Notes from Ruby/Rails group, 15-Feb-2010

[Ted Roche]

Despite being President's Day, February 15th's meeting of the NH Ruby 
group got 14 attendees. Held as usual at RMC Research in Portsmouth and 
hosted by Tim, a good time was had by all.

There was a round of introductions so everyone got to know each other. 
Announcements including the NHTweetup calendar 
(http://www.google.com/calendar/embed?src=cnIybTF0bm12ZHRoaWhtMzA1OWVsdTBrNW9AZ3JvdXAuY2FsZW5kYXIuZ29vZ2xlLmNvbQ), 
and the Rails Camp NE http://railscamps.com/#us_ri_march_2010. Rails 
Camp is only a month away and down to a few last seats - don't miss it!

Brian presented "Object Oriented Programming." Slides: 
http://www.slideshare.net/bturnbull/oop-intro-in-ruby-for-nhruby-feb-2010. 
The presentation was a good high-level overview of OOP, as seen from the 
Ruby perspective. Brian noted ways in which Ruby differed from C++ and 
Java. There was some very good discussion and some teasers of future 
presentations: Mixins were thought too complex to try to squeeze into 
this presentation, and Brian is developing another talk, "The Complete 
Class," which will include a discussion of all of the methods, 
properties and attributes a class should have to encapsulate best 
practices. I'm looking forward to both of those discussions.

Brian finished with two puzzles. Having covered the principles of OOP, 
Brian handed us an interesting problem to code: a 3x3 sliding tile 
puzzle. It's great to see actual code being developed at the meetings, 
and it was a good challenge in that we all understood the problem 
domain. We broke into small groups and started modeling the problem. At 
the end of the evening Brian presented his model and code: 
http://github.com/bturnbull/tile_puzzle

The bonus  challenge involved taking Brian's model and solving a 
particular puzzle in the fewest steps possible. I worked with a team 
that took this one on, repeating the classic steps of invoking 
recursion, discovering the limits of Ruby on the Mac for stack overflows 
;) and failing to properly store and retrieve states and scopes diving 
in and out of the recursion. While we didn't finish the solution, we had 
some great discussions on the various ways to solve the problem and all 
learned more about working with Ruby, which after all is the point. 
Brian also posted his solution to finding the shortest steps to solving 
this problem: http://github.com/bturnbull/tile_puzzle_solver/

Thanks to Tim and RMC Research for hosting us in their fine facilities, 
to Brian for organizing, pizza and the presentation, and to all for 
attending and participating!

-- 

Ted Roche
Ted Roche&  Associates, LLC
http://www.tedroche.com