Sunday  30 Nov 2014           comment?

Saturday  29 Nov 2014           comment?

Uncyclopedia, a site of parody articles, came to my attention this past week. I'm a tough customer when it comes to comedy, so if I found the site only occasionally funny that may say more about me than it does about Uncyclopedia. I can find a line like this (from the article on chess) somewhat amusing, but an article's worth of lines in a similar vein starts to wear thin:

Machines have made great strides in freeing humans from the burden of playing chess.

I was tempted to add to an Uncyclopedia page but I desisted. The joke I had in mind was derisive—not excessively so, just derisive in the way that much of Uncyclopedia is, which feels as sporting as shooting fish in a barrel.

Ten days ago, I said I wasn't offended by jokes made at the expense of a sheep. I didn't go into why, partly because I leave things for readers to consider rather than spelling everything out. And I thought it was kinda obvious that the sheep in question wasn't going to read what was written about her on the web.

It remains a bit of a mystery to me that so much humor is cruel: a mystery in the sense that I don't fully get what it tells us about ourselves.

Friday  21 Nov 2014           comment?

Our current definition of the second (in terms of the frequency of radiation associated with a transition in cesium atoms) dates from 1967. Considerably more accurate time standards have since been developed, making the definition somewhat anachronistic.

We can now measure time with an accuracy of about two parts in 10¹⁸: more accurately than we can measure length, mass, force, current, voltage, or temperature.

Einstein showed that time passes more slowly with increasing gravitational field strength. The effect of Earth's gravity on time is small, but our best time standards are accurate enough that a clock near the floor of a room runs noticeably slower than one near the ceiling.

The relativistic effect of gravity affects every kind of time standard. It's a different (and much smaller) effect than a pendulum clock's sensitivity to the strength of gravity. I have a mechanical wall clock that lost about two minutes a week after I moved from sea level to Lone Pine, in proportion to the increase in distance from the center of the earth.

Wednesday  19 Nov 2014           comment?

A climbing buddy's alma mater is in the news for what students are referring to as the sheep incident. From conversation while out climbing yesterday:

cb:	I would've guessed it was an ag student. But this was a computer engineering major.
Tommy:	This feeds into the stereotype that geeks don't have the social skills ...
cb:	... to get someone of their own species.

Modern-day scruples against considering sexual assault a laughing matter are evidently somewhat relaxed when a sheep is involved. I am amused rather than appalled by this cartoon in the campus newspaper, and by the hashtag #BaaaMeansBaaa started in reference to this affair.

Happy nineteenth, everyone.

Thursday  13 Nov 2014           comment?

There's a world chess championship in progress, not that you'd know it from reading The New York Times. Game results get no mention under sports, nor in the chess column that the Times unceremoniously discontinued last month. The October 12 installment was followed by a terse statement: "This is the final chess column to run in The New York Times." Sounds like they mean for ever and ever.

Some chess devotees shrugged at the column's demise because they weren't impressed with it anyway. You can't judge a page by its URL but perhaps the Times' attitude toward chess was reflected in the column's path (see the link above, that files chess under "crosswords"). Sad part is, there's no chance for the column to improve now that it doesn't exist. It's worth remembering that the Times had a serious chess column decades ago when Samuel Reshevsky was writing it.

US news media often don't do a stellar job when it comes to chess. A 60 Minutes piece on Magnus Carlsen described the number of possible chess moves as "infinite" (it's not) and in a supplement the reporter said, "chess is all about deception" (huh?).

The championship in progress is a rematch between Magnus Carlsen and Viswanathan Anand. I've been watching the match live—well, just the tail ends of games as they are playing in Russia and start at 0400 PST. The match's official site has live video and commentary; other live presentations (not interrupted by Russian commercials) are available on the web as well, e.g. moves and commentary at twitch.tv .

Wednesday  12 Nov 2014           comment?

"White mates in two" is my favorite type of chess problem. The example here is one I saved from a magazine in 1972. I was thinking last year about what it would take to write software to create ("compose", in chess parlance) problems, but I didn't get very far.

Just this week I read about a program that composes chess problems better than at random, i.e. taking aesthetics into account. The program's author, Azlan Iqbal, did his Ph.D thesis on computational evaluation of aesthetics in chess. Dr. Iqbal said

Even though there may be philosophical objections as to whether or not aesthetics can be defined in the first place, much less estimated using mathematical formulas, my research suggests that a computer can indeed be programmed to identify combinations and compositions, at least within the specified scope of mate‑in‑3, that the majority of human players with a reasonable knowledge of the game would likely consider beautiful.

So far so good, but now for for the strange part. Dr. Iqbal's software somehow pairs attributes of chess problems to attributes of just about anything people find aesthetically appealing:

The beauty of the approach is that neither the two domains nor their attributes need have anything in common semantically. This should be no more surprising than when an artist says he painted the magnificent landscape after having had two drinks and thinking about a former girlfriend. Why the DSNS approach works at all is still an open question. We have only been able to demonstrate experimentally that it does work.

I had a dream this afternoon in which my brother gave me a cardboard model of a polyhedron and asked me which song it corresponded to. It didn't correspond to any song I know of, not even by means of some inside joke my brother and I had. But dreams don't have to make sense, and evidently neither do heuristics for composing appealing chess problems.

I don't know which is stranger: composing problems by likening them to pictures of people, or computer Go programs that evaluate games played largely at random. But I suspect that if and when we ever start to understand the mechanisms by which our brains gauge aesthetics, they will stike us as bizarre. Some aspects of the nervous system that we already understand have an I‑can't-believe-this-works quality to them.

Saturday  08 Nov 2014           comment?

I had occasion to use Firefox today and was greeted with the text shown here. "Kind of like you!" reminds me of an encounter with a politician I blogged about nine years ago.

The blue text linked to a page at thunderclap.it about how wonderfully independent and non-corporate Firefox is and won't you help spread the word. While the page was loading I saw it was hitting Google analytics, which made me curious enough to read Thunderclap's privacy page (which is effectively part of their Terms of Use). One section says

Use of Personal Information
You agree that we may use your personal information to:

• provide the services and customer support you request;
• ...
• for any other purpose for which the information was collected

A charitable reading of the last bullet point would be to interpret it as identical to the first, in which case why say it at all. I have no choice but to take it at face value, i.e. you agree that we may use your personal information for anything we want to do with it.

I find that insulting. In linking to a Thunderclap campaign, the Mozillians evidently don't consider its terms as repulsive as I do (or perhaps they didn't bother to read them).