<Rolls 1DFibonacci in his mind>
It's a three. Have fun!
---
Unless it's something I've actually implemented before - this is pretty much what I do these days. You can sprint plan all you like - but in my arena (front-end) there are so many things that can mess up your best laid implementation plans. So in such a scenario - I agree with the article. Just get on with it. Do a quick, rough plan and get to coding up a rough PoC that validates the approach as quickly as possible.
Folks get anxious when there is implementation uncertainty. And I've learned that voicing this uncertainty just erodes their trust. If I do discover a problem with my implementation that is going to cost some extra time - then they generally have no problem with that as long as I communicate the discovery promptly along with a confident appraisal of the effectiveness of my new approach.
Folks are fine with mistakes that happened in the past. They only fear the possibility of mistakes in the future.