Arvind Narayanan (arvindn) wrote,
Arvind Narayanan

  • Mood:

Innovation on the Internet -- some lessons

ESR is exultant that the Cathedral-Bazaar metaphor is being used in the MSM by people with no clue about its original context, proving that the idea has had much bigger success than could have been envisioned. In my own small way, I can say I know that feeling.

Two years ago I invented a way to make chess diagrams on Wikipedia as an HTML table using the existing wiki markup. Previous methods were very cumbersome, such as making screenshots of chess software. A solution using "wikitex" had long been promised, but never materialized. As such, my proposal was enthusiastically debated by the community and began to see some use; however soon after that I cut back my Wikipedia contributions to the occasional paragraph or two and so didn't keep track of what was happening on the chess diagram front.

A little while back I was pleasantly surprised to see that the seed I planted had grown into a giant oak tree, if you will pardon the cliche. In typical bazaar fashion, the proposal accrued improvements one after the other: someone made it possible for the diagram to be of different sizes, the colors of the images and the stylesheet to be used were voted upon and standardized, my cumbersome input format was simplified, automated converters were writen from other chess formats, the images were uploaded to the "wikimedia commons" and standardized accross all Wikipedias, and the format was extended to incorporate other board games like Go. When I last checked, there were over a thousand chess diagrams in the English Wikipedia alone, and I'm guessing at least that number in other languages, and an unknown number of diagrams of other board games. And growing rapidly. There have even been correspondence chess tournaments on Wikipedia. All told, that's at least a thousand man hours of work saved by the format.

I'm not alone here -- all over the Internet, people have found that little ideas that they had went on to have lives of their own and have an impact on the way things are done. What lessons can we learn from this?

The first is that a lot of new ideas are overlooked for so long because they are so obvious no one thought of them. Mohan and I once noticed, when we were in the math olympiad, that one of our fellow competitors was successful because he was good at Noticing the Obvious. I started to train myself to do this, and when I competed the following year the results were way beyond my expectation. In general, it is far more important for anyone who wants to be an innovator to Notice the Obvious than to come up with Brilliant New Ideas.

A related lesson is that the Right Way to do things is often the one that works. If there is an Easy Way and a Right Way, always try the Easy Way first. You might discover that it has unexpected advantages purely by virtue of its simplicity.

Thirdly, one good idea with a working implementation is better than ten great ideas on paper. I wrote about some of the ideas now found in Nautilus/GNOME about two and a half years back, but I didn't produce any code, so they were never implemented until someone else thought of them. Developers are typically inundated with ideas about how to revolutionize their software, so put your money where your mouth is if you want to make a difference.

Finally, your ideas have a far better chance of success when they are free. If you want monetary compensation for your ideas you're better off in a big company where your ideas can be part of a bigger product. I like to think that the days of a genius inventor making millions off of a single great idea on the Internet died with the bust and aren't coming back. This is the way it should be -- it's sort of like the reverse of insurance.

Since I talked about ESR and Wikipedia in the same essay, I can't resist pointing out that ESR doesn't "get" Wikipedia -- he once posted on his article's talk page under the heading "Response by the the target of the article", displaying an "us vs. them" mentality, as if Wikipedians were a group out to slander him. ESR not getting Wikipedia is deliciously ironic on multiple levels -- Wikipedia would probably not have come into being if not for ESR's paper on the bazaar and his evangelism of the concept, and also because he's so fond of pointing out who "gets" open source and who doesn't!
Tags: collaboration, hack, innovation, internet, software, technology, wikipedia
  • Post a new comment


    Comments allowed for friends only

    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded