Sorting Out the Truth: The DICOT Credibility Score

Posted on by
Reply

In this new era where anyone can publish anything any time, the ability to discern what’s probably true and what’s probably bullshit is more important than ever.

In an article titled “Sorting Out the Truth”, Jerry Michalski proposes a quick way to score an item that you come across:

A quick credibility score

We need a way to evaluate quickly the credibility of items as they cross our path, the way newborns are evaluated with the Apgar Score.

In the interest of creating such a score for news, allow me to suggest DICOT (no idea if there’s a floral metaphor here; work with me):

  • Detail: 0 = important details missing; 2 = all questions addressed, given situation
  • Identity: 0 = no disclosure; 2 = reliable disclosure of trustworthy person(s)
  • Context: 0 = spin condition red; 2 = few reasons to suspect malice
  • Openness: 0 = no contact or info shared; 2 = open content, all questions answered
  • Tone: 0 = could be a joke; 2 = straightforward and serious

So a perfect DICOT score is a 10.

Michalski uses last year’s  “Flying Like a Bird” video – which garnered millions of views but turned out to be a well-made hoax – as an example of how his DICOT score could be employed:

DICOT would be a nice way to do a quick assessment of initial news reports and rumors, as well as deeper reports later. Let’s put “birdwings” through it.

  • Detail: 0 – important details missing everywhere
  • Identity: 1 – mysterious disclosure – I’m being charitable
  • Context: 1 – it’s a big claim; people lie a bit in these situations, to get attention
  • Openness: 0 – no info shared
  • Tone: 0 – could be a joke

The noble birdwingers would rack up an impressive 2 out of 10, which would be a nice warning to all.

Read Jerry Michalski’s full article at Forbes.

Quantifying Consciousness: Giulio Tononi’s “Phi”

Posted on by
Reply

In an article in The Atlantic titled “Awakening” Joshua Lang describes how patients undergoing surgery sometimes become painfully aware of what’s being done to them despite being under the influence of a general anaesthetic, and the ongoing effort to devise a way to detect whether a patient is conscious.

To me, the most interesting part of the article is the final section, which discusses Giulio Tononi’s theory of consciousness. The remainder of this post is quoted from Joshua Lang’s article.

At the heart of Tononi’s work is his integrated-information theory, which is based on two distinct principles, as intuitive as they are scientific. First, consciousness is informative. Every waking moment of your life provides a nearly infinite reservoir of possible experiences, each one different from the next. Second, consciousness is integrated: you can’t process this information in parts. When you see a red ball, you can’t experience the color red separately from the shape of the ball. When you hear a word, you can’t experience the sound of it separately from its meaning. According to this theory, a more conscious brain is both more informative (it has a deeper reservoir of experiences and stimuli) and more integrated (its perception of these experiences is more unified) than a less conscious one.

Compare the brain to New York City: just as cars navigate the city’s neighborhoods via a patchwork of streets, bridges, tunnels, and highways, electrical signals traverse the brain via a meshwork of neurons. Tononi’s theory predicts that in a fully conscious brain, traffic in one neighborhood will affect traffic in other neighborhoods, but that as consciousness fades—for instance, during sleep or anesthesia—this ripple effect will decrease or disappear.

In 2008, in one of several experiments demonstrating this effect, Tononi pulsed the brains of 10 fully conscious subjects with his electromagnetic gun—the equivalent of, say, injecting a flood of new cars into SoHo. The traffic (the electromagnetic waves) rippled across Manhattan (the brain): things jammed up in Tribeca and Greenwich Village, even in Chelsea. Tononi’s EEG electrodes captured ripples and reverberations that were different for every subject and for every region of the brain, patterns as complex and varied as the traffic in Manhattan on any given day.

Tononi then put the same subjects under anesthesia. Before he pulsed his gun again, the subjects’ brain traffic seemed as busy as when they were conscious: cars still circulated in SoHo and Tribeca, in Greenwich Village and Chelsea. But the pulse had a drastically different effect: This time, the traffic jam was confined to SoHo. No more ripples. “It’s as if [the brain] has fragmented into pieces,” Tononi told me. He published these findings in 2010, and also used them to file a patent for “a method for assessing anesthetization.”

