I touched on this in Principle 12 - All prototypes should be throwaway, or else you're just doing emergent design.
And build them for low effort/cost.
It's about quickly integrating feedback. If we spend a long time building the prototype, we'll not get information needed to make decisions. We're building the prototype to help drive decisions. If we can't make decisions, the product development stalls... It's not good.
Build 'em fast, hack 'em together - Get the info to be able to make the decision.
And the book says...
Sme thing. I do like one call out it has - Use any language. We need information to make a product decision; slam out a prototype in any language; whatever supports getting it into the hands of those who can provide information.
I think that's a really good addition to consider.