Welcome to 2004! I’m looking for something new to do, as in job-hunting. [Update: I’ve accepted a job, starting March 15th. I’ll write about it then.]
It turns out I couldn’t write this up briefly, so here’s a table of contents:
On Leaving Antarctica · Well, I’m not really. I’m still a shareholder and go to Board meetings and will even carry the CTO card for a while longer. But they’ve got Barry Yates, who’s a master of explaining why visual interfaces are a good idea; Bill Schonbrun and Mike Sorensen, who are outstanding sales pros; and Dave Ashworth, who’s a slightly better programmer than me and a terrific project manager. These guys don’t need bright new visionary ideas, they need to focus on being a software sales machine; and I’d like to turn to something new.
In an Ideal World · What turns my crank is a chance to change the world; so my ideal job description could fit in five words: Leverage and a good salary.
But I can be more specific than that. Here’s what I’d really like to do: Over the last few months, I’ve written a long essay here at ongoing on the subject of search technology (“Slashdotted” on December 18th, thus read by tens of thousands); I concluded with a vision of what the engine, whose interim name is BRF for Basic Resource Finder, should look like. Now I want to build it. Obviously, lots of the pieces you’d need (maybe most of them) are already out there and ready to use. So it would be running and useful inside of a year.
In an ideal world, somebody hires me on the understanding that my main focus of work for at least a year would be BRF. Maybe not do-able; but we’ll see.
What I Can and Can’t Do · I have in-depth experience with lot of different kinds of infotech, including XML and databases and full-text search and user interfaces and the Web and typography and performance optimization and disk I/O. I’ve done system architectures and go-to-market strategies and technology evangelism, all with pretty good success; and written a lot of code.
I’ve been the CEO of three companies; while all are still here and one is public, I just don’t enjoy that role. But I can read balance sheets and prospectuses and licensing agreements competently.
I wrote a filesystem exerciser in 1990 or so named Bonnie that has been called a reasonable disk performance benchmark by no less than Mr. Torvalds. Now called Bonnie++ and maintained by others, it’s in fairly wide use by people who write Linux filesystems and build disk drives.
I know a lot of people in the business, which means that I know who to bother when I don’t know the answer to a question. I currently serve on the W3C TAG, to which I was appointed by Tim Berners-Lee.
Also, after twenty years in the business, I find that I still can’t not code: I’ve used dozens of languages and operating systems but am most comfy in C and Java and Perl and Elisp. If I had to highlight one thing: I tend to write code that runs fast, and be successful at tracking down and fixing performance problems in other people’s code.
I also can’t not write—English I mean, as opposed to code. I’m not a great writer but I’m pretty coherent and very fast. I’ve really enjoyed writing ongoing and can’t see stopping for the next decade unless something unforeseen comes up. As of now, ongoing has just over two thousand subscribers and gets fifteen or twenty thousand different visitors in a typical week; popular articles are read by many more. I think I can claim that my readership is an influential group.
I enjoy public speaking and have given many keynotes and I think I can claim that the audience usually goes away happy.
Finally, I have experience with standards processes—in particular at the W3C—and know how they work. I don’t want to be a full-time standards person, but would be happy to continue putting some work into this area.
What I Can’t Do · For family reasons, I can’t really live anywhere but Vancouver (Canada) or Melbourne (Australia); on the other hand I’m a seasoned and efficient traveler. I can’t go to work full-time much before mid-February 2004. I don’t enjoy project management, which requires struggling with the youthful programmer’s over-optimism that I’ve never managed to lose.
I would prefer not to do another startup just at the moment. When I left Open Text (which I’d co-founded) in 1996, I said I’d never do another one but in 1999 did, so I’m not going to say “never” again, but my short-term appetite for that isn’t high.
Who Should Hire Me, and Why · It seems unambitious just to shout “Will do technology for food” at the world; so I thought I’d pick a few candidate employers and explain why they should hire me. There are lots of other interesting organizations that I’d love to work for; the idea here is to provide some examples.
Why an Academic Institution Should Hire Me · While my list of peer-reviewed publications is short, it contains work that has been noticeably impactful on the practice of information technology. Furthermore, I think there is still lots of unexplored research territory in the land of text. First, we are basically nowhere in practically-useful natural language processing to support search. Second, doing a good job on XQuery and other structure is going to require serious new work in the basic-algorithms space. Third, for markup languages and document data we don’t have consensus around anything like the relational algebra or calculus, or the notion of normalization. Why not? There’s work here that seriously needs doing, I think.
With reference to search, I think that if something like BRF existed, it would provide a good sound infrastructural basis for future research.
And stepping outside the CompSci box, I think that in ongoing I’ve explored a class of interesting issues in the murky territory where technology, discourse, and economics meet; maybe some interdisciplinary institute would like to turn their eyes on this stuff.
Why Apple Should Hire Me · To start with, when something in OS X irritates me, I’d be bugging the right people at Apple internally instead of hanging my gripes out here in front of the whole world (snicker). More seriously, I think I’d be pretty effective externally both as an advocate of OS X and internally as a source of ideas for moving it along.
But the real upside for Apple would be BRF, which would run beautifully on OS X. At the current time the search capabilities in Apple’s tech-support space are really not very good and take little advantage of OS X’s strengths; BRF would.
So Apple should cut a bargain with the world and hire me (mostly) to work on BRF; it’ll be open-source, but they get first crack at it and OS X is the development platform.
Hey, Steve tried to hire me to work at NeXT in 1989, who cares if it takes me 15 years to say “yes”?
Why IBM Should Hire Me · To help IBM Global Services make money. Right now, GS customers are writing big checks to search-engine companies for software licenses for their Web deployments, both inward and outward facing. Instead, those customers should use BRF, which will be free and just as good, and write those checks to Global Services for deployment and integration services. The return on investment should be enormous.
Why an Information Company Should hire Me · By which I mean Thomson, Lexis-Nexis, that kind of operator. These people live and die by search, by text processing, by XML, by database optimization, by their Web presence. Sounds like more or less exactly what I’ve spent my career becoming an expert in.
Why an Intelligence Agency Should Hire Me · First of all, see the previous note on Information Companies, which are essentially the private-sector’s intelligence agencies.
In my life as a search-engine, content-management, and user-interface vendor, I have repeatedly had large globs of opaque data land on my desktop with instructions to find some value in it and make it useful to information consumers. I’ve been pretty successful; in fact two companies have been launched based partly on my ideas in this area.
Consulting? And the Short Term · I’d really like my next thing to be the right thing, so it may take some time to nail it down. In the interim, if you want to retain my services to give a speech or advise you on XML or search or optimization or whatever, why not? If you’re in the States you might have to do a bit of border-related paperwork. I’m pricey but I work fast. Lauren and I have an an incorporated company and it might be simpler for some organizations to deal through that.
In the long term, though, I’d like a long-term relationship with an employer I respect.
Track Record/Contacting Me · I have an online resumé which has all the basics. Email works best.