On my last day in San Francisco I visited Jans Aasman at the Franz.com labs. They have a very nice place on the 15th floor of the Oakland City Center building with a beautiful view across the bay!

I presented and demoed the personalized medical journal I've been working on for a while. We had a lively discussion about technology and marketing aspects of this service.

Jans showed me some of the projects he has been working on. One is a full text index on top of AllegroCache. Blazingly fast! We discussed aspects of data model evolution using CLOS and AllegroCache. He also showed me the new SQL interface for AllegroCache done by Intelligent Handbook in Belgium. The library produces query code on the fly and works for AllegroCache and Oracle using the same API.

Over lunch (happy birthday again Michael!) we also talked about implementing a revision control system on top of AllegroCache. I would like to revision the source code along with the object code, the test data, the test cases and binary patches. Jans told me that Joe Marshall is already working on something like that, though not on top of AllegroCache:

Joe Marshall's versioning system is written in Lispworks on top of a monotonic database that he wrote himself. It is fantastic...

Interesting how things sometimes just come together. Essentially I want something that is totally integrated with my development environment, so I can do diffs on everything and go back in time with the whole system, not only the source code. Annotations would be important too.

Recurring themes in presentations and discussion in the hallway during ILC-2005:

• Keep the source code with the binary and version both
• Annotate code and define execution environments
• Make patching live systems reversible and keep track of all changes
• Code comprehension. Create programs that understand programs. Maybe teach CYC about programs and programming and let it read it's own code
• Code verification and correctness proofs

I'm totally convinced that, if we don't get at least engineering AI (as defined by Drexler) or usable program verification/proofs, the complexity of our systems will eventually make maintenance impossible.

I picked up some books while browsing book stores in San Francisco:

• FAB - The coming revolution on your desktop - From personal computers to personal fabrication by Neil Gershenfeld
• Marooned in Realtime by Vernor Vinge.
• Einstein's 1912 Manuscript on the Special Theory of Relativity. This is a facsimile reproduction of the original manuscript
• A Short History of Progress by Ronald Wright
• Nanofuture by J. Storrs Hall
• Stalking the Riemann Hypothesis by Dan Rockmore
• A World Without Time - The Forgotten Legacy of Goedel and Einstein by Palle Yourgrau
• Incompleteness - The Proof and Paradox of Kurt Goedel by Rebecca Goldstein
• The World's 20 Greatest Unsolved Problems by John Vacca

Reading the above list I'm ready to agree with Erik Naggum, who wrote about book stores in a posting to c.l.l.:

I just bought a book on a hunch the other day while trying to get out of the magic spell that bookstores seem to cast on me, robbing me of my free will and causing me to buy books for no good reason. I should have all my credit cards say "not valid in bookstores unless holder is accompanied by a responsible person".

I've updated the link to my US t-shirt shop. There were some requests for Polo shirts at ILC05. I've added a black Polo shirt with LISP lettering on the front side and a cons cell on the back. Suggestions for other designs are always welcome.

ILC'2005 was such a blast! Though ten hours of talks each day about theorem provers, correctness and other fancy stuff have taken their toll (image by Kevin Layer). I'm back in SFO giving my weary neurons a well deserved rest.

I was busy recording some of the talks, but I promised Carl Shapiro to wait for his approval before publishing any of the audio material. He will process the video recordings and get the approval from the speakers. I'll post a list of the talks I recorded shortly.

I got my copy of Anatomy of Lisp signed by John Allen! I'm allowed to bother him about a reprint, though he was not promising anything. We'll have to clarify the copyright situation first.

Wearing the John McCarthy shirt I approached John McCarthy and asked him for permission to keep producing the t-shirts. He accepted the t-shirt I brought along as a present (but unfortunately didn't wear it for his presentation). I have his talk on tape. Here's a quote regarding the Common Lisp standard (paraphrasing from memory):

If someone was to drop a bomb on this building, it would wipe out 50 percent of the Lisp community. That would probably be a good thing. It would allow Lisp to start over.

If you listen to only one podcast this month, choose the presentation by Daniel Hillis at the 2005 O'Reilly Emerging Technology Conference.

It's a place Bran Ferren and I started because we weren't having enough fun at Disney. --Daniel Hillis

I'm currently in SFO heading for ILC'05 in Stanford. That promises to be a lot of fun. So drop me a line via email or Skype (id: lispmeister) if you'd be interested in meeting over a beer. Of course that applies regardless of whether you're attending the conference or not.

I just found a very interesting recording of an interview with Colin Campbell, the Peak Oil expert.

Understanding depletion is simple. Think of an Irish pub. The glass starts full and ends empty. There are only so many more drinks to closing time. It's the same with oil. We have to find the bar before we can drink what's in it. -- Colin Campbell

Elmar recently inspired me to listen to the Tom Leykis Show. This is by far the funniest radio show I've heard for a long time! If you don't live in California you can download current recordings of the show at leykisonline.com or some hilarious clips from the Tom Leykis Show Unofficial Website.

Here's a taste of Leykis reading the Tom Leykis 101 FAQ. Or read what Wikipedia has to say about Tom.