Thursday, August 13, 2009

The Amazing Meeting 7: Sunday paper sessions, Million Dollar Challenge

This is the sixth and final part of my summary of TAM7, covering the last day's events on Sunday, July 11. Part 1 is here, part 2 is here, part 3 is here, part 4 is here, part 5 is here, and my coverage of the Science-based Medicine conference begins here.

Sunday's continental breakfast was served while an old James Randi television appearance on the Oprah Winfrey show from 1986 was shown. This brought back some old memories--I think I have the show on videotape in my archives, as I think we showed it at a meeting of the Phoenix Skeptics. Randi appeared with a faith healer ("Amazing Grace"), a psychic (Joyce Keller), and an astrologer (Irene Hughes), which led to some entertaining and ridiculous exchanges of words. Randi showed his footage that exposed Peter Popoff using a wireless transmitter and receiver to fake the "word of knowledge," and did some spoon bending. Joyce Keller claimed she was entitled to his $10,000 prize, and Oprah mistakenly claimed that Randi had brought his own spoons, which she corrected herself about after a commercial break.

This was followed by the Sunday refereed papers, which were again organized and moderated by Ray Hall, professor of physics at California State University, Fresno and at Fermi National Labs.

Don Riefler, "Teaching Critical Thinking in a Therapeutic Setting"
Don Riefler, Direct Care Supervisor at the Jessie Levering Cary Home for Children in Lafayette, Indiana, gave a talk about strategies he's used to teach critical thinking to underprivileged/institutionalized children at the Cary Home, complete with positive reinforcement in the form of candy distributed to members of the audience who gave good answers. He discussed several categories of common "thinking errors" which included both logical fallacies and heuristics that lead to problems when overgeneralized. As part of his teaching, he has kids conduct ESP experiments with Zener cards, which he uses to teach them about erroneous inferences they draw about their skills. This provoked the first critical question (from regular ScienceBlogs commenter Sastra), asking whether his referral to "success" and "failure" in the Zener test suggests to kids that it's a matter of effort. (I neglected to record his response.) In answer to a question of how he deals with religion he said that he avoids it and shuts down talk of religion or ideology.

David Green, "Patently Ridiculous: The Perfect Sommelier"
David Green, a Senior Patent Examiner at the Canadian Intellectual Property Office, gave a talk that was essentially a sequel to a talk he gave at TAM5. He spoke about "The Perfect Sommelier," a product that claims to "align tannin molecules with magnets to age wine faster." He compared how the patent application for this product was handled in the U.S. vs. Canada.

In the U.S., patent examiners made two objections to the application, first, that it was obvious or already known, and second that the "subject matter is inoperable--the theory of operation cannot be correct." The first objection failed, since the invention was sufficiently different from prior art in various ways (such as having magnets at both ends of the bottle, not just at one end). And, based on the Longer ("lawn-jay") test, under which the description of the invention must be accepted as true unless there's a reason to doubt it, it passed on the second as well, and was granted two U.S. patents. Green said that it essentially comes down to a he-said/she-said debate, and the patent office has to be biased towards issuance of the patent.

In Canada, the same objections were made as in the U.S., along with a third. David Green had read a Swift article about a test of the product, so the third objection was a rejection on the basis of double-blind research evidence showing that the product doesn't work, published in the Journal of Wine Research. That study concluded that "no evidence was found to suggest that The Perfect Sommelier improves the palatability of cheap red wine." The manufacturers responded to the first two objections in the same way they did in the U.S., but for the third, they asserted that their evidence in the form of testimony overrides the double-blind research.

And then they abandoned their patent claim in Canada.

The reason they did this, Green explained, is because of "U.S. file wrapper estoppels"--that what you do in a foreign patent application can affect your patent in U.S. court. If they had continued with their claim in Canada and been denied--or if they had failed to file a response to the objections--that could have impacted their U.S. patent.

What this demonstrates, Green argued, is the importance of doing solid investigations and research on such products, and getting them published and spreading the information around (e.g., online), so that patent examiners can find it. It can make the difference between a nonsensical product getting a patent or being denied a patent.

At this point I took some time to chat with Ray Hyman, and came in a little bit late for the next presentation.

Adam Slagell, "Fear, Uncertainty and Doubt: The Pillars of Justification for Cyber Security"
Adam Slagell, Senior Security Engineer at the University of Illinois at Urbana-Champaign, spoke about claims made for security and security products that we should be skeptical of. He pointed out that there's no such thing as perfect security, and there are always tradeoffs to be made between security and usability/convenience/etc. He spoke a little bit about TSA "security theater," pointing out the gaping flaw in the "no fly lists" that comes from the separation of checking ID and boarding pass at the security checkpoint from checking your boarding pass at the gate. He also questioned the point of shoe removal, which led to the first comment on his talk from Ian, an airport security officer at Gatwick, who argued that forcing shoes to go through the X-ray machine does close a genuine vulnerability. (Ian also argued that the liquid restriction makes sense, though he didn't respond to Slagell's point that you can carry multiple 3-ounce containers and combine their contents with those of your associates after you go through screening. Most interestingly, Ian said that airport metal detectors go off randomly in addition to when they detect metal.)