Tononi is to his neuroscientist peers as the 18th-century philosopher Immanuel Kant was to his empiricist counterpart David Hume. Like most modern neuroscientists, Hume saw only the “easy problem.” He proposed that consciousness was nothing more and nothing less than the bundling of various bits of experiential knowledge, or, as he called them, “perceptions.” Using this logic, my physiological argument against free will could stand.

Kant, however, believed that the mind is more than an accumulation of experiences of the physical world. Like Descartes 150 years earlier and David Chalmers 200 years later, Kant focused on the “hard problem,” making the logical argument that something beyond sensory inputs must account for the subjectivity of conscious experience—what Kant called “transcendental” consciousness. Tononi’s theory hinges on a similar conception of consciousness as something more than the sum of its experiential parts—leaving room, then, for the possibility of free will.

The amount of integrated information in the brain—the quantity of consciousness—is what Tononi calls phi. He believes he can approximate it with a combination of his TMS‑EEG technology and mathematical models. Many well-known philosophers and neuroscientists, however, remain skeptical. Chalmers has praised Tononi for his bold attempt to quantify consciousness, but he doesn’t think Tononi has come any closer to solving the hard problem. And even Tononi admits that, in scientific-research time, his theory is still in its infancy. What Tononi has made progress on is neither the easy nor the hard problem: it’s the practical problem. He is currently developing a machine that has the potential to end anesthesia awareness once and for all. Like the BIS monitor, the device would provide a numerical assessment of a patient’s awareness, and would be simple and compact enough to become a regular fixture in operating rooms. Unlike the BIS monitor, it would also be relevant outside the operating room. Whereas the BIS is rooted in data specific to surgery, Tononi has developed a comprehensive theory of consciousness that could, with appropriate technological tweaks, be applied in any number of medical, scientific, or social settings.

While subduing consciousness is the most urgent aspect of Tononi’s work, he is especially animated when discussing consciousness in its fullest, brightest state. In his office in Madison, he described a hypothetical device called a “qualiascope” that could visualize consciousness the same way telescopes visualize light waves, or thermal goggles visualize heat. The more integrated the information—that is, the more conscious the brain—the brighter the qualiascope would glow. Using the device in an operating room, you would watch a patient’s consciousness fade to a dull pulse. If he woke up mid-operation, you might see a flicker.

But if you turned your gaze away from the operating room, you would gain an astonishing perspective on the universe. “The galaxy would look like dust,” Tononi told me. “Within this empty, dusty universe, there would be true stars. And guess what? These stars would be every living consciousness. It’s really true. It’s not just a poetic image. The big things, like the sun, would be nothing compared to what we have.”

VISUALIZING CONSCIOUSNESS In an experiment on vegetative patients, researchers pulsed one subject’s brain with electro­magnetic waves on three different days as the subject emerged from a coma. The resulting EEG patterns reflected Giulio Tononi’s theory of consciousness: they became more complex and widespread as the patient became more conscious.

VISUALIZING CONSCIOUSNESS
In an experiment on vegetative patients, researchers pulsed one subject’s brain with electro­magnetic waves on three different days as the subject emerged from a coma. The resulting EEG patterns reflected Giulio Tononi’s theory of consciousness: they became more complex and widespread as the patient became more conscious.

Read the full article at The Atlantic

On Gurus

Posted on by
Reply

In the documentary film Kumaré, a fairly ordinary guy named Vikram Gandhi, born and raised in New Jersey but of Indian ancestry, invents a false identity as a “long-haired, orange-robed, heavily accented Hindu guru” called Sri Kumaré. In this guise he goes about attracting a following, but eventually reveals himself in a “Great Unveiling.”

