Buysiders.com » Science

DLD Conference 2012

IP — Sun, 22 Jan 2012 17:59:44 +0000

DLD 2012 has started today in Munich and runs until Jan. 24th. In it, people as diverse as Sheryl Sandberg, Arianna Huffington, the Dyson family (Freeman, Esther and George Dyson) and Hiroshi Mikitani (Chairman & CEO of Rakuten) share their views on what matters to them. The themes are also varied – augmented reality, the future of cities, citizen science, epiphanies – and the program is packed with interesting talks and panels.

In the age of multi-disciplinary events, this is one of the best. You can keep track of it live on Livestream.com, check out the event’s blog on Tumblr, its Twitter feed or Facebook page. Enjoy!

The dangers of models

IP — Fri, 02 Dec 2011 18:01:39 +0000

Less than 24h after publishing our rant on Economic models (it’s the first story there), we get John Kay’s brilliant piece in our inbox – “A wise man knows one thing – the limits of his knowledge” – courtesy of the Financial Times. It is the ultimate summary of the many dangers of modelling in general, not just in economics – among which dangers we count over-complicating things, but most importantly over-estimating a model’s value as predictive/forecasting tool. In fact, as the article argues and as we’ve seen countless times, we tend to over-estimate models even in their ability to analyze the past, especially if one is asking the wrong questions. And since we’re taking a turn to the more conceptual in this post, we cite a few quotes and allegories by Taleb and a long article on Edge.org by Emanuel Derman, himself a former quant trader at Goldman Sachs.

“Nothing substitutes for thinking”, as Munger and Buffett have said, and someone has also said that “not everything that counts can be counted, and not everything that can be counted counts.”

Taleb’s book “The Bed of Procrustes” is full of aphorisms, allegories, metaphors and whatnots. In a sense his entire body of work is about the dangers of false empiricism & science, and poor knowledge of, well, everything. And yet we select below just a few aphorisms that apply to this case.

“Suckers problems: when the map does not correspond to the territory, there is a certain category of fool, the overeducated, the academic, the journalist, the newspaper reader, the mechanistic scientist, the pseudo-empiricist, those endowed with what I call epistemic arrogance, this wonderful ability to discount what they did not see, the unobserved who enter a state of denial, imagining the territory as fitting his map.”

More attentive readers will remember our “map allegory” in our Q3 2008 report, a time of extreme market turbulence:

“At IP, we are lucky to be able to draw on a simple “map”, which we take off our pockets on each and every occasion. Or better: it is usually more efficient in moments such as now, when the rationality of a battered and bruised market simply wanes.

Independence. Rationality. Patience. Discipline. Conservativeness. Long-term horizon. And the mandate to buy stakes in good businesses that have honest and competent managers, requiring a discount, a significant safety margin compared with its value. It can’t get any simpler.”

The good news is that this map also works in good times.

Other quick Taleb quips:

“They think that intelligence is about noticing things that are relevant (detecting patterns); in a complex world, intelligence consists in ignoring things that are irrelevant (avoiding false patterns).”

“Randomness is indistinguishable from complicated, undetected, and undetectable order; but order itself is indistinguishable from artful randomness.”

“To bankrupt a fool, give him information.” — Absolutely fantastic!

Finally, the long article on Edge.org (it is rather redundant to use the word “long” before mentioning an article on Edge.org). Emanuel Derman was the Head of the Quantitative Strategies Group, in the Equities Division of Goldman Sachs. If tempted to distill the article into two paragraphs, they would be:

“Theories deal with the world on its own terms, absolutely. Models are metaphors, relative descriptions of the object of their attention that compare it to something similar already better understood via theories. Models are reductions in dimensionality that always simplify and sweep dirt under the rug. Theories tell you what something is. Models tell you merely what something is partially like.”

and

“Given that finance’s best tools are shaky models, the best strategy is to use models as little as possible, and to replicate making as little assumptions as you can.” — This particular sentence is in the last part of the text, section 9, and immediately after this quote Mr. Derman lists a few great tips for those of us who have to use models – in summary: make it simple, don’t over-reach, and most of all don’t idolize neither the model nor the process of modelling.

Follow-up: reading lists and the speed of light

IP — Thu, 24 Nov 2011 06:00:56 +0000

Not to be outdone by our recent post about reading lists, Booz & Co.’s Best Business Books 2011 feature is a sprawling selection of books covering Management, Economics, Marketing, Ethics and other business-oriented fields. Some of the books were already in our wish lists, others were pleasant surprises. There’s a 35-page PDF of the feature here, but it requires (free) registration. Inside we highlight the second follow-up.

