"

I know the piece well. Back when, my university used "Oh, Pascal!" to teach introductory programming to programmers. For worse reasons than Kernighan, I never liked that the length of an array was part of the type in Pascal, but, in retrospect, that might've saved a couple Trillion in buffer overflow problems.

At the price of being fundamentally unable to deal with variable-length arrays? That seems like taking away a rather fundamental ability.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.