In an insightful article published in Aeon Magazine, Erik Davis thoughtfully examines Vikram Gandhi’s deception and its effect on his followers. Davis writes (in this highly abridged version of his article):

Rather than setting up an atheist’s honey-pot, Gandhi actually staged something more interesting, and more ambiguous: a theatre of awakening that transforms himself as well as his students. His sceptical and rather self-serving prank turns out, from a certain angle, to be weirdly spiritual, stirring up, at least for people familiar with modern gurus such as Gurdjieff or Bhagwan Shree Rajneesh, the prickly conundrums of trickster spirituality. The irony is that it’s not clear that Gandhi himself really grokked the implications of his ruse, or the depths contained within his alter-ego’s self-reflexive teaching that ‘you are your own guru’.

Kumaré’s core teaching is pretty consistent. He tells his students that he is an illusion, nothing more than a mirror or a symbol, and that the guru they seek is inside of them. At one point, he gets a student, whose eyes are closed, to sit down before an altar. When the student opens his eyes, he sees framed photos of Barack Obama, Osama bin Laden and Kumaré. ‘All symbol, just symbol,’ lisps the guru.

What is marvelous about this ‘teaching’ is its perfect duplicity, which paradoxically allows both Gandhi and Kumaré to speak true. For seekers, Kumaré is offering a version of the familiar Vedantic message that the Atman (or soul) within is the Brahman (or guru) without; meanwhile, Gandhi foregrounds the prank for his viewers. Whereas many yogis and other spiritual teachers are arguably more ‘authentic’ in their presence and practice than in their often clichéd rhetoric, Kumaré is most truthful when proclaiming his own status as an illusion. Earnest spiritual teachers, even flawed ones, also want to wake up their students, whereas the awakening of Kumaré’s students to the truth of his illusion becomes, for Gandhi, something of a problem. Unlike Sasha Baron Cohen, whose pranks often depend on an almost contemptuous detachment from the tentative, affective bonds of trust that grow between strangers, Gandhi grows to feel a genuine connection to his students within the rigged theatre he has created. He likes them and wants them to discover this truth within themselves before he rams it down their throats with his Great Unveiling — a final act of truth-telling that will, Gandhi knows and fears, probably make them feel ‘like idiots’.

Gandhi makes much of the fact that, as Kumaré, he ‘made up’ his chants, yoga moves, and meditation rituals, including the core practice of the ‘blue light’, a fairly heavy-duty group visualisation exercise… But there is a deeper conundrum here, one that Gandhi seems to miss. When he ‘made up’ his moves, Gandhi did not, of course, invent them out of thin air. Instead, he drew on what he was familiar with from studying religion, taking yoga classes, following gurus, watching movies, and just paying attention. He imitated, adapted, and modified, just like anyone else working in a genre — and more or less like all innovative spiritual teachers do, especially when they are working within popular traditions or the grab-bag of esoteric or New Age spirituality. He was making up stuff, but the stuff he was drawing on to make up his stuff had the force of tradition.

This is what sceptics naively misunderstand about spiritual ‘authenticity’. Creative fabrication, intentional or not, is part of the spiritual tradition, as is the necessity of some sort of studied engagement (after all, Gandhi did not just wake up one morning sitting in full lotus position). If yoga, chanting, and meditation practices have any power outside their strictly traditional contexts, then that power emerges from the practices themselves as they mutate over time. The point is that while sincere and skilled teachers are great, even an unethical blowhard can teach you to circulate energy through your body.

The 20th-century Hindu master Sri Nisargadatta (1897-1981) attacked ‘the self-appointed guru’ who, in contrast to the true guru, is more concerned with himself than with his disciples. Here we encounter another one of Kumaré’s unintended ironies. The fact that Gandhi is performing a character means that, unlike all but the most perfect masters (if they exist), his persona is not burdened with actual self-striving. Because his real ego was elsewhere, Gandhi could afford to become selfless, especially because — given the guilt he inevitably feels for conning his students — he wants to make their experience as valuable as possible. In other words, his lie inadvertently became what Buddhist teachers describe as upaya, or skillful means.

