The paper by Backus to which this post links is one I had not seen before. I read through the whole thing, and it struck me instantly as a classic which every hacker here should read. It is surely one of the best things Backus ever wrote.

Interspersed among long-obsolete technical details are gems of insight about software and programming languages that are worth their weight in gold. The final paragraphs articulate Backus' critique of "Von Neumann languages" in a way that even his (much better known) masterpiece on FP doesn't achieve.

I had no idea that, prior to Fortran I, the very possibility of a workable compiler had already been widely rejected as impossible because of grandiose claims made by vendors. In other words, vaporware is older than high-level languages.

And behind the dry academic tone one gets a feel for the the brilliant design work, all-night debugging sessions and camaraderie of one of the greatest teams of all time - as well as, to judge by his consistently humble self-effacement, one of the greatest team leaders.

So to the chorus of chicken littles incessantly bemoaning the decline of Hacker News, now's your chance to redeem yourself: shut up and read this paper!

Would have passed over it if not for your comment.

Read, worthwhile, archived. Thanks.

One of the problems FORTRAN was trying to solve (from Backus's paper): "The cost of programmers associated with a computer center was usually at least as great as the cost of the computer itself." Interesting how things change; the ratio would be more like 100:1 now.