The slice is not the whole: Folk Pedagogues and Open Source

Alan Kay sometimes quotes the Talmud: “We see things as we are, not as they are” I’ve been thinking about a corollary of that quote: “We see the slice, but the slice is not the whole.”  Alan’s quote is about seeing things in terms of ourselves, not as the thing is.  We also tend to look at our vicinity, and convince ourselves that our experience generalizes more broadly. This is one of the most prevalent misconceptions that I see computer science teachers make.

Ray Lister referred to most teachers as “folk pedagogues” in an earlier comment.  He talked about how teachers make decisions based on their past experience, without taking into account current situations.  I responded with a story about Davide Fossati who is studying how computer science professors decide to change their curriculum.  Decisions are made about changing curriculum, changing languages, and changing approaches based on “talking to a few students.”  What makes those students representative of the class?  Talking to a self-selected few is a terrible way of understanding how the class is a whole is doing.

Several people contacted me after my post about open source to tell me how their open source project cares about usability, or has lots of women on it.  I am sure that those people are right.  However, those are slices and don’t change the results upon which I was commenting. Mike Terry and James Howison are trying to understand open source as a culture, as a movement, as a whole.  They are doing careful sampling to get a picture of the whole.  They are finding few usability development practices. That’s not to say open source software is unusable. I use it — I’ve been upgrading by Ubuntu installation this weekend, and I still use Squeak regularly.  But those are my slices, and Mike and James are trying to understand a whole that’s larger than my experience, too.

I’m not immune to getting caught up in the slice.  I whined last week about the self-centeredness of my senior design students, and Briana Morrison joined me in her comments.  We’re both computing education researchers, so we know that our experiences of talking to a few students does not reflect the attitudes of the whole.  Ellen Zegura offered a great counterexample of students signing up to help others. In reality, the majority of students in my class did sign on to solve someone else’s problem. None of us is studying these classes, trying to get a clear over-all picture.  We’re talking about impressions.  We’re not making decisions to change things based on these impressions, which is a good thing.  We’re just noting slices.

I disagree with the title of Tom’s blog post: “The Open Source Community is us.”  We’re not.  We may be part of a community, but the community is much larger than us.  We’d be making a mistake if we used ourselves as examplars and generalized to the whole community based on ourselves and our projects.  I suspect that part of what Tom is suggesting, which is what I think Erik Engbrecht is suggesting too, is that we can change the community, that we can create projects that value usability and diversity.  Yes, we can create projects, but that probably won’t change the community.  That changes a slice, not the whole.  Maybe the values of the new project will spread throughout the community — maybe, maybe not.  How to change an all-volunteer social network is a fascinating question, but changing open source wasn’t the point of my post.

It’s dangerous to take one’s slice and think it’s the whole.  From the SIGCSE list and being at Georgia Tech (where the freshman class grew this year), it’s easy to believe that the undergraduate enrollment crisis has ended.  Then I talk to teachers at smaller colleges in Georgia, and I hear about plummeting enrollments and faculty being fired.  I live in Georgia where white politicians have believed that they were making decisions in everyone’s best interests, when they were really making decisions that were just in the best interest of their community, because they never spoke to the minority communities and never understand their needs.  It’s easy to confuse the slice for the whole.

That’s one advantage of mandatory standardized testing — it’s a clear view of the whole.  As a teacher, it’s easy to believe that the students who come up and talk to you represent the class.  It’s hard to ignore the results of a standardized test that all your students take.  How do we get university faculty to see the whole, to not be distracted by the slice?  Maybe that’s even harder than trying to change open source development.