After the Great Unveiling, while a few of Kumaré’s students fled the room in anger, the majority stuck around and engaged the Indian-American dude from New Jersey on his own terms. After their pursuit of an authentic spiritual connection was essentially betrayed, most of Kumaré’s students trusted their hearts and embraced the latest mindfuck rather than retreat into the shell of the wronged self. In other words, they were able to integrate Gandhi’s media con as part of the path. Gandhi might have set out to expose the deluded projects of foolish New Age seekers, but their reaction to the Great Unveiling speaks at least as much to the strength, intelligence, and experimental grit of the majority of his disciples. Deluded literalists, locked into fetishising the exotic and magical other, could not have pulled this off.

Here, then, is the greatest irony of Kumaré: what appears on the surface to be a debunking of gurus winds up underscoring the ongoing resilience of seeker spirituality.

One of the reasons the trickster plays an important role in this evolving spiritual culture is that an important current in that culture uses scepticism, disenchantment, and even pranks as opportunities for liberation — the swords that slaughter the Buddhas you meet on the road.

 Read the full text of Erik Davis’ article

 

For a few years when I was in my mid-twenties, I spent a considerable amount of time checking out various forms of Eastern spirituality. Among other experiences, I attended a talk by Jiddu Krishnamurti, hung out with the sannyasins at one of Bhagwan Shree Rajneesh’s ashrams, often visited a Hare Krishna center in San Diego for their vegetarian meals and chanting, and for a while regularly attended meditations at a Siddha Yoga center (Swami Muktananda’s organization.)

While I felt a certain degree of affinity for each of these gurus and their groups, ultimately I could not stick with any of them. What kept me from fully integrating into them was the nearly allergic reaction I felt when encountering the unquestioning devotion of any guru’s followers.

And in fact, many spiritual leaders who have been subjects of such devotion have later been found to have engaged in various shenanigans, including financial misdeeds, sexual exploitation of their followers, and other abuses of their power. In other words, they have turned out to be human after all.

Humans are only human. None of them are perfect, and we do both them and ourselves a disservice when we fail to recognize this. Even Tenzin Gyatso, the 14th Dalai Lama, who is perhaps the living person who I most admire, has his failings, as he himself (to his great credit) admits.

What I have come to understand is that we need to separate the teachings from the person. Looking back on my life so far, I recognize that many, many people have been valuable and influential teachers for me. Some of them, like Tenzin Gyatso, occupy fairly exalted positions; others are ordinary people of no particular renown. All of them are humans;  none of them are infallible. Some of them are not even particularly admirable. In every case, it’s up to me to determine what learning I can take away from whatever they might impart.

I find it helpful to consider that every person holds a piece of the truth. And reminding myself of my own combination of talents, knowledge, foibles, and limitations helps me to have compassion and to look for the good in those I encounter in my journey through life.

School Placement Lotteries: Problems and a Solution

Posted on by
Reply

In 1994 when my son was about to reach kindergarten age, I realized there were  glaring problems with the way the Minneapolis Public Schools operated their school choice placement program. With a bit of research I discovered a solution, patterned after the National Resident Matching Program that matches medical school graduates with residency programs in the U.S. I proposed this method to the school district. Although the guy who ran the school lottery program recognized that this would be a significant improvement, the idea never went any further due to bureaucratic inertia.

A couple of decades later, the problems I identified still exist. So does the potential solution. In fact, the designers of the NRMP algorithm, Lloyd Shapley and Alvin Roth, won the 2012 Nobel Prize in Economics for their work.

 

Mr. Denny Lander
Minneapolis Public Schools
807 NE Broadway
Minneapolis, MN 55413

Feb. 21, 1994

Dear Mr. Lander,

I am a computer consultant and (more importantly) a Minneapolis parent.  My son will be entering kindergarten in the fall of 1995.  My wife and I feel that choosing the right school for our son is important, and since we don’t want to be rushed in the decision we have already begun investigating the options available to us.

As I’ve talked with other parents who are going through this process, I’ve become aware of a dilemma that confronts many parents.  Certain schools are very popular, and the chances of  getting into them are slim.  Parents who list such a school as their first choice are not likely to get it.  But worse, because of the way the placement lottery works, they also take the risk that their second choice might fill up during the first round. Parents who list three popular schools as their 1st, 2nd, and 3rd choices are unlikely to get any of them.  They might therefore choose to list as their 1st choice a school that actually falls considerably further down their list of preferences.

