Viswanathan K's Thoughts

The famous “runaway trolley problem” goes like this: an out of control trolley on tracks is hurtling down towards 5 people and will certainly kill them; would you be OK killing some 6 th person to save the five? Robert Sapolsky in his best-selling book, Behave , analyzes it at length with more and more variations. First, it matters how the scenario is phrased: 1)       If the option is to pull a lever that diverts the trolley onto a different track where the 6 th person is standing, 70 to 90% would do it; 2)      But if the option is to push the 6 th person physically into the path of the trolley, 70 to 90% would not do it. As Sapolsky says: “The same numerical trade-off, but utterly different decisions.” So is the key difference is that we don’t want to get our hands dirty (literally)? They changed the option from pushing physically to push with a pole. No difference: people were just as reluctant as the push physically option. Did physical proximity to the

Increasingly, I’ve begun to encounter those awkward moments when the school book says something wrong, the kid detects it and asks if that’s really true. I am stuck between a rock and a hard place. Do I lie, “Yes, the book’s right”? Or do I tell the truth and risk the kid: -           Ask why she needs to learn something that’s wrong? -           Abuse the one instance as grounds to not believe/learn everything else? Take these two recent examples. In the word meanings section, the word “wagged” was defined as “waved from side to side”. “That’s correct for a dog wagging its tail”, my daughter countered, “but if I wave my hand to say bye,   that’s not called wagging my hand, is it?”. Fortunately, that one got resolved easily because she seems to have a feel that English is a crazy language. The other instance involved poetry where the answer to the question as to how a butterfly gets its colors was that it “takes” them from the flowers. “Really?”, she asked. See, I wanted

A while back, I heard this very interesting Guardian podcast on the magnetic north pole. In case you’re wondering, “magnetic north” refers to wherever the compass points. As opposed to “geographic north”, which is the top point of the axis around which the earth rotates. Edmund Halley convinced the British Navy to fund his idea of plotting points on the globe by comparing the deviation between magnetic north and geographic north. The obvious use of this was in sea navigation. In theory, that was a good idea. In practice, it didn’t work. Why? Because the data was collected over a few years. Trying to combine data collected in Year 1 with that from, say, Year 3 led to all kinds of weird errors! This may “sound like a failure of epic proportions” but it led to a major scientific finding. The errors led to the realization that the magnetic north moves continuously, ergo data from different times wasn’t consistent! Today, we know the extent of the movement over the last few cen

No triumph is preordained, unless one side is overwhelmingly stronger. Strike that: even then, the occasional David does defeat the Goliath. And yet, the fact that we don’t know the outcome until the fight ends doesn’t stop us from believing that the victor’s victory was preordained. And so we continue to write history as if the victor was known all along! I read this blog that cited the triumph of capitalism as an example of what many now believe as something that was inevitable all along. The blog mentions multiple examples that proves the reality of how things were really perceived while the capitalism v/s communism fight was still ongoing, even in the US: “(The US fought the disastrous Vietnam war) because it feared communism would succeed, not that it would fail – that communism could supplant capitalism… When Khrushchev spoke of “burying” and “overtaking” western capitalism, nobody laughed. The danger was a serious one. And the launch of Sputnik suggested to the world

When I was a kid, I remember reading that plants grow faster if they hear classical (Western) music. Turns out those claims “have never been substantiated by rigorous experiments,” says Richard Karban from the University of California at Davis. Ed Yong says those unscientific, yet hugely popular claims weren’t just wrong; they did damage : “They tainted the entire field of study, making scientists skeptical about the very notion of plants exchanging signals.” But now two researchers in Israel, Lilach Hadany and Yossi Yovel, have found that plants can hear the sounds of pollinators (aka bees). You may be wondering how someone could possibly know that, right? “In both lab experiments and outdoor trials, they found that the plants would react to recordings of a bee’s wingbeats by increasing the concentration of sugar in their nectar by about 20 percent. They did so in response only to the wingbeats and low frequency, pollinator-like sounds, not to those of higher pitch.” Fur

All of us have run into systems that worked perfectly upto a point but then broke down after that. Simply because some resource turns out to be inadequate beyond that point. Or to put it in engineering terms, the solution wasn’t “scalable” beyond a point. But have you ever encountered a scenario where a surplus of a resource caused a degradation of performance? If you’ve used the Internet, then yes, you’d have run into that scenario, writes Brian Christian in his book, Algorithms to Live By . Here’s how the modem was designed to work way back: 1)       You want to upload a really big file. 2)      The network is busy, so your modem can’t transmit just yet. 3)      So the modem stores the content in its buffer, planning to transmit whenever the network allows. Sounds good, right? What could possibly go wrong with that approach? Aha, when it was originally designed, memory was expensive, so how much data the modem could store in its buffer was relatively small. B

