Tuesday, December 01, 2009

The Grand Prize for Epic Fail goes to ...


... the Infosys Science Foundation!

For its immense failure -- Epic Fail! -- to find a single Indian researcher for the Infosys Prize in Engineering and Computer Sciences.

The Infosys Science Foundation is being recognized for:

  1. Its failure to grasp the magnitude and the extraordinary symbolism of its assertion that there's no one in India deserving of the Engineering Prize right in Year One.

    [It's not as if this Prize has already been awarded to some five people in the past, and the jury is now unable to find anyone in the same league as those past winners. This is Year One we are talking about, and the jury has declared there's no one who measures up to The No One!]

  2. Its failure to realize that if no one deserves this Prize in Year One, its Prize in Year Two is in deep doo-doo.

    [If the Prize in Year Two goes to someone from the present set of nominees, the question would be, "What changed in the last year?". If it goes to someone else, the question would be, "Why was he/she not considered last year?" Either way, the jury for Year One ends up looking incompetent, if not downright stupid.]

  3. Its failure to, simply, get Prize-worthy nominations.

    [What good is a jury if they could not reach out to people who can nominate Prize-worthy candidates? I mean, what good is your Prize -- that too, one that carries the highest monetary award in India -- if all you can attract is spam?]

  4. Its failure to see the irony in no one getting the Engineering Prize from a foundation set up by what is arguably India's leading tech company.

That was one moronic decision from you, Infosys Science Foundation! Perhaps we should have expected a screw up from you, but we could never have expected such a big one right in Year One.

Congratulations!

Monday, November 30, 2009

DST nails an IIT-KGP professor


The verdict:

India’s leading science funding agency has withdrawn a prestigious fellowship it had awarded last year to a professor at the Indian Institute of Technology, Kharagpur, after an investigation charged him with scientific misconduct.

Who:

Suman Chakraborty, a professor of mechanical engineering at IIT-KGP.

Why:

Two panels of experts set up by the DST found that Chakraborty had portrayed research he had completed before 2005 as the outcome of a DST-supported project which he had undertaken between 2005 and 2008, sources told The Telegraph.

His DST-supported project involved the prediction of how biological tissues interact with lasers, a science that may lead to new strategies to fight cancer. In a project completion report submitted to the DST earlier this year describing his work during the 2005-2008 period, Chakraborty had reproduced verbatim his own research paper published in a scientific journal in 2005, according to the findings.

G.S. Mudur of The Telegraph has the full story.

Some quick observations / comments:

  1. First, a disclosure: Suman Chakraborty is an alumnus of IISc's Department of Mechanical Engineering, where he finished his thesis in record time (IIRC, less than a year). I have had casual chats with him a few times when he was a student; I do not recall any interaction with him after his graduation.

  2. Chakraborty has been accused of passing off his old (pre-2005) research as something that was done under a project whose funding started only in 2005. At least two panels have found him guilty of this charge. From The Telegraph story, it appears that Chakraborty himself is not disputing the fact behind the charge.

  3. I know of situations where people do quite a bit of groundwork before the project begins officially; they are then in a position to start publishing within a short time after the project starts. Chakraborty seems to have taken his groundwork to publications well before the official project start. In his mind, the whole thing -- his groundwork as well as the work he did after the project started officially -- may have formed one long continuum, but the funding agency (and the people it appointed to investigate this issue) has not bought into this view.

  4. In comparison with the "high crimes" in science -- fabrication, falsification or plagiarism -- Chakraborty's offence is a minor one. If I may use an analogy here, he has not been accused of murder or manslaughter; he has been found guilty of a Section 420 offence.

  5. Section 420 is an apt analogy here: he has been accused of cheating the funding agency. The penalty for it is in the nature of a fine -- the Swarnajayanti Fellowship, which he won before his offence blew up, has been withdrawn.

  6. This penalty appears to fit the offence here; it's far lower than what the 'high crimes' attract. [For example, fabrication and falsification might attract an outright dismissal, while plagiarism is punished with not allowing any student to come anywhere near the culprit.] While he has lost out on getting a grant from DST for some years to come, his right to work with students (as well as to seek funding from other agencies, including private industry) has been left intact.

  7. The Telegraph story talks about how this case has led to a debate among Indian scientists. I don't really get what the 'debate' is really about; maybe it's it's about whether the punishment is appropriate.

    If Chakraborty is really guilty as charged, not punishing him would make the funding agency look foolish. I mean, someone cheats you, and you still go ahead and reward him with a Swarnajayanti? WTF?

    Chakraborty has been unlucky in that the agency that felt cheated and that which made the Fellowship offer happened to be the same -- DST. The Swarnajayanti might still have been his if it was awarded by a different agency.

