I think a copy of Sam Harris's Letter to a Christian Nation ought to be available, for free, to anyone who wants one; but especially to children. We probably can't do much to change the ideas of millions of religious adults, but we certainly ought to expose as many children as possible to the idea that their parents might just possibly be delusional lunatics when it comes to the idea of "god". Planting that idea in the minds of as many kids as possible is the one thing that I think might avert a future religious-ideology-driven apocalypse.

I just wanted to mention that my absolute favorite record of all time, without any real competition other than a couple of other albums in the same artist's ouvre, is The Dreaming, by Kate Bush.

Apparently a lot of people, even Kate's fans, don't particularly like that record. I'll admit that on first listen, it was a bit off-putting, but having heard it several hundred times now, it would definitely be the record I'd choose to take with me to a desert island, if I could only choose one.

On another day, I might choose Kate's Hounds of Love or Aerial as my favorite record, but today (and most days) it's The Dreaming.

While I'm on Spolsky, who thinks that CS programs ought to teach students things that will actually open their minds and broaden their ideas of computation, I'm reminded of one mind-broadening experience I had recently: Chapter 3 of SICP: http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-24.html#%_sec_3.5
This chapter discusses "streams", which are essentially just lazy lists. The notion that any recursive algorithm can be represented as a stream defined in terms of itself took a little getting used to. The section on sequence accelerators blew my mind.

I think SICP is a great book, in part because of stuff like streams, but also because of its whole organization: it takes the reader from a very high-level model of computation (the Scheme language), and ends up at the low level of simulation of a virtual machine (the explicit-control evaluator). This is quite different from a lot of CS texts which begin by talking about machine organization, and end up discussing some half-assed high-level language like Pascal.

I got every question right on Joel Spolsky's freshman UPenn CS exam: http://www.joelonsoftware.com/articles/TestYourself.html.

Now why didn't Google want to hire me? Oh yeah, my brain turned to oatmeal the minute I walked into the interview room :-( It's been six months, and I still have horrible flashbacks to that day. Beautiful corporate campus, amazing lunch... brain-dead interviewee. For example, when asked to write a function to remove duplicates from a list, I gave the "obvious [O(n^2)] recursive solution" -- what the heck was I thinking? Ever heard of hashtables, dufus?

Oh well, maybe I'll try again this year.

I had a very interesting experience last night: just for the hell of it, I decided to try writing a quine (http://en.wikipedia.org/wiki/Quine) in Scheme. It took just a few minutes for me to think of a reasonable approach, and it worked the first time under DrScheme (http://www.plt-scheme.org/):

((lambda (x) (print `(,x ',x))) '(lambda (x) (print `(,x ',x))))

This week I've engaged in a silly thread about quining on comp.lang.forth (archived here http://groups.google.com/group/comp.lang.forth/browse_thread/thread/ba6fe1a735c3dca6/8bee19a004b7cbcd#8bee19a004b7cbcd), in which I argue pointlessly with another poster about the exact definition of "quine". I think what bothers my correspondent about my "context-specific" quines is that it is the context, rather than the quine itself, which contains the knowledge about how to reproduce the quine's source. My position was that that is true of all quines, to the extent that they depend on external tools (such as a runtime library) in order to run. However, in a "context-specific" quine, the self-referentiality is trivial, since it is built in to the runtime environment. The Scheme quine above does not share that characteristic, although it's hard to see exactly where the dividing line is.

For Christ's sake, *when* are we going to impeach these people?

http://www.amconmag.com/2006/2006_10_09/print/reviewprint.html

I did it. It took me two months, but I finally finished _Goedel, Escher, Bach_. It was a lot of fun, and it took me to some interesting places. It's also the first non-fiction book I've actually finished in a really long time. Probably years.

I've decided that from now on (that is, until there's a good reason to either change the plan or admit that it isn't going to work), I'm going to try to finish one non-fiction book every week. I can easily gobble up two or three SF novels in a week, but non-fiction presents something of an obstacle, so this is going to be a real challenge.

Fortunately, I've got a *big* pile of non-fiction books that I've started, but never finished. I get further along in some than in others, and if I choose my victims carefully, I can probably chalk up a finished book-a-week for many weeks without even starting on new material. But that would feel like cheating, so I'm going to start with something I have never read before. If I can find such a thing in my house, anyway. Maybe _Introduction to Expert Systems_ by Peter Jackson -- it was actually the textbook for an ES class I had in grad school, but I don't think I ever actually cracked it open, and even if I did, I don't remember doing so. And I have an ES project in mind, so it would be apropos.