This has undesirable consequences for both parents and the school district:

  • Parents who understand the system are put in the uncomfortable position of weighing the odds.  They must decide whether to gamble or play it safe.
  • These parents have to decide without sufficient information.  Since there is no way of knowing how many people will apply to each school, it’s difficult to evaluate the risk involved in requesting a popular school.
  • Less sophisticated parents who naively list a popular school as their first choice are, in effect, penalized for their lack of understanding of probability.
  • The school district receives invalid information about what parents want because many people are likely to list school choices that do not reflect their true preferences.

I believe there is a solution.  The National Residency Matching Program (NRMP) manages the placement of  medical school graduates in residency programs at hospitals around the US.  Applicants rank hospitals in their order of preference.  The matching algorithm works in such a way that applicants do not need to know or worry about their chances of getting into particular hospitals;  they are free to honestly rank their preferences without regard for the odds.  In fact, doing so gives an applicant the best possible chance for getting a hospital he or she likes.

This algorithm is perfectly suited to public school placement.  The only significant difference between the two situations is that hospitals submit their own lists of preferred applicants in rank order.  The public schools do not do this;  however, they do something similar by giving preference to siblings, children with special needs, etc.  A public school’s preference list can be generated automatically using these preference rules and random assignment of rank order within each preference group.

The beauty of this is that it requires no change at all from the viewpoint of parents.  They will still rank order their preferences just as they always have. The process actually becomes simpler for parents, because they no longer have to worry about the odds of getting into particular schools.  Of course, it would be good to publicize the change in algorithms so that parents know that they are now free to list their honest preferences.  And it might be desirable to allow parents to list more choices (say, up to five) in case their top three choices are all popular.

I’m enclosing some information that details how the matching algorithm works.  I hope you will take a few minutes to look it over and give me a call.

Sincerely,

David R. Woolley

 

The NRMP recently posted a new explanation of how the matching program works. But the following fable the NRMP published in the 1990′s also illustrates the process in a very easy-to-understand way.

 

The Matching Tale

How it started…

Once upon a time a shoemaker was in need of an apprentice. To find one, he went to the fair, set up his booth, and began extolling the virtues of his profession. Soon the young men gathered around him to sign up. The next year the plumber and the carpenter followed his example, as did the butcher, and various others. Over time, the fair became quite chaotic, and it was not easy to determine which of the conflicting claims sounded the best. Some groups took their booths outside the fair so they could catch fairgoers before the others did.

This went on for many years. Some of the young men learned to manipulate the system. To get the best place at the plumber’s booth they sported a button “I Love Plumbing.” As they walked to the next booth, they quickly turned their lapel showing another button with “I Love Carpentry.” Over time, someone thought, “There has to be a better way!” That is why the current system evolved.

First, it was decided that no contracts could be signed on the spot. During the fair everyone walked around with a notebook comparing the best of the preceptors and each of the applicants had to offer. After the fair, all went home to study their notes. A week later, they gathered again at the town square. Benches had been set up on either side, one side for the preceptors, one side for the future apprentices. In the center stood the town clerk, who had given each preceptor some tokens, one for each apprentice position he wanted to offer.

At the meeting…

As everyone was seated, the town clerk asked for silence and called the first preceptor. The first preceptor had three tokens. He looked at his notebook, and offered one token each to the three applicants at the top of his list. The applicants accepted the token and bowed politely. Then the town clerk called the second preceptor, then the third, fourth, and so on.

When the fifth preceptor stepped forward he offered a token to one of the youths who had received a token already. The youth stood up and said, “Thank you, Sir. I’m honored, but I already have a token that I like better.” The preceptor looked at his notebook and offered the token to the next one on his list.

