> This is why you usually assume infinite memory in most problems, and why I consider limited-memory problems like in-place sorting to not be very interesting. Time limits space, so time is more interesting.
Yeah, I'm not sure I agree with this statement. Citation needed.
After this initial O(M) time and space cost, you do additional operations which only take up time, not space, so the claim Time >= Space holds here as well.
The inverse however has very few implications. You don't need much space to get into huge time complexities. Something like prime factorization has very little space requirements, but becomes very time-expensive very quickly.
Though I'm kind of disappointed that this article doesn't touch on the practically weird parts of complexity theory. For example how at times, the asymptotically better algorithm isn't used in practive, because the preprocessing only barely starts to break even with an asymptotically worse algorithm at input sizes of millions and billions, while practical problems are at input sizes of 10k - 100k.
