Changing of the guard
Recently Kurt Zeilenga, Founder of the OpenLDAP Project, informed me that he needed to step down as the technical leader of the Project. Today he appointed me as the Chief Architect. Woohoo...
In reality this isn't much of a change; Kurt had effectively been deferring the technical guidance of the Project to me for at least the past year already. So this just makes it official.
It's nice to be working on this Project in this role. It's nice to have taken what was at best a mediocre proof-of-concept code and turning it into the fastest, most scalable, most secure LDAP server in the world. Not because those achievements in themselves are so unusual. I had already done the same thing with the AppleShare server for Unix at my old company. What's nice is that this code is completely out there in the open - anyone in the world can check and verify that my claims are true. My AppleShare server was the first to support TCP/IP, even before Apple's own server, but besides a few engineers inside Apple and a few customers, nobody ever knew about it. It was also the fastest and most scalable AppleShare server in the world, easily handling tens of thousands of clients when even Apple's best server could only support a hundred or so, with response times two orders of magnitude faster than anything else.
What's nice is that I can happily tinker away at what I enjoy, being the best in the world at what I do, and now I know the world is seeing it. The world is watching, noticing.
And this *is* what I do - create the most efficient designs in the world. Real designs, that quickly become real implementations, not just fancy theories that never pan out. It's what I've always done, and there's never been anyone who could even come close. Nobody can design as comprehensively, nobody can comprehend as quickly, nobody can implement as quickly, and nobody's implementations can run as quickly.
I remember once, back in Michigan, saying to a friend "just once, I'd like to be in a singing group where I'm the *weakest* member." I remember thinking how nice it would be to meet a fiddle player who could keep up with me. It hasn't happened. It won't happen. Nobody else's brain runs near my speed, that's just reality.
A side lesson from all of this, a point I've made many times before - don't work in secret. Closed-source proprietary code is closed because it has something to hide. Take pride in your work, excise the warts. Make everything you do something to be proud of, something to show the world, and show it. Otherwise, your work is likely to disappear into obscurity, like my old AppleShare server, and then it will have all just been for nothing.
In reality this isn't much of a change; Kurt had effectively been deferring the technical guidance of the Project to me for at least the past year already. So this just makes it official.
It's nice to be working on this Project in this role. It's nice to have taken what was at best a mediocre proof-of-concept code and turning it into the fastest, most scalable, most secure LDAP server in the world. Not because those achievements in themselves are so unusual. I had already done the same thing with the AppleShare server for Unix at my old company. What's nice is that this code is completely out there in the open - anyone in the world can check and verify that my claims are true. My AppleShare server was the first to support TCP/IP, even before Apple's own server, but besides a few engineers inside Apple and a few customers, nobody ever knew about it. It was also the fastest and most scalable AppleShare server in the world, easily handling tens of thousands of clients when even Apple's best server could only support a hundred or so, with response times two orders of magnitude faster than anything else.
What's nice is that I can happily tinker away at what I enjoy, being the best in the world at what I do, and now I know the world is seeing it. The world is watching, noticing.
And this *is* what I do - create the most efficient designs in the world. Real designs, that quickly become real implementations, not just fancy theories that never pan out. It's what I've always done, and there's never been anyone who could even come close. Nobody can design as comprehensively, nobody can comprehend as quickly, nobody can implement as quickly, and nobody's implementations can run as quickly.
I remember once, back in Michigan, saying to a friend "just once, I'd like to be in a singing group where I'm the *weakest* member." I remember thinking how nice it would be to meet a fiddle player who could keep up with me. It hasn't happened. It won't happen. Nobody else's brain runs near my speed, that's just reality.
A side lesson from all of this, a point I've made many times before - don't work in secret. Closed-source proprietary code is closed because it has something to hide. Take pride in your work, excise the warts. Make everything you do something to be proud of, something to show the world, and show it. Otherwise, your work is likely to disappear into obscurity, like my old AppleShare server, and then it will have all just been for nothing.