The lecture itself was interesting - he argues that the key to the OSS shift is not the licences but is best understood as an enabler. The real killer OSS apps are services provided on them - like Amazon and Google. Bind enabled the DNS (but the authors and maintainers never made any money from it). The alternative model is the "one ring to bind them all" of Microsoft ...
These, then are content delivery platforms - but they also benefit strongly from the contributions of users (Google through links and Amazon through reviews) - so maybe the value is in the content and the network afterall.
The advice to developers is to build things for maximum hackability and experimentation and collaboration and to avoid data-locking.. He wasn't too vocal on how the companies and developers that build such open systems actually benefit from the applications that are delivering value over their infrastructure.
The method I think would work is to leverage your closeness to the community that engages with and builds on your platform and then partner with, or buy them.
There where about 10 people at the lunch earlier - mostly research bods. I reminded Tim that we'd met before - at the first O'Reilly P2P conference when Marshall and I did the Napster Fabbing presentation. He laughed and remembered us hacking pieces of foam onstage with an axe. So that was cool. Although he asked me if I'd kept up in that area - and I had to admit that I hadn't - which is probably a mistake. So stay tuned for more on personal fabbers in the future.
He talked about Patents and everyone based SCO a little. I couldn't resist bringing up the BT hyperlink patent - which probably cut my career at BT fairly short - but it was too tempting. You could tell by the body language that that event was a seriously embarrassing episode for those in research - it still pains me to think that an organization could undermine its standing in a community on the very off chance of a gain - not a sensible cost/benefit/risk analysis.
I also asked him if he got annoyed when people criticize O'Reilly for advocating the Founders Creative Commons license (copyright lasts for only 14 years). The basic criticism is that it is easy for tech books - they go out of date quickly. Tim laughed and replied that he didn't mind - his industry does well when interesting things are happening - so why not help them to happen. Stirrer. He also pointed to the abundance of works out of print and argued that if the don't have commercial value they should be public domain.
So nothing hugely new there - but good to hear from the man ;)
I got a chance to put in a plug for our "(Crowston, Annabi and Howison, 2003) Defining Open Source Software Project Success">Defining Open Source Software Project Success paper - we where talking about what success means and it was nice to be able to mention that we'd done a full paper on it - he also thought it was cool that we'd turned a problem in getting a definition into an opportunity for a paper.
Currently listening to Human Fly from the album "As Heard On Radio Soulwax Part 2" by The Cramps
Posted by james at June 24, 2003 10:40 AM
| TrackBack