One thing not emphasized enough in this interview: SQLite is public domain. So was the web, which also won against its better funded licensed competitors at the time.
The lesson is clear: if you want to win in the long run, you need not just great skills and tech, but you have to go public domain. Or to put it bluntly: #LicensesAreForLosers.
I think I've heard SQLite creator Richard D Hipp say in the past that he thinks using the public domain declaration rather than a regular open source license was a mistake.
Public domain raises all sorts of challenges for potential adopters of the software that aren't an issue with a more deliberately designed license.
UPDATE: I misremembered this. He does talk about some of the surprise challenges in this interview, but does not go as far as saying that he regretted it: https://changelog.com/podcast/201#transcript-215
Some countries don't have Public Domain, and that can be a problem. I didn't know that. SQLite's response (from that transcript):
""" “We’ll sell you a license for SQLite.”
We do our best to talk them out of it and explain they don’t need this, but for a lot of people it’s cheaper to pay the fee and get the license than it is to convince their lawyers that they don’t need one.
"""
Are you talking about CERN's decision to put their web client and server source code into the public domain in 1993? It seems like a stretch - to me - to attribute the success of the web to that decision.
It was probably a big factor. Gopher was a real competitor initially but the University of Minnesota which owned the IP started trying to charge license fees in Feb ‘93- I suspect CERN’s decision was a response to this as well as MOSAIC’s similar efforts around the same time.
You couldn't predict that CERN's protocol would win, but you COULD predict that a public domain protocol would win.
Proof:
For every pair {protocolX,protocolY} where functionality(protocolX) = functionality(protocolY) && isPublicDomain(protocolX) == true && isPublicDomain(protocolY) == false, then speedAndUtility(protocolX) >> speedAndUtility(protocolY).
The lesson is clear: if you want to win in the long run, you need not just great skills and tech, but you have to go public domain. Or to put it bluntly: #LicensesAreForLosers.
https://breckyunits.com/how-the-public-domain-can-win.html