For people who want to read but just can’t seem to get (make?) the time, the authors of the Farnam Street blog suggested having a target of 25 pages per day . Additionally, they listed some ways to get the time for those 25 pages/day. Ryan Holiday raised a problem with the 25 pages per day approach: “Broken up into too many pieces, you’ll miss the whole point of the book, like the proverbial blind man touching an elephant. Those who conquer long books know that it’s not a matter of reading some pages before you fall asleep but rather, canceling your plans for the night and staying in to read instead.” Valid point, responded the Farnam Street guys. But the point of 25 pages a day was slightly different: 1)       Partly, it was to get over the starting problem that many face (“If I can't read for hours on end, why bother starting?”). 2)      The number was the minimum pages to read a day, not the maximum! The other problem with the advice was caused inadvertentl

Everyone complains about fake news. But Tim Harford worries more about the dangers of the over-reaction to fake news rather than fake news itself. Harford starts with some history of the term and its usage: -           The usage of the term “fake news” spiked after, no surprise, Trump won: “The interest was driven by astonished people looking for an explanation. Fake news was not the only scapegoat but it was, and still is, a popular one.” -           Over time, it has become a catch-all umbrella term for politicians making misleading claims and media outlets with a political slant as well. -           Yes, it is “cheap to invent lies, and eye-catching lies are a reliable source of clicks and thus advertising dollars”. This inevitably has led to resentment from journalists, who use the term indiscriminately to try and tar every source other than the “established” ones. -           When people rant about fake news going viral, they forget a less mundane explanation for

Anytime we plan to call someone in another country, whether for work or for personal reasons, we have to check what time is it out there . Call center operators work night shifts for the same reason. But at least the time difference between two places remains the same, except for the Day Light Saving adjustments made by some countries. Imagine how much tougher it would be if the time difference between two places changed on a daily basis ! That’s exactly the scenario that NASA engineers working with the Red Rover module on Mars have to face. Why? Because, as Nagin Cox, said in her TED talk : “Our day is 24 hours. It takes Mars 24 hours and approximately 40 minutes to rotate once. So that means that the Martian day is 40 minutes longer than the Earth day.” But why should that matter? Aha, because the plans/instructions for the next Mars day have to be relayed at (Mars) night, while the Rover “sleeps”, i.e., takes a break from her operations on Mars to recharge her batteries. L

Here’s how Deep Blue, the computer that famously defeated (then) world champion Garry Kasparov in the pre-Big Data age “learnt” to play, writes Steven Strogatz: “Principles, which have been refined over decades of human grandmaster experience, are programmed into the engines as complex evaluation functions that indicate what to seek in a position and what to avoid” Sure, it was insanely fast, but was “utterly lacking insight”. Today’s chess engines “learn” by playing and looking at games, “discovering” for themselves patterns. It’s how all of machine learning works: the more and better the data you throw at it in the learning phase, the better it gets. With machine learning techniques, computers often appear to identify insights never known to humans! Take this description of AlphaZero, Google’s machine learning algorithm world champion at chess, shogi (Japanese chess) and Go. It crushed its “programmed” learning counterpart, Stockfish by: “thinking smarter, not faster; i

If Salman Khan is found guilty and convicted, we expect producers of his ongoing films to suffer. But nobody asks for his already released or completed films to be taken off the stores or TV channels or the Internet, right? But that’s India. Contrast that with what happens in Hollywood these days, laments Lionel Shriver: “Artists’ misbehavior now contaminates the fruits of their labors, like the sins of the father being visited upon the sons. So it’s not enough to punish transgressors merely by cutting off the source of their livelihoods, turning them into social outcasts, and truncating their professional futures. You have to destroy their pasts. Having discovered the worst about your fallen idols, you’re duty­bound to demolish the best about them as well.” Worse, this happens without even waiting for a court verdict. Just an accusation can set this chain off: “In numerous instances during the #MeToo scandals, accusation has stood in for due process, and criminal offe

There’s this BBC podcast called The Infinite Monkey Cage that had an episode on the immune system. I found these aspects fascinating: -           The body produces cells with receptors that “protrude out” of the cell. These receptors can lock onto specific cells of the intruder the way a jig saw piece fits into only one other piece. If they find a fit, they know it’s an intruder and go into attack mode; -           Ok, but don’t the intruders/germs mutate over time? How does the body create new receptor shapes for the mutants? Aha, the body shuffles around the genes thereby creating random shaped receptors on a continuous basis. And the rest is luck: the hope that one of those randomly shaped receptor would lock into the mutated germ! -           That’s impressive, but couldn’t such randomly shaped receptors accidentally turn out to be a match for a body part? Would the immune system then end up attacking the body itself? Great question, and the answer is that the body has a