Absolutely, but I'll add this posit: What "viable" means is heavily context driven. If the goal is to learn about whether or not your specific take on things is valid, then what your MVP will need to look like will depend on the kind of market your pursuing, whether or not their are already existing players in your space, how radically different your approach is, etc.

IOW... if you wanted to launch an MVP of a "Facebook Killer" and the "secret sauce" was something, like, oh... some gamification thing... badges or accomplishments related to making new friends, let's say. That's enough like Facebook (or g+, etc.) that the expectation for a potential user would be based around Facebook... your "secret sauce" is only an incremental improvement, so you'd almost have to roll out something as feature rich and visually polished as Facebook, in order to find out if you can compete. That sucks, but one that's why there are both pluses and minuses to entering an established market with an incremental improvement. :-)

Now if your "secret sauce" is something world-changing and totally new, like say, "automatic Frombleglizzits when you Gwizztickum" (whatever that is), then you may ONLY need to build enough to make people understand what "automatic Frombleglizzits when you Gwizztickum" is, and then you can find out if that's compelling to them or not, in isolation. If it turns out it isn't, you would have wasted a tremendous amount of time building a complete Facebook clone just to find that out. You have more uncertainty in this case, but this is exactly where the Lean Startup approach helps... eliminating waste in the face of extreme uncertainty.

good news: no one knows who you are when you launch, so you have nothing to lose.

Exactly.

Unfortunately I think what we're seeing is the lean startup backlash where people use the terms but don't really take the time to understand what they mean.

Absolutely.