Comment Re:The biggest Mistake Today (Score 1) 467
I totally agree with you (well except for your last paragraph.)! I used to work with C/C++, and switched to working mostly with R (interpreted). Working with an interpreted language makes you lazy. You write an algorithm, and try to make it work by trial and error, because you think that it's almost there. (Just like in the binary search example he showed). But that's the wrong approach. You don't want your program to work, you want to work correctly. You can never make sure it works correctly by trial and error. You have to take a minute or an hour, think about what the code does, and why this unexpected behavior happens. When compiling takes 10 minutes, you sit and think before trying. When it is instantaneous I get lured by the trial and error approach.
Another problem is that to invent a new algorithm you have to think. You have to understand what an algorithm does, how it can be improved.
But a lot of his ideas are cool, too. In some cases (a debugger for example) immediate feedback is great.