So anyway, if you're one of the possibly two other people on this planet that reads this blog, you can expect a weekly update on my reading material. I'll even try to make some reasonably intelligent comments about it. Starting with _GEB_:

_GEB_ is a hard book to absorb. It's complex, intricate, and contains an enticing soup of ideas that have simmered for years in the author's mind. I'm sure I didn't get all the connections (sometimes not even the ones he explicitly pointed out). But I got a lot of it.

One of the major themes is the interaction between different levels of organization in a system. Hofstadter believes that such interactions -- especially those that allow a system to reflect on or observe itself -- are the basis of consciousness. I agree; in fact, I had independently toyed with some of the same ideas before encountering _GEB_, though Hofstadter has taken the notion much further.

In the last couple of chapters of the book, he discusses a very important point: that physical reality is the base case against which all apparent infinite regresses fetch up, in (automated) logic, mathematics, biology, and so on. For example, we can question whether our reasoning is valid when we assert that ((A -> B) & A) -> B -- do we need a higher-level axiom, (((A->B) & A) -> B) & A) -> B ? In principle, yes; in reality, no, because that's the way logic works in our world. Our axioms of formal logic are based on that innate (or maybe learned) sense of how causality and logic work in the physical world, and for the purposes of actually doing formal logic, that's good enough.

However, Hofstadter did not take the next step and ask *why* that should be. That is, why does our method of reasoning, the one that we use everyday for getting along in the world (whatever it is), actually work? Of course, Darwin answered that question: organisms whose reasoning about the world (whatever form that reasoning takes) does not actually reflect the structure of reality, do not live long enough to pass their insanity on to offspring. (However, some theistic philosophers have questioned this idea as well -- the theory of evolution, being based on reason, might actually be wrong, if our reason does *not* reflect reality. In other words, I guess, it might just all be an elaborate joke by god.)

I think that considering the relationship between rationality and evolution would be a fruitful avenue of research in AI. Probably lots of people have had that thought already.

I just read a Weekly Standard article by Bill Kristol http://www.weeklystandard.com/Content/Public/Articles/000/000/012/711lzwcj.asp to which Andrew Sullivan http://www.andrewsullivan.com linked today. When I got to the part about the "vague Article 3 standards", I decided to, y'know, Google the Geneva Conventions (http://www.unhchr.ch/html/menu3/b/91.htm). And now I really have to wonder if Kristol's editors, or any of his readers, or even Kristol himself have ever bothered to do that. I wonder which provisions of Article 3 he thinks are "vague"? Would it be the part where "the following acts are and shall remain prohibited at any time and in any place whatsoever"? Or could it possibly be the descriptions of the acts themselves, which include "Outrages upon personal dignity, in particular, humiliating and degrading treatment"?

Apparently Kristol doesn't expect anyone to notice this discrepancy between Article 3 and his characterization of it. Which makes me wonder: are Weekly Standard readers *always* so easy to manipulate?

But of course, he's just parroting the Bush administration's take on the matter. Any reasonable and reasonably intelligent person would have to agree that Article 3 is perfectly clear about what it prohibits. Therefore, we are left with two possibilities: the Bushies are either (a) stupid, or (b) unreasonable. In the first case, the USA is a laughingstock for electing such a nincompoop. In the second case, we are setting an example of "creative" interpretation of international law that will make a mockery of the Geneva Conventions themselves. If it's not obvious to these guys that such a course is going to come back to bite us in the ass, we're back to case (a): they're stupid.

I recently investigated version-control systems for personal use. I didn't spend a lot of time on any particular one, just enough to absorb the general ideas (or at least, I think so). I knew I wanted to step away from CVS, but I didn't know where to go. Simplicity was my strongest desire, although I do also have some need to manage changes both at home and at a client site.

The systems I looked at:

* subversion : http://subversion.tigris.org
* arch : http://wiki.gnuarch.org
* darcs : http://www.darcs.net

Executive summary: I'm using darcs. Read the flipside for my reasons.

I've abandoned Pomo (see below) in favor of Lisp. See why on the other side: