brokenco.de/_posts/2012-02-09-fosdem-smalltalk...

2.6 KiB

layout title tags
post FOSDEM: Smalltalk Pairing
smalltalk
fosdem
fosdem2012
tripdiary

As I previously mentioned, I've been learning Smalltalk lately, in an attempt the understand the language that inspired two of my other favorite languages: Objective-C and Ruby.

This past weekend at FOSDEM, I was fortunate enough to attend a few sessions in the "Smalltalk devroom", which from my understanding made its debut appearance at this year's conference. I'll save my thoughts on Amber.js for another time, but in this post I wanted to talk about the Smalltalk Workshop which was held at the end of the day.

One of my major stumbling blocks with Smalltalk has been general unfamiliarity with the development environment, the workshop was the perfect opportunity to resolve some of this. The structure was to pair one experienced Smalltalker with one noob, and for both to work through a pre-planned exercise with an existing image and application set up.

Pairing on Smalltalk
Haxx haxx haxx (Photo courtesy of Adriaan van Os)

While my partner (Norbert) and I did not complete all the exercises, we did spend a good amount of time discussing and working through the "Smalltalk way", or at least Norbert's Smalltalk way, of solving particular problems, refactoring and method structure.

One of the things that struck me as we worked through the exercises was how small the ideal method body is for most things in Smalltalk. I don't think there was a single method that was longer than 10 lines, except for some test methods which certainly had a bad code smell.

Below is a sample of one of the longest methods I wrote in the entire workshop:

vote: aVote for: aUser
  author ~= aUser ifTrue: [
    self votes
      detect: [ :vote | vote author = aUser and: [ vote direction = aVote direction]]
      ifNone: [ self votes add: aVote ]]

I'm quite happy with the way things turned out, while I didn't anything I didn't know already about Seaside, I did learn a lot about using the development environment effectively and test-driven development "the Smalltalk way" which as it turns out is quite impressive.

I'm looking forward to learning more about the ways of the Pharo, maybe next year I'll count as one of the experienced Smalltalkers.