Thursday, July 16, 2015

  • Intro: Last build season, we discussed an ideal strategy.  This strategy had a number of leap-of-faith assumptions about how long things would take and what we could do.  Try to describe our ideal strategy, including those assumed numbers that helped us calculate how many points we could earn in a round.
Our ideal strategy was to build large stacks at the feeder and score them, we figured we could score at least 40 points a game, if we were able to get the bottom tote to land flat.  We assumed these numbers based on roughly how long it would take to lift a tote using our protobot to test.

  • Intro: At what point in the build season did we find out how accurate our assumptions really were?  Was it possible to accelerate this?  Could the assumptions be broken down into small experiments?  If so, how?
I would say we found out about half way through the build season, this definitely could have been faster.  If we had built a full protobot on the first/second week we would have been able to realize many of the important design decisions that it took us a long time to find.  We could have broken some of the assumptions down which we did do, however I think in retrospect that we should have done these small experiments sooner, if we had built a protobot quickly we could have tested the extensions idea/center of gravity problem.

  • Intro: Explain the build-measure-learn feedback loop.  What is the purpose of this loop?  Why is it a loop?
In the bml (build measure learn) feedback loop, you start by building a minimum viable product, testing it, and seeing where you can improve it.  Then you are able to improve it and make a better build, test it, and find where you can improve, making a perpetual loop of improving your product.  The purpose is to be able to build better and  better product by learning what customers want/what does and doesn't work.  It's a loop because you can improve pretty much everything, and you can keep using this process to get a better and better product.

  • Going back to some of the leap-of-faith assumptions, what would some minimum viable products (MVPs) look like that could validate these assumptions?  What would we measure with the MVPs?  Think specific to last season's game.
A MVP in First most likely will look like a protobot, being able to test anything that isn't fact, whether its being able to straddle the scoring platform or if we will tip over if we try to drive straight onto it.  If we had built our protobot sooner we could have found many design changes that would improve our real bot and have fixed them.  We did that too an extent but I think we should have done this sooner as it takes a lot of the unknown out of design.

  • Explain the words "pivot" and "persevere" in the context of a team's way of doing things.
For us to pivot is to realize that a different design choice would be better than what we are currently doing and use that idea.  Persevering for us would be deciding although something is hard and there are other options we continue to push forward because it is for the best.

No comments:

Post a Comment