Sunday, November 29, 2009
Saturday, November 28, 2009
- Sharing information in scientific research: yes/no/when. Interesting, but the analysis here is applicable only in the context of some sciences (arguably, not most). Moreover, sharing by means of co-authorship is discussed at best indirectly (if one is willing to expand on their repeated interaction game thread...). But my major "disagreement" has to do with their upshot: my conjecture is that know-how sharing is, ceteris paribus, over all disciplines, U-shaped in the degree of competition (and, since I also believe that, overall, very high competition dominates very low competition equilibria, you already know what my prior of the ideal is). Remains however an open empirical question for now.
- We are what we ate: Tony Judt's culinary autobiography. Reminds me that (fortunately) I have only got to know top Indian restaurants in London (in general, Londonese Indian cuisine is probably, up to now, my favourite in the world; though surely there is a variance): a lot must have changed.
- David Brooks on "the other education": concise and well written. I usually call this second education "exposure". The vast majority of people I've met hardly had any.
- Cheaptalk on the election process of Econometric Society Fellows. Not extremely surprising or, for that matter, singular within Economics; after all, they almost forgot Hurwicz for the Nobel Prize..., for largely the same reason: most of the people who used to propose and lobby for him died at some point... I think Ely is right: focus on the young people-- at least that would be an attempt to solve the problem for the future... And yes, for potential candidates why wouldn't you just look to the Econometrica editions (Lones Smith's suggestion in the comments)...
- Music critic review: mode d'emploi, by John Adams.
- Philip Greenspun on universities and economic growth; via Razvan, on Ad Astra. First impression: he writes much and he misunderstands a lot; particularly the Economics of it all (no, he is not qualified to understand what is clear and what is controversial in Clark's book, to give but one example). Also, doesn't seem to realize (not sure whether qualifying this as voluntary misrepresentation would be better or worse...) the difference between statistical and anectodal evidence. And, in general, he doesn't seem to have decided whether his target is to make people incensed at or interested in what he has to say. BUT, although he errs nearly everywhere else, I agree that a. much change is needed in the way teaching in most universities is done nowadays (see also the 3rd bullet point here on opinions on the value of college education & all that jazz); b. he has some very decent ideas there (others had/have them too) and c. these changes would not cost too much, with the benefit very likely to outweigh that cost. (Probably) Inadvertently, Greenspun is actually arguing for a "Japanese approach" (which the Japanese apply to both teaching and on-the-job training): give a rather broad ("customer-based", if you want some context) training, be able to/ focus on study/work in teams, always help the new/junior ones etc. etc. There is however a known problem with the (standard) Japanese perspective to (life-long) education that I am not sure Greenspun is aware of... Simply put, you really do not want to give no/wrong incentives at the very top of the ability distribution.
- One of the two prediction markets on the 2009 Romanian Presidential Election is now closed and cashed-out (the other one is also "closed", but waiting for the final results on Dec 6th), as the official First Round results of that election are out. Several participants won (virtual) money (yours truly included), but the congratulations go to Dan, as he is the one who won the most (a fortune!). Which means that he is obviously going to pay for the (very good: e.g. the French on this list?) wine, with the occasion of our next meeting :-).
Wednesday, November 25, 2009
Monday, November 23, 2009
Friday, November 20, 2009
- Acemoglu on how nations can get rich: "Fix incentives and you will fix poverty. And if you wish to fix institutions, you have to fix governments"
- Blogossary. With some definitions completely redundant.
- Carlsen is finally right where he should be. For now in Blitz.
- Diploma mills and degree mills. Romania has plenty, Denmark at least one very famous case (though Wikipedia is too slow for that).
- Caballero on sudden financial arrest. I don't agree with everything: the moral hazard is not merely fuzzy reasoning in the context.
- What's up with all the vampire hype: Tyler Cowen's answers. If you want the perspective of a vampire on why vampires are popular, you can meet me in private.