Finally, it appears that the speed of light (as a limit) is indeed in question as recent studies were able to repeat the first experiment’s results. We’ll certainly hear a lot more about this in the near future, as there is still plenty of doubt. In the meantime, for those concerned with the state of a world in which not even the speed of light is a certainty, we suggest a further cause for concern: our degrees of separation, which used to be 6, are now down to 4.74, according to researchers from the University of Milan and Facebook. Yes, apparently Facebook has a research department.

World welcomes “Citizen 7 Billion”

IP — Mon, 24 Oct 2011 19:09:20 +0000

Or doesn’t it? The UN expects “Earthling #7 Billion” to be born this week – here’s a link to the report itself, to be released tomorrow (Tuesday, Oct. 26th 2011). Yet it’s just a symbol to spark a stream of articles about overpopulation – demographics, at this level, is open to much debate… and noise. Clive Cookson in the Financial Times had a short piece with a brilliant infographic, and the New York Times today has an entire debate series about it, with at least six articles. One of them is by Joel Cohen, who also has the honor of introducing the debate in this relatively balanced article. But there’s more.

BigThink.com also jumped on the opportunity and highlighted this 2min 40s video with Professor Dickson Despommier from Columbia University. From there you can view many other videos with him.

Going back to the NY Times, this blog of theirs – Dot Earth – is a pretty interesting source (we are linking to a specific post but you can navigate from there – and the lecture seems to be worth the time, judging by the first 10-15 minutes. In fact, we’ve embedded the whole 80-min video below):

Wrigley Lecture Series – Andy Revkin and Braden Allenby from Sustainability @ ASU on Vimeo.

Faster than light

IP — Fri, 23 Sep 2011 09:00:22 +0000

In what is admittedly light reading after yesterday’s trouncing of the equity markets – but loyal readers have read our “Don’t panic” post, right? – there’s bound to be intense debate around this proposition by scientists: it IS, apparently, possible to achieve speeds in excess of the speed of light in a vacuum (at least for neutrinos, that is). It was considered impossible until now (you know, E=mc²) and, if confirmed, it may change quite a few theories. Talk about a ‘black swan’ of enormous proportions. By the way, remember when phosphorus ceased to be a life-defining element? This is bigger, by far. It all hinges on the margin error: the scientists who disclosed the “discovery” have estimated it to be one-sixth of the discrepancy found, but they could be wrong. In fact, a large margin of error is precisely what destroyed a similar claim in 2007. The CERN scientists claim to have “checked and re-checked” but have asked US and Japan labs capable of running roughly the same experiment to confirm the findings.

It’s one of those occasions where we’re supposed to remind you to seek the source material, remain skeptical and filter the signal from the noise. However, the ‘multidisciplinary geek’ in us wants to see some scientific foundations shaken up.

Taleb and the planning fallacy

IP — Wed, 21 Sep 2011 09:00:53 +0000

We’ve been arguing that “to finish first, first you must finish“ for over twenty years. We’ve also recently argued for cash as a strategic weapon with both defensive and offensive features: it is as much “cushion” as it is “cannon”. Nicholas Nassim Taleb has given a lecture recently at the Univ. of Pennsylvania (the 68-minute video is embedded inside), and the main takeaway is this: “(…) you live long by not dying, you win in chess by not losing—by letting the other person lose. So negative investment is not a sissy strategy. It is an active one.” — It is always interesting to hear him as a reminder of the constant mental traps we fall into, be it “fooled by randomness”, the difficulty of dealing with “black swans” and the concept of fragile/ anti-fragile systems. It also highlighted the planning fallacy, something David Brooks wrote about even more recently, so we’re taking the opportunity to post both together. H/T Farnam Street Blog for linking to both stories over the weekend.

Here’s the embedded video:

Praising the multidisciplinary approach

IP — Mon, 11 Jul 2011 22:37:20 +0000

Sunday’s post in the great “Epicurean Dealmaker” blog opens with a Richard Feynman quote, something that guarantees our attention, but it’s a great post in itself. In his comments to a Tim Harford article in the Financial Times – also worth the time, by the way – “Mr. Epicurean” highlights that science is increasingly about multidisciplinary collaboration, which brings its own set of pros and cons. The risk, simplifying the arguments a bit too much, is that scientists are becoming so specialized that nobody truly has a “bird’s eye” view of what’s being done – i.e. no one can question the collective idea because no one can know enough about each piece of the puzzle. And if the scientific method is all about questioning, what does that mean for the future of science? No easy answers, but the quest to achieve as much of a multidisciplinary knowledge as possible – as argued by Feynman in the introductory quote and by Charlie Munger repeatedly over the years – gains importance day by day.

