The Cathedral and the Bazaar

I was pleasantly surprised by how much is in Eric Raymond’s now classic The Cathedral & The Bazaar. Thinking I was going to get one essay on Linux and the open-source movement, instead I was treated to four! I’m surprised how unaware I was of this book, given that I used various versions of Unix back in the day before Linux came together. This book brings back memories of earlier days, and although I was never a bona fide hacker I found the history fascinating. I was actually pointed to this book from reading about management, creativity and disruptive new tech. The book that led me to this gem was Scott Berkun’s The Year Without Pants, which chronicles his time with Wordpress.

The premise of the first essay, which also takes the title of the book, is to contrast two approaches one could use in designing a computing platform. The old-school cathedral-building approach uses highly controlled project management – an ethos that still reigns in many I.T. departments. The bazaar approach, on the other hand, is more like the Wild West. Can a bazaar-like approach build a coherent system that is economically viable, works well, and won’t suffer from the “tragedy of the commons”? It is clear that at least in the tech world, Linux is a marquee example that the bazaar approach can work, although other earlier open-source attempts failed to gain enough momentum (or ran into problems).

The second essay in the book titled “Homesteading the Noosphere” was even more interesting than the first. The author delves into the social workings of why and how a bazaar-style open-source approach works. Clearly it doesn’t work in all cases, and the author outlines the advantages and disadvantages of both closed and open-source approaches. My first thought when reading the chapter was to think about hacker culture from a craftsmanship model (since I recently read Shop Class as Soulcraft and blogged about it here). But Eric Raymond had considered it already, and had a better explanation on hand – a gift culture with quirks unique to the hacker world. I had previously stereotyped hackers as narrowly interested in hacking, but the subculture turns out to be fascinating – and the author has a coherent thesis analyzing anthropologically his own “tribe”. He anticipates what we’re seeing in the Big Data and Sharing movements of the last several years.

Raymond’s book has gotten me thinking about whether a bazaar approach could work in higher education. From a knowledge/information point of view we see parallels with Wikipedia although there seem to be a few differences in the control system and culture (from the little I know). A ranking system of helpful educational videos or books (e.g. YouTube or Amazon) makes use of crowd-sourcing, but not in the same way that hackers contribute to building and maintaining source code, distributions and writing patches. The videos and books aren’t crowd-edited – certainly not any good ones.

As a college professor I have control over my own materials and how I will teach my classes. (I suppose that if I was teaching younger children in a public school I may be constrained by having to deliver a specific curriculum possibly in a particular way.) While I would search the World Wide Web for helpful illustrations, animations, strategies, source-material, I am still the final arbiter for my class. I choose how to arrange the materials and activities, and they are tailored to who I know to be my “audience” (albeit, one that interacts with me as the instructor). However, I can only do so much to customize the learning process for each of my students. (It would work great if I only had 3 students per class, but this would not be an economical model!)

This is where adaptive learning technology may prove to be superior – customization for every student. I mentioned ALEKS and Knewton in a previous post and how these systems may put us teachers out of a job, even as we contribute to it – at least if I’m feeling dystopian. At the moment, these systems are proprietary (closed-source) and the behemoth publishers (McGraw-Hill and Pearson) are banking on holding on to the keys to make their profits. This sounds strangely reminiscent of the cathedral-building approach in the tech world. Could the bazaar work here? Is there a community of hacker-professors out there in critical mass? Or can a platform be built that allows non-hacker professors to collaboratively leverage an open-source adaptive system built by hackers?

Maybe. Raymond’s book suggests the ingredients of the culture that might sustain such an endeavor. There certainly are some parallels between academic and hacker culture, but also distinct differences. The lack of a clear objective standard in higher education of what should take place to best enhance learning perhaps contrasts to clear and efficient code that works smoothly – something hackers of different stripes could probably agree on. The Minerva Project is banking that they might learn something by collecting lots of data monitoring both students and instructors in their online-only courses (they’re sort of the anti-MOOC). Could a Big Data feedback approach help coalesce educational mechanics?

If I think up a Killer App for all of this, should I post it and sound a clarion call for the open-sourcers to gather around? Or go start my own company? The possibilities!