- Very little time left for you to trade Romanian Potential President stock. You have to act NOW! :-)
Wednesday, November 18, 2009
Irving Berlin and Cole Porter were two of the great experimental songwriters of the Golden Era. They aimed to create songs that were clear and universal. Their ability to do this improved throughout much of their careers, as their skill in using language to create simple and poignant images improved with experience, and their greatest achievements came in their 40s and 50s. During the 1960s, Bob Dylan and the team of John Lennon and Paul McCartney created a conceptual revolution in popular music. Their goal was to express their own ideas and emotions in novel ways. Their creativity declined with age, as increasing experience produced habits of thought that destroyed their ability to formulate radical new departures from existing practices, so their most innovative contributions appeared early in their careers.
This is the abstract of David Galenson's new paper on the two creativity patterns in songwriting. Extremely interesting, very convincingly argued, and at the same time a crash course into the musical biographies of Berlin, Porter, Dylan, Lennon and McCartney. See also a previous entry on Galenson's research.
Monday, November 16, 2009
Sunday, November 15, 2009
- One of the reasons I just love reading Landsburg: he is acid and funny. Here's something that hopefully will open a longer exchange between him and Krugman: I really think Steve Landsburg is the proper counterpart to Paul Krugman in any debate (NB: Mankiw is great, but too serious and not engaging--understandably-- enough in the type of debate Krugman seems to love). And I tell you beforehand that I shall bet all my money on Landsburg :-). This for instance should be kept for posterity: "But sometimes I think Paul Krugman is out to top them all, by excelling in two activities that are not just disparate but diametrically opposed: economics (for which he was awarded a well-deserved Nobel Prize) and obliviousness to the lessons of economics (for which he’s been awarded a column at the New York Times)."
- Who should go to college and who should pay for it: with many specialist opinions (via MR). Signalling seems to be dominating as point of view, though I myself think that university also has a genuine positive effect (I agree that matters most for the students in the top of the ability distribution). Else, all might be boiling down in the limit to competition between cats with fraudulent diplomas (via boingboing, via MR).
- Kirman has been writing about these things for quite a while now (with high frequency recently), and almost always on such a revolted tone; these things are happening within Economics, for quite some years now, much of what he mentions is mainstream or closeby (think of social interactions and networks, herding behaviour in finance etc). Plus arguing for discarding these old and well known models (my belief is that most serious economists are very well aware of their limitations for each specific context), just for the sake of discarding them, is madness (who is ideologue here?...). I really don't think this is the way one should argue for different approaches. All my respect for Alan Kirman's scientific work, but with his (especially recent) comments he does seem to go/fall a long way in the direction of non-Economists misinterpreting Econ fundamentals (first bullet point), or, worse, of the several nonsense people writing now and then in The Guardian, e.g. here (6th bullet point) or here.
- The "10 simple rules for scientists" collection at PloS Computational Biology (via MR). Some are redundant, others are great; within the latter category, I think everybody should memorize the "10 simple rules for doing your best research, according to Hamming" (though ok..., rules 5 and 10 are very much location-specific :-)).
- La mémoire refoulée de la Roumanie-- le communisme reste "une abstraction".
- Tim Harford on the "Jamie Oliver Feed Me Beter" experiment in UK schools. Features recent econ research analysing effects of that experiment, by Michele Belot and Jonathan James; you can download a draft version here. The preliminary results suggest that good food has considerable positive effect on educational outcomes. Extrapolating: so stop telling me not to spend my money on Michelin-starred restaurants :-).
- The young Viking is on top of the world right now, according to the unofficial 14 Nov ELO ratings (though I am dissapointed he let Kramnik win the Tal Memorial this year, Carlsen tying for second with Ivanchuk). See also a (happy) Norwegian article on the matter(via Susan Polgar).
- I think "wine critics/commentators" should move-- what they should have done a long time ago-- to identify / classify wine quality ranges, rather than preserve the current practice of grading on a 1 -100 scale. In any case, the last paragraph in this article is the one to retain.
- Less then a week left before the Prediction Markets on the upcoming Romanian Presidentials close: don't waste the opportunity to trade your favourite Presidential stock! :-).