Here’s a quote dear to our hearts: “the most important phrase in science is not ‘Eureka!’, it’s ‘that’s strange’… “

LINKS:

Richard Feynman at his best in this 49:37 video, “The Pleasure of Finding Things Out”. There’s a wealth of stuff on Feynman online, and we wouldn’t do him justice by trying to quickly compile some list today. If you don’t have 50 minutes readily available, jump to the last 7 minutes (starts at 42:54) – but come back someday for the rest. This last bit starts with pseudo science and “experts” – about how hard it is to really know something. It seems that smart, experienced people are extremely careful in the process of forming their views, opinions and “truths”. It continues with the need to keep an open mind – the need not to “marry” your ideas and frameworks. He says about physics: if it turns out that there’s one universal law, so be it. If it turns out that it’s a multitude of laws, so be it. It doesn’t matter, it’s important that nature reveals itself as it is, not as one group of scientists or another wants it to behave. Intellectual integrity: you must doubt, you must question even yourself. “When you doubt, you ask, and it gets a little harder to believe. I can live with doubting and not knowing. (…) I don’t feel frightened about not knowing things.” – that’s the spirit.

Great books and anti-libraries – About things you realize you don’t know and unread books.

Hammers and nails – About avoiding the trap of the “man with just a hammer”.

Edge.org’s question for 2011 – One of the best answers to their 2011 question is “independent thinking”.

Multidisciplinary approach and communication – Should we seek to be multidisciplinary or is it enough to build a multidisciplinary team?

Classics: Psychology of Human Misjudgment – Charlie Munger’s timeless speech (it can also be found in the more recent Munger link above).

Winning by losing

IP — Wed, 29 Jun 2011 20:06:08 +0000

Interesting preliminary finding, as reported by Scientific American, by researchers looking at 18,000 professional basketball games: ending the first half losing by one point actually increases a team’s chance of winning the game. If a team was winning by 4 points at halftime, the chance of winning the full game was 70%. If the advantage was 6 points, the chance of winning jumped to 80%. However, when the margin was 1 point then the team had a better chance of losing… The study seems well done but, of course, take it with a grain of salt before applying it to motivational “tools” inside your company.

Back to the study, whose abstract can be found here: they also looked at >45,000 NCAA (US college-level basketball) games and found the same effect, albeit smaller.

The hypotheses to explain these findings were loss aversion and diminishing sensitivity: that is, losing by 1 point made teams seek to reverse the situation and losing by a “lot” made teams care less or subjectively try less hard as those losing by 1 point, tied or winning. The researchers then devised a “game” to test subjects’ motivation in similar situations and they did find that those losing by little increased their efforts.

Even more interesting is the finding that people with higher confidence in themselves increased their efforts more than those with less confidence. In the article’s own words: “Being slightly behind won’t increase motivation if you think you’re not capable of turning things around. And, similarly, a large deficit might also be motivating if you think you’ve got what it takes to stage a glorious comeback.”

Motivation isn’t an easy issue – not even as a study field – so it’s only natural that it seems so hard to instill it in oneself and/or others. When you mix the issue with confidence, it’s hard to know what comes first – this study seems to say that confidence is the fertile ground for motivation.

More on Education

IP — Fri, 13 May 2011 14:17:27 +0000

BigThink.com has two interesting videos on Education. We’ve written before about Salman Khan and the future/ quality of Education, and he is the subject of the first video, a much shorter version of the TED speech we posted in the link above. From the first video we navigated to another video – this time with Harvard’s Louis Menand, whose full interview can be seen here. Both videos embedded inside.

Customization/data analytics coming to Education is great news.

Salman Khan’s 2:32 video on BigThink:

Louis Menand’s 4:13 video at BigThink:

Making real “stuff”

IP — Mon, 25 Apr 2011 02:27:18 +0000

UPDATE (April 25th, 16:18): Incredible 4-minute video posted inside, sent by a reader. Many thanks!

Very interesting story in the NYT this Sunday: how the Web is making it easier to fund/ develop companies that aren’t web-based at all, and actually make tangible products. It reminded us of a conversation we had internally last Tuesday about “open-source chips” and other electronic hardware available for a fraction of the cost of a few years ago, and how much easier it is nowadays to build prototypes. All that was needed was a way to organize like-minded people and help their ideas find interested investors, however tiny the capital needs. No longer. In the words of an MIT professor, “The situation resembles the way that anyone with a laptop and an Internet connection can now start a Web-based company” – and we all know what innovation and wealth creation can come out of that “mold”.

TED Talk by Marcin Jakubowski on “Open-sourced blueprints for civilization” – Feb 2011 (4:11)