  8. Frankly, I'm surprised that DST's project review panel caught Chakraborty on this offence; typically, end-of-project reports get only a cursory look to see if the broad objectives have been met. The scrutiny that's required for catching this sort of offence must have been pretty deep indeed. Either the panel was amazingly effective (very admirable, but unlikely) in its scrutiny, or its members must have had some other reason to subject Chakraborty's project to a higher level of scrutiny.

  9. It looks like this stuff -- the investigation, the verdict and the Swarnajayanti withdrawal -- happened several months ago. Due to lax (or, lack of) disclosure norms, DST hasn't made its decision public. It is this lack of disclosure that makes The Telegraph story read like a major scoop.

    Compare this with what the NIH does: it discloses on its website, as a routine matter, the results of all its misconduct investigations -- see this example.

Sunday, November 29, 2009

Talent, Expertise, Excellence


Here is a bunch of links:

  1. Daniel Chambliss: Mundanity of Excellence. A classic paper on how experts differ from novices -- based on a long-term study of competitive swimmers at different levels. There's a fabulous section on why 'talent' does not lead to excellence. [You might want to get hold of the full original from the source; requires subscription, though.].

  2. Greg Downey at Neuroanthropology.net: Talent: A Difference That Makes A Difference. Here's an excerpt from the section on "What is talent and how to identify it":

    ‘Talent’ or ‘potential’ are ways that some of us think about inequality in ability, or variation in the way that different people seem to benefit from training. ‘Talent’ is alleged a potential trait, a symptom of nascent ability, a foreshadowing of future greatness, or a way of explaining someone’s early achievements or performance advantage. On the other hand—paradoxically—the concept of talent is a way of understanding why some experts are more proficient than others; unlike a concept like jeito, a Brazilian term for something like a ‘knack,’ ‘talent’ is usually quite task specific or specialized, even though a ‘talented’ person is often quite versatile.

    ‘Talent’ is typically contrasted with ‘hard work’ or ‘determination,’ suggesting skill is some mix of natural ‘talent’ and ‘hard work,’ in various proportions. The cultural concept of ‘talent’ is a bit unstable; no one would expect a talented musician to simply pick up an instrument and play. Rather ‘talent’ is usually an idea that some people learn quicker, more effortlessly, and with greater effect. In some ways, ‘talent’ can be like a multiplier, allowing a person to get more out of formative experiences and instruction.

    At times, ‘talented’ seems to mean little different from skilful, but ‘talent’ also has a bit of an edge: it can be an evaluation tinged with disappointment, ’squandered talent,’ a suggestion that a person has potential which may not have been fully developed because of other failures, like an absence of hard work or discipline.

  3. Anders Ericsson: Updated Excerpts (2000) on Expert Performance And Deliberate Practice. [You may also be interested in popular science articles covering Ericsson's work at CNN and Scientific American; see also this post for further links.]

  4. Dr. Doyenne at Women In Wetlands: Is Talent Overrated? (and Part 2).

  5. Fabio Rojas at Orgtheory.net: You Need More Than Talent to succeed in academia.

  6. Malcolm Gladwell in New Yorker: The Talent Myth:

  7. Gladwell's book Outliers has several chapters on successful people at the very high end. From what I know, his summaries of academic research are fairly accurate, though I'm not able to evaluate his spin on that research. For a flavor of what he covers in this book, take a look at these reviews.

Lilavati's European Daughters


Women in Science:

For much of human history, women were officially excluded from the scientific realm. However, in spite of their invisibility in the history narrative, this did not mean that science was exclusively a man’s world. Many women, throughout the centuries, have managed to overcome their marginalisation and excel in their chosen field, making vital contributions to the sum of human knowledge.

With this book we would like to celebrate European women scientists throughout the ages. The book tells the compelling stories of some of the heroines of European science – some sung but many unsung – and, through their narratives, it enriches and completes the history of scientific knowledge by highlighting its female face.

The entire book is available as a pdf; the website also has audio-chapters.

Links ...


  1. List of Professor-Approved Holidays at PhD Comics.

  2. Raghu at A Heuristic Viewpoint of Life: Graduate School Mahabharata.

  3. Johann Hari in The Independent: A morally bankrupt dictatorship built by slave labour: "Dubai is finally financially bankrupt – but it has been morally bankrupt all along. The idea that Dubai is an oasis of freedom on the Arabian peninsular is one of the great lies of our time."