Saturday, November 14, 2009
First, I think this is a very welcome, very open interview (several questions/comments are just great, congrats to the interviewers!) and it is extremely interesting to see the opinion of this great mathematician (inter alia, Fields Medalist in 1982) on a wide range of topics. The pros: I completely agree with Connes's view on the distinct (ir)relevance of String Theory for Mathematics and respectively, Physics. I also like his (humorous) detachment from being considered the guru of NCG (of which unfortunately I know currently epsilon, albeit once I was almost sure this is what I wanted to do...) and from the tendency of always looking for / looking up to the one mastermind, in general, in any (sub)discipline. So no more on those, read for yourselves in the transcript of the interview. What I don't quite agree with is summarized below:
- resources (money) in research are not important (Connes's context has to do with the large interest/funding in Bio-Mathematics; he actually says "nothing", which I take as far stronger than "not important":-)): to the extent that it shapes incentives, I think it is actually very important. Intrinsic motivation is (the most) relevant, but it is not everything. The marginal (very able-- let us simplify) scientist can be moved into one direction or another by means of designing proper (or improper; but then again, who is to decide what is proper/improper in the context: I think we ought to take the view that lots of money is being thrown in one direction, because there is a lot of interest in that particular direction) extrinsic rewards. That being said, I personally (also) think there are a lot more interesting things in/to Maths than Biomaths :-).
- the European academic system is better than the US one. Hmmm, this is an endless debate and, as always, the truth is probably somewhere in the middle. Inter alia, it goes back to whether you need/want tenure or not in the academe (see for instance such a debate I've earlier linked to, especially within Economics) and to what goals you expect researchers to meet. And this also goes beyond one or another discipline, although it is perhaps interesting to discuss it indeed in the light of fundamental Mathematics, given its very abstract nature. Now, Connes believes that a system such as the French CNRS (which possibly is in the process of changing since 2005, when this interview took place) is perfect for mathematicians working on extremely complicated issues, that take years and years and years, since they are insulated from being subjected to those "n publications" requirement per year and in general from the eternal harassment of frequently showing how you compare to your peers, something specific to the top US institutions (Connes dismisses that the US places are ultimately inherently better in producing top scientists, because they get all the top European people-- not (entirely) true and to a great extent working eventually against his thesis, e.g. need to justify preferences of those very high European achievers for the US places, but let us not get also into that). The potential problem (sacrifice), as acknowledged by Connes, is the cost of such a practice, given that a lot of people might end up not producing anything and that the vast majority of them will be very far from getting Fields Medals or similar recognition among their peers... I say that the main problem is who bears that cost, namely the taxpayers here; and the public (not all of them having the same goals as Connes or as the specific, minority, group of the scientists, in general) is justified in knowing and assessing (whenever it so pleases) where its money is going and what precisely it pays for (if the funding is private, all this discussion has a completely different flavour-- remark that the US top academic places are privately funded, while all European examples Connes mentiones are public institutions; in my view, this again tips the balance towards the US academia). Related, but extremely surprising, Connes seems to be nostalgic after the Soviet Union academic system, but I think he deeply confuses things-- anyway, let us just say for the sake of this brief post that, fortunately, France was never quite like the Soviet Union, despite its tendency to lean extreme left, particularly within its academe... As for the claim that the Soviets would have been far ahead US and everybody else, if their system remained in place, I guess we'll never know (though I have opposite priors). And I think it is better we don't... So I am rather dissapointed that one of my idols in Mathematics has/had (this was '05) such, hmm: uninformed, views. But then again, I've always thought Economics (Not Politics. Politics is just a surface, not relevant in the long run, ultimately all boils down to Economics-- really!) is far less intuitive than Mathematics or Theoretical Physics :-).
I am sure one can go on and on, but I trust the main ideas are all outlined above (read also between the lines).
