Also his Riffle example is thinking about the problem the wrong way round. It looks like he tried to turn code into english, when it should be the other way around. Here is an actual english version:
1. Start with a purple square
2. Clone the square, rotate it by 0.1 radians and scale it to fit the previous square
3. Also alternate the colour between purple and yellow.
4. repeat clone process until square is just a dot in the middle of the screen
Note how easy it was to understand what I meant, and yet a computer would fail at so many of the instructions: scale how? how to test if it fits? alternate whose colour? repeat what steps? 'clone process'? clone same square or newest created square? what defines a dot? screen? middle?All these ambiguities we can solve, but a computer cannot hence why our programming languages are so specific to the point of cognitive overload. Also none of his later examples show to me that Wolfram is better than English, unless of course you use wolfram daily (as he 'the creator' likely does/did), in which case you're brain is already optimized to read it...
This isn't even mentioning that language is only part of how we communicate. We can often leave so much out because our visual system, body movements, previous experiences, memory and other senses fill in the rest.