Knuth is suffering from extreme isolation from the world around him.
On unit testing:
As to your real question, the idea of immediate compilation and "unit tests" appeals to me only rarely, when I'm feeling my way in a totally unknown environment and need feedback about what works and what doesn’t. Otherwise, lots of time is wasted on activities that I simply never need to perform or even think about. Nothing needs to be "mocked up."
...and multiprocessing:
I might as well flame a bit about my personal unhappiness with the current trend toward multicore architecture. To me, it looks more or less like the hardware designers have run out of ideas, and that they’re trying to pass the blame for the future demise of Moore’s Law to the software writers by giving us machines that work faster only on a few key benchmarks! I won’t be surprised at all if the whole multithreading idea turns out to be a flop, worse than the "Titanium" approach that was supposed to be so terrific—until it turned out that the wished-for compilers were basically impossible to write.
Multithreading and multiprocessing aren't going away. Unit tests aren't some optional artefact, a crutch of the weak and/or stupid. Knuth is wrong. Think I'm taking this out of context? Read for yourself.
Anyone who believes that multithreading is a flop must be using indexes at the back of books to find information every day, or some search engine other than Google. Multithreading and multiprocessing scalability is the backbone of many of our favorite businesses, internet services, games, and passtimes. Got a 360? You've got six cores and a GPU. Got a PS3? Multiple SPUs, a CPU, and a GPU, the first time in my memory asymmetric multiprocessing has ever been quite this extreme *and* commercial. You've got Google, capable of providing near-instantaneous results of searches on vast chunks of the internet, and services like Last.FM, capable of tracking what every one of its users listens to and using that data provide recommedations. We live in a society steeped in parallel computing infrastructure and applications, disguised as the everyday applications we know and love.
Moreover, I'd like to emphasise the importance of unit tests in team-based development where the project or product outlives the developer. Knuth can literately document every brain cell in his head and it won't protect the code from changes from others once he's dead. And most programmers leave a project - they do not, unlike Knuth, spend their entire life refining a handful of projects. Knuth is a throwback, a window on the way things used to be - but doesn't in any way reflect the world as it is.
Knuth's own statements stand as proof that not only do programmers only change under huge amounts of pressure, and only under necessity, but that concepts in computer science that were once au fait become outdated and outmoded as the science/art form evolves. Developers are more mobile than they used to be - a lot of people will change jobs every two to three years, and development processes have changed both to enable that kind of mobility and to protect the product from the rate of change. This isn't a weakness - new blood, new ideas, new directions keep products fresh and young, and introduce changes into a product that are vital to keep those products alive.
Knuth couldn't be more wrong - but he's totally capable of building an ivory tall enough that he never needs to look out the window. Let's face, it folks. He's Sauron, not Gandalf, a once great man who went astray. His products are stable, sure - they're made of granite, immovable objects. The rest of the world, however, remains the irresistable force, and that force is change.