Slagell argued that signature-based antivirus products are obsolete, since polymorphic malware and use of packers are extremely effective at eliminating signatures, and observed that companies are starting to create products based on white-listing, only allowing pre-defined sets of software to run on a machine. (At last year's New Mexico InfraGard conference, Anthony Clark and Danny Quist spoke in some detail about different kinds of packers, and offered a set of criteria for measuring AV effectiveness that included use of methods other than signature-detection, such as anomalous behavior detection.) He unfortunately didn't have time to talk about passwords.

Another questioner asked what users behaviors are useful to stay secure, to which Slagell replied that you should keep systems patched and backed up. (There is actually some argument, at least for corporations, to be somewhat selective in patching, since many patches aren't applicable, have other mitigations, and have potential for reducing availability themselves--but there is no substitute for having a vulnerability management program in place.)

Steve Cuno, "The Constructive Skeptic: Rebranding Skepticism at the Grassroots Level"
Steve Cuno, chairman of RESPONSE Agency, Inc., gave an excellent talk last year at TAM6, and he gave another great presentation this time as well. He started by saying that skeptics have a branding problem.

What is a brand? Is it a name and logo? A great slogan? What you say about yourself?

He gave some counterexamples for each of these, including some nice vintage ads (e.g., "They're happy because they eat lard" from the Lard Information Council). AIG had the slogan "The strength to be here." (He didn't mention any of my favorite unintentionally ironic bank slogans.)

He gave an example slogan for skepticism: "Skepticism: Doubt worth believing in." He called all of these proposed brand definitions "brand flatulence: you may like the sound and smell of your farts, but nobody else does."

He gave as his prototypical example of what branding really is the example of Nordstrom's. There's no particular logo or slogan involved, but people think of Nordstrom on the basis of the values that are expressed by the company through its employees and the experience you have as a customer. The essence of creating a brand is creating a positive customer experience.

And the way for skeptics to give skepticism a good name is by self-policing "to deliver positive brand experience."

He suggested that the way to do this is to delay giving yourself a label, and when you do identify yourself with a label, anchor it in something positive. Instead of saying "I don't believe in ...", think through and express what you do support. For example:
  • I believe in what the evidence supports.
  • I believe in honesty, integrity, equal rights, and treating one another with dignity and respect.
  • I believe in and defend the right of all people to believe as they choose.
Do things that are positive. He gave the example of the GLBT protests at the annual April Mormon Church Conference, which, rather than picketing and protesting, engaged in protest by cleaning up parks, visiting shut-ins, and doing positive and helpful things in the name of their cause. The result was to get tons of positive press.

He heartily endorsed TAM7's vaccination support and food drive, and further added that we should play nice. Being controversial and using insults may work for media figures, but not for the grassroots. Be sure that messages are well-timed. And remember that some people just don't care--to quote Will Rogers, "Never miss a good opportunity to shut up."

A summary of Cuno's talk may be found on his blog.

Brian Dunning, "What Were the 'Lost Cosmonaut' Radio Transmissions?"
Brian Dunning's talk was a sequel to one of his Skeptoid podcasts on Achille and Giovanni Judica-Cordiglia, a pair of Italian brothers who built equipment to monitor radio transmissions from spacecraft at an installation they called Torre Bert. They successfully recorded the October 1957 launch of Sputnik I, Sputnik II with Laika the dog in November 1957, and then a few oddities. In February 1961, they recorded what they reported as a "failing human heartbeat," when there was no known flight. In the same month, they recorded a "voice of a dying man," again with no known flight. In May 1961, they recorded the voice of a woman, Ludmila, speaking about how she was "going to re-enter," which they attributed to a secret female cosmonaut mission that resulted in her death.

There are no corroborating reports of these transmissions, despite the fact that the U.S. Defense Early Warning system began in 1959. And there were no female cosmonauts in 1961. The female cosmonaut program wasn't approved until five months after the recording, and the first five women selected for the program a year later. Yuri Gagarin had just launched in Vostok 1 in May 1961, and for the Vostok 2 launch in August 1961, they had to scavenge Gagarin's space suit to make a suit for the second cosmonaut. So there was no way there was a female cosmonaut launch in May 1961.

At the time, the U.S. was flying X-15s. Did the Soviets have some kind of space plane? The Soviet Kosmoplan never got off the drawing board, and its Raketoplan was developed, but wasn't ready for testing until 1962.

A jet fighter? The YC-150 didn't fly high enough. Dunning also ruled out the Mig-21 and high-altitude balloons.

