shock collars

New Iron's Janelle Klein hosted a roundtable discussion at Lean Software Austin on Monday. The topic was on mistake-proofing.

The discussion was framed such that we were to talk about what are and how to implement shock collars in lieu of Tupperware's Shape-o-balls. Let me explain the difference. Janelle envisioned two forms of mistake-proofing while putting discussion emphasis on the latter of the two. Janelle's two varieties of mistake proofing:

1. Shape-o-balls only allow you to put pieces in the holes made to accommodate those pieces. This is simply an analogy for software that is restrictive in a healthy way. (i.e. well architected)
   
2. Shock Collars in contrast don't prevent you from making a mistake. They signal you when you are about to stray outside of the invisible fence of good practice letting you know that you're about to be walking into PAIN, PAIN, PAIN. Now what do these look like? Are these tests asserting how things should be? The group brought this up and then quickly dismissed it. One could stray into pain and have tests reinforcing the course of action the entire time. With tests ruled out it quickly seemed like shock collars were in fact not of an engineering solution but instead of environment and culture and hence naturally more fuzzy in definition. Janelle seemed to suggest that a good culture could serve as a shock collar the same way a good upbringing might discourage one to lie/cheat/steal.

The event started at 6:30 in the evening and I left a little after 9. The discussion lasted later into the night, but by the time I left there had been much productive discussion yet not a single example of a shock collar offered. The group wrestled with the topic. Many suggested that cultural solutions lay in crafting an environment that empowered persons to do well in contrast to shocking them before they do bad. One attendee suggested: "There is nothing you can do to make a plant grow as a gardener. All you can do is create an environment where a plant can grow." Scott Bellware who runs Lean Software Austin suggested that geeks will respond poorly to negative enforcement (compared to positive enforcement) as so many have historically fit the stereotype of being that guy bullied by the jocks in high school. Another attendee suggested in contrast that geeks will respond negatively to negative enforcement due to histories of being the smart guys who usually come out on top in academic completion and are used to winning and having their ways.

The thing I took away from the meeting is that we shouldn't try to be making others wear shock collars, and yet, I can think of so many times when I've been ahead for telling myself "Com'on stupid, you need to be doing better." Yet, there is a big difference between browbeating yourself and either browbeating others or having them browbeat you. It was an interesting talk. It was just understandably hard for the group to put its finger on a consensus for recommending any concrete shock collar-based solutions. Steering our culture seems challenging stuff... worthy of debate!