My MVP for a transportation device would not be to give the customer a wheel, then two connected wheels, then a car body, and then a car. This type of large product design is unacceptable in modern product creation, where an end product is delivered after a long time.

If the requirement was to have a transportation device, then a skateboard is relatively minimal because it transports one person and it steers (else it would be shit on a non-straight stretch).

I don't necessarily agree that customers hate MVP; just sometimes, MVP doesn't provide a lovable experience. Depending on what is being built, if it answers a real problem where nothing else exists, then customers are willing to accept a period of transition where the product improves.

Good article and insight into building things better. MVP is a very engineering approach, where the technologies don't give a fuck about first impression and loyalty. If you're building a competitive product, you must make it good from the start, or you're unlikely to retain users when an already competitive and stable product exists.