Friday, November 13, 2009
So the loud cough, most likely completely unconscious, is a way of saying “I can’t handle this, folks. You all may be crowding round Mahler’s deathbed for one final intimate confession. You may be letting Debussy whisper opium secrets into your ear. Perhaps you like being ravished by Takemitsu’s lush penumbras. BUT I AM OUTTA HERE!"
Excellent!-- I don't think anyone could have written this better than Adams.
PS. Unfortunately, I think there are even more types of concerto coughers. For instance the person who just cannot stop coughing-- desperately wants to, but simply cannot-- somebody who should have stayed at home, or even better, in a hospital. Provided s/he does not die before, s/he is usually gone after the break...
Thursday, November 12, 2009
- On the poetic prehistory of development economics: a very interesting post by Adam Martin on Easterly's Aid Watchers blog.
- They cannot be blamed for defending/arguing what's in their best interest (click on the British flag in the upper right corner to get to the article in English, if you do not see it immediately through the link above); however, in my opinion, their general case is overstated: most of the translation services as hitherto understood are redundant. There is still a case (and a premium) for translation services from/to Arabic, Mandarin, Japanese & the like, but certainly not for European to European languages, and certainly not in countries from Scandinavia or Benelux...
- Top 100 best films of the decade (?). To be honest, I have seen to date only one film from their top 10. But I did see quite a few from the range 10-20 and, if this hierarchy makes any sense, I ought to definitely see anything that beats the movies at positions 16, 20 or 21, which are all in my very own all-time top 26. Quite impressive is also that Romanian director Cristi Mungiu's film "4 luni, 3 saptamani si 2 zile" is number 14 in this ranking (this movie is in my priority to-see list). See also my earlier entry on The December Children.
- Did you forget about the Romanian '09 Presidential Election Prediction Markets?! Participate: it's fun!
Sunday, November 08, 2009
Grazie, Elena-- it is indeed high time I was reminded of Tenco's superb voice and style.
Saturday, November 07, 2009
- "Famous economists" easy quiz. In fact too easy, as presented, because you can see the questions before you start the clock. So, in order to seriously test yourselves, just start the clock and then read the questions (you have 3 minutes in total). Took me 1 min and 14 seconds to answer all questions correctly, because I needed three trials for one of them-- yes, yes, I admit I have not read at all that book :-). Cowen claims it took him only 49 seconds. By the way, in case you didn't know of it, here's a great resource on the history of economic thought.
- The Levitt vs. Heckman leitmotive (actually the rest of that Univ of Chicago Magazine article is more interesting; in general, I think people put too many resources into this sort of personalized academic fights). Earlier on the same topic (3rd bullet point). See also an essay on the economics-made-for-fun genre, kind of vague / unfocused, but some parts are well worth your time (such as the beginning summary of the books in this econ-made-for-fun area)
- The three (only?!) habits of highly irritating management gurus: very well written, welcome article from the Economist. We love it!
- A nice recap post of Terry Tao on the "no self-defeating object" argument in mathematical proofs.
- Kahn has an interesting argument, but I find him overly pessimistic. I also think Cowen is somewhat out of touch with what is really going on with academic Economics. Very briefly, here's what I think: nothing is amiss with the current direction of research in Economics (as long as these ideas/ forecasts do not turn into self-fulfilling prophecies...)
- Best US cities for classical music, in '09. Worldwide, I would place London, Amsterdam and Vienna somewhere in top 5. Not sure about Tokyo yet, since I was for too little there (and still have to put my impressions on paper... well, blog).
- Pure selection of 100% go-getters, if you ask me.
- Don't forget about the Prediction Markets on the Romanian 2009 Presidential Elections-- some seem to be making already (fictitious) fortunes there! :-)
Friday, November 06, 2009
PS. The hilarious thing is that some would like to add even more journals that nobody reads/cites to these two categories.
Tuesday, November 03, 2009
Sign up and trade!
PS1. Hope I didn't make any mistakes; I put these up in less than 5 minutes :-).
PS2. Although you might really not share this opinion, I 'conjecture' that, if the outcome is not clear in the first election round, the future President will eventually be the first round's runner-up. Wanna trade against that? See above :-).