Post-P2P Conference Comments from HailStorm Architect Mark Lucovsky

Email.Email weblog link
Blog this.Blog this

Tim O'Reilly
Nov. 12, 2001 08:39 AM

Atom feed for this author. RSS 1.0 feed for this author. RSS 2.0 feed for this author.

Just got this mail from Mark Lucovsky after the P2P Conference, and asked him if I could share it in my weblog. I got to have dinner with Mark after his keynote, and found him honest, thoughtful, and engaging--not to mention someone with a deep UNIX background. His followup email shows those same characteristics--not at all the picture of Microsoft employees that many open source developers carry around. I'm really glad I've been reaching out to Microsoft, and that they're reaching out to the developer communities O'Reilly represents. As Rael Dornfest likes to say, a lot of interesting things can happen once we get past the marketing firewalls on both sides.

Rael, Tim, Thank you for allowing me to participate in the p2p conference this year, and for inviting me to dinner Monday evening. I had a wonderful time during my short visit.

I enjoyed having the opportunity to see things from a perspective that isn't always available to me. I found my time in the speaker room with Brandon Wiley and Stephen Hazel, two Open Source P2P hackers working on their Tristero project, particularly beneficial. I have their cards in my luggage and hope to get to know them both better. What I was trying to do is get them psyched to spin up an open source client project for caching and replicating HailStorm services on a wide variety of platforms. You can get to HailStorm from any platform with the Open Access model, but you can do even more with some client smarts. I would really like it to be a grass roots thing though, not something that Microsoft has to start. I guarantee my support and the support of my technical and program management teams though.

I can't say I understand why so many are so turned off by Microsoft and PowerPoint. I like using the best possible tool for the job at hand. To me, that means on a laptop, I'm going to run XP. When I do a presentation, I'm going to use PowerPoint. If I need to use associative arrays, heavy regular expressions, I'm going to use Perl. If I need to produce HTML from XML, or transform XML from one shape to another, I am going to use XSL. Just my style I guess. I got a kick out of watching people not use the best tools. Lots of fun joking around with them as they struggled to do, what for me, comes easy.

Just so you know, I really do love UNIX. First studied UNIX internals from the Lions books. Then deeper study from BSD releases, System III, System V, etc. Pretty much memorized how everything worked back in those good old days. Spent many years in the kernel myself. In the early 80's I worked at one of those mini-super startups. We had an exotic array processor front-ended by a sun1/sun2 box. I did all the kernel work. Had to make it work in asymmetric MP mode to support front-end/backend system calls. Built the memory management, context switching and state management for the array processor. Extended the image format and loader to support dynamic microcode segments, etc. When NFS was first released, I took all my kernel changes and diffed them into the sun2 code, and got NFS running on our machine. After this company flamed out, I did similar work at Vitesse Electronics where we were building an even more massively parallel pipeline processor based on SunOs. I was the kernel guy and also the admin for all of our workstations and servers. During the transition from DEC to Microsoft, Dave Cutler and I toyed with the idea of a startup where we were going to build a system and use Mach as the kernel. Applied for and got Mach source. I pored through the system learning all that I could.

I understand why Linux/FreeBSD is exciting to these guys. I know it is deeper than my simplified view, but I know what’s really exciting... Everyone wants to be a kernel guy. I see it all the time while interviewing, I'm excited about it, and so is every good computer science graduate. These systems offer them an opportunity to work in a kernel and experience this thrill. Having worked in commercial software at the kernel level for so many years, I know the kind of discipline that is needed at this level of the software stack and I have views that are impossible to agree with unless you have shared my experiences.

Don't mean to ramble on about this. I took a different path and would not change it for the world.

What's exciting to me is the energy level of your attendees. I want to find ways to harness that energy in a way that’s beneficial to the next generation of software consumers and would appreciate any feedback on ways we (Microsoft) can do a better job at doing so. I believe that projects like HailStorm, innovative p2p applications, and exotic uses of XML and XML web services provide us with a great sandbox. I am committed to keeping my platforms openly accessible so that all of these guys can play with my toys.

Thanks again for everything.


P.S. Mark couldn't remember the exact titles of the Lions books, but checked his bookshelf and sent them in a later message: Unix Operating System Source Code Level Six, and A Commentary on the Unix Operating System, both by John Lions, Department of Computer Science, University of New South Wales, June 1977. (For those who are interested, Peer-to-Peer Publishing, which was started by Dan Doernberg and Rachel Unkefer after they sold Computer Literacy Bookshops to republish out of print computing classics, issued the 6th edition of the Lions Commentary on UNIX in 1996, and has it for sale over the web.

Tim O'Reilly is the founder and CEO of O'Reilly Media, Inc., thought by many to be the best computer book publisher in the world. In addition to Foo Camps ("Friends of O'Reilly" Camps, which gave rise to the "un-conference" movement), O'Reilly Media also hosts conferences on technology topics, including the Web 2.0 Summit, the Web 2.0 Expo, the O'Reilly Open Source Convention, the Gov 2.0 Summit, and the Gov 2.0 Expo. Tim's blog, the O'Reilly Radar, "watches the alpha geeks" to determine emerging technology trends, and serves as a platform for advocacy about issues of importance to the technical community. Tim's long-term vision for his company is to change the world by spreading the knowledge of innovators. In addition to O'Reilly Media, Tim is a founder of Safari Books Online, a pioneering subscription service for accessing books online, and O'Reilly AlphaTech Ventures, an early-stage venture firm.