The conclusion--get your own Russian translators. Dunning got four Russians to listen to the recording, and found that it didn't say what was claimed, but instead was almost 99% unintelligible, with the rest being numbers. He also found that the source of the transmission was not moving, but was at a fixed position.

Although he didn't come to a definitive conclusion, he was able to at least eliminate a number of possibilities--sometimes that's the best you can do.

Christian Walters and Tim Farley, "How Are We Doing? Attracting and Keeping Visitors to Skeptical Websites"
Tim Farley was another return speaker, this time with Christian Walters. They talked about how the over 650 skeptical websites should measure acquisition of visitors and take actions to keep them and to obtain high search engine rankings.

First, how you're acquiring visitors can be measured by looking at rankings on search engine result pages (SERPs), Google PageRank, and Yahoo link strength measurements. These measures are all increased by receiving links from other web sources, of which important sites are social media sites like digg, reddit,, Facebook, and Twitter.

Another important factor is having good page titles, which include popular search terms. The META keyword tags are no longer so important. By using the Google AdWords Keyword Tool, you can find what popular search terms are. Sometimes they are surprising--for instance misspellings of some terms (like accupuncture) get more search hits than the correct spelling.

It's also a good idea to put the keywords from your title into the URL, rather than use URLs as some blogs do that only have a page ID in them.

The anchor text of hyperlinks to your pages should also contain the appropriate keywords, and so your internal links within a site should make a point of using them.

It's important to describe your site with an XML SiteMap or via RSS feed, which you get for free with blogs. When you link to other sites, you are dividing up your own link strength among the sites you link to, unless you use the NOFOLLOW tag, which you should do when linking to sites you don't want to promote in search engine results. NOFOLLOW is also a good idea when linking to sites that may engage in spam or other abuse, to prevent that abuse from reflecting on your site, as it might in Google search engine results, for example.

The Million Dollar Challenge: Dowser Connie Sonne
Everyone had to leave the auditorium for preparation for the JREF Million Dollar Challenge, with Danish dowser and former police detective Connie Sonne (who has described her alleged powers in an interview with Alison Smith of JREF). Everyone had to sign an agreement to remain silent and not disturb the proceedings before filing back in--and everyone remained quite quiet for the hour or so that it took for the test.

This was a preliminary test, with a 1 in 1000 probability of success by chance, which, if successful, would allow Connie Sonne to go on to the official challenge for the JREF's $1,000,000. The protocol for the test was developed in conjunction with Connie Sonne and both sides approved. She signed paperwork describing the protocol and agreeing that she woudl go ahead with the test.

Connie Sonne claimed to be able to use a pendulum to identify playing cards without looking at them, and she successfully did this when she was able to see the cards. Sets of playing cards, A-10, for each of three suits were placed separately into envelopes. Each of those envelopes for the same suit was placed into a larger envelope, with the suit written on the outside. Banachek ran the test (I thought to myself at the time that this was a likely source of future complaint, given his skill at illusion), opening each of the three suit envelopes, one at a time, and rolling a 10-sided die to indicate which card from the suit Connie Sonne was required to locate. The ten individual card envelopes were spread out in front of her, and she used the pendulum to identify which envelope she believed contained the appropriate card. For the first set, she was supposed to find the 3 of hearts, for the second, the 7 of clubs, and for the third, the ace of spades. The cards she picked were the 2 of hearts, which was in the second envelope of the first set, the ace of clubs, which was in the seventh envelope of the second set, and the 2 of spades, which was in the first envelope of the third set. Banachek opened all of the envelopes from each of the three sets so that she could see that there was no trickery, and she agreed that all was done fairly.

At the subsequent press conference, she continued to maintain that all was fair, but that there was some reason she wasn't supposed to reveal her powers to the world yet.

But by the next day, she decided that she had been cheated somehow by Banachek. Her main point of evidence was that Banachek identified the ace of spades from the third set before pulling the card out of the envelope--but it was the last card of the set to be opened, and he identified it after the end of the envelope had been cut off and as he started to pull it out. The cards were visible inside the envelopes once the ends were opened.

On July 13, she made her accusation of cheating on the JREF Forums:
Hi out I know why Banacheck was "the card handler". I have been cheated. I did find the right cards. And there is one more thing. At the stage, Banacheck said to me BEFORE he even looked in the envelope I had cut...and here is spade ace, the one you looked for!!!! I first hit me now about that ....but maybe you can see it yourself if someone get the video. I don`t care about the money, that wasn`t the reason why I came. So no matter what you think out there......I was CHEATED!!!!!

It was a typical response to the Randi challenge from an honest proponent of a claim who doesn't understand why the claim failed under test conditions, resolving the cognitive dissonance by placing blame on the experimenter.

That concludes my summary of TAM7--I look forward to attending TAM8 next year.

1 comment:

BathTub said...

I didn't know she had decided to use the cheater excuse. That's really sad I thought she held herself up really well during the press conference, loony of course, but accepting.