Several preceptors later, another youth also received a second token. This youth replied, “Thank you, Sir.  I already had a token, but I prefer yours since you are closer to where my sweetheart lives.” The youth accepted the new token and handed his first token back to the town clerk, who returned it to the preceptor who had given it out earlier. That interrupted the sequence for a moment, since the earlier preceptor now had to look in his notebook to determine the next person to whom he would offer his token.

As the day progressed, this happened more often. Each time an applicant received a token that was higher on his list, eh accepted it and turned in the earlier token;  that token then passed to someone else. In time, many applicants received better offers and were able to get closer to the top of their list. Several preceptors had to go farther down their lists than they had planned, but all were pleased with the way the town clerk conducted the meeting. There was no more screaming and yelling to get to the front row.

What happened…

This is how the first matching program was born. Let us take a look at some of the things that happened and at those that did not.

1. First, the town clerk did not have the power to make any decisions. His function was to make sure everyone followed the rules.

2. Preceptors did not ask “Who loves me the best?” Nor did they give their token to the one who yelled the loudest. Applicants did not need to approach each preceptor telling him that they liked him better than the next one. To whom each preceptor would give the next token was determined only by the list in his notebook.

3. Also, preceptors were not penalized for offering tokens to applicants who might well return them. Each token retained its full value. As their list came down to an applicant who wanted their token, that one would keep it, no matter how many had rejected it before.

4. Similarly, applicants did not need to worry that they used space in their notebook to list preceptors from whom they might not receive an offer. Those extra names had no effect. The only thing that mattered was the position of preceptors from whom they did get offers.

5. Neither did the applicants worry about the sequence in which the preceptors made their offers. Even if their favorite preceptor was the last one to step forward, they could still accept his token when it was finally offered and return the token they had held to that point.

How it evolved…

The town square meetings worked well, but after a while preceptors noted that it was not really necessary to go to the town square themselves, since all they did was read the lists they had prepared in advance. They could give this list to a friend or to a broker and ask him to go in their place. They would still be assured that the outcome would be exactly the same. Very soon the applicants reached the same conclusion. They, too, gave their lists to a broker to represent them.

This is exactly how the first manual matching programs were run: two persons, one with a stack of training program lists, and one with a stack of applicant lists, exchanging notes until all possible offers had been made.

Not long thereafter it was decided that not even the town clerk and the two brokers were needed. Since their actions were entirely predictable, they could be replaced by a computer program. Fortunately, the town clerk and the brokers are not out of work. They were rehired to become the computer programmer and data input operators.

This is how the Matching Program operates today. The computer program runs faster, but the steps it goes through are exactly the same as in the meeting on the town square. If you understood the town square meeting, you will understand the matching program logic.

Web Discussions: Flat by Design

Posted on by
Reply

In his Coding Horror blog, Jeff Atwood makes the case that online discussions should be flat rather than threaded. “Flat” means a discussion consists of a single linear chain of messages in chronological order. “Threaded” means replies can be inserted anywhere, so sub-discussions can branch off the main discussions, sub-sub-discussions can branch off from those, etc.

I have maintained the same position for years. Almost 40 years, in fact — I designed PLATO Notes discussions to be flat, and I’ve always argued that a flat structure is best for promoting group discussion because it most closely replicates the experience of holding a face-to-face discussion with a bunch of people. I allow that threading has some valid uses, particularly for tech support, question-and-answer sorts of applications where you’re not there to engage in prolonged discussions but just want to drill down to a specific answer as quickly as possible and leave.

Anyway, I appreciate Atwood’s post because he goes into the flat vs. threaded argument in more depth than I ever have. I agree with all of his key points:

  • Browsing a tree is complicated because you have to constantly think about what level you’re at, what’s expanded, what’s collapsed.
  • Replies can arrive any place in the tree any time. How do you know if there are new replies? Where do you find them?
  • Indented replies to replies to replies push the discussion off your screen.
  • Replying to a reply gives you the impression you’re only talking to that person, but in reality you’re talking to everyone.
  • Scrolling down is the most natural way to get through a discussion. Having to collapse dozens of random tangents is too much work.

Here’s what I’ve had to say on this topic in the past.

Read Jeff Atwood’s full blog post.