  4. Anubhuti Vishnoi in The Indian Express: Brand IIT goes to the world, gets OK to set up a campus in Qatar. From the Wikipedia entry on Qatar, we learn the following:

    Qatar University was founded in 1973. More recently, with the support of the Qatar Foundation, some major American universities have opened branch campuses in Education City, Qatar. These include Carnegie Mellon University, Georgetown University School of Foreign Service, Texas A&M University, Virginia Commonwealth University, Cornell University’s Weill Cornell Medical College and Northwestern University.

  5. Aaron Swartz at Raw Thought: How I Hire Programmers: "There are three questions you have when you’re hiring a programmer (or anyone, for that matter): Are they smart? Can they get stuff done? Can you work with them?"

  6. Geoff Maslen in University World News: AUSTRALIA: Collapse spreads around global village:

    News spreads fast in the global village created by the World Wide Web. And bad news always travels that much more quickly than any other kind - as the Australian government found to its likely cost this month when a Chinese-owned company called the Global Campus Management Group that ran a series of vocational education colleges in Melbourne and Sydney for foreign students suddenly shut its doors and went into voluntary liquidation.

Saturday, November 28, 2009

Annals of MBA Scams


Volume XIV: This University Operates from a Car Shed.

Conference with No Participation


We have already seen one version -- the scamference, which is essentially a fraud.

Thanks to an e-mail alert from a colleague, I became aware of a second kind: Journal Conference.

Journal Conferences are virtual conferences, where blind papers are peer reviewed by at least two/three experts and accepted papers are published in International Journals. No paper presentation is required.

Don't you love that part about "blind" papers?

Going by the names of people in the National Committee, it seems like an Indian operation. Poking around that site, I couldn't find a physical, "real world" address, though.

Anyways, the conferences (in four areas!) "didn't happen" in 2009. I wonder if they will be repeated next year ...

Manhattan Project at Chicago


... [T]his trouble with graphite [that boron impurities in it quenches a chain reaction even before it gets started] is one of the reasons America far outpaced Germany in the race to obtain a sustained nuclear chain reaction. The arrogance of German physicists made them suppose they alone had the intellect and talent required to achieve their goal. They didn’t bother to consult chemists and engineers, the very people who made the American project feasible. Tests by German physicists showed that graphite was unsatisfactory as a moderator in a reactor, but the physicists did not realize that the trouble lay in traces of boron. So they opted for heavy water, a material available only at great expense from one Norwegian hydroelectric plant. Sabotage prevented the Germans from ever attaining enough heavy water for a nuclear reactor. Pure graphite can be made in quantity far more cheaply and easily than heavy water.

That's from Work on the Manhattan Project, Subsequent Events, and Little Known Facts Related to its Use, a short article by Lawrence Bartels, University of Michigan chemist who's "one of the few remaining survivors of the war-time project."

Bad day for the IIMs?


India Today:

The Common Admission Test (CAT) that was supposed to go paperless from today has been rescheduled at some centres as the servers developed some glitches.

Reports are pouring in from several places in the country that because of network problems, the CAT examination has been put off.

At centres in Chennai, Kolkata, Bangalore, Indore, Pune and Bhopal the test has been put off.

Meanwhile, IIM authorities, have denied reports that online exam servers crashed. They have told the HRD Ministry that the exam was not cancelled but has been rescheduled at some centres.

Friday, November 27, 2009

Philosophy Quotes


Over at 3 Quarks Daily, Gerald Dworkin has put together a wonderful collection. Great fun!

Here's a geeky put-down:

He is a quantum philosopher. I can’t understand him and his position at the same time.
-- S, Morgenbesser

Here are a couple of other quotes:

Philosophy is the cure for which there is no adequate disease.
-- Jerry Fodor

Metaphysics is almost always an attempt to prove the incredible by an appeal to the unintelligible.
-- Mencken

Links ...


  1. Suzanne Franks at Thus Spake Zuska: Lives of the Saints of Science: Darwin:

    This past year the scientific community has been engaged in a massive telling and retelling of the story of one of those key figures - Charles Darwin. All year long, I have been reminded of my first encounter with the actual writings of Mr. Darwin, as opposed to the presentation of his myth. It happened in a women's studies class.

  2. Charu Sudan Kasturi in The Telegraph: Foreign University Bill in Deep Freeze:

    A proposed law to allow the entry and regulation of foreign universities will now be reviewed afresh under a panel of top bureaucrats, threatening to delay indefinitely a legislation scheduled for cabinet approval. [...]

    The decision to refer the bill to a committee of secretaries effectively means the proposed legislation, which may now undergo fresh changes, is unlikely to see the light of day for some time.

    Typically, committees of secretaries take several months to review a proposal or draft legislation before finalising their reports. But they have no time restrictions within which they are required to complete the review.

  3. Thomas Frank in WSJ: A Liberal Thanksgiving: "We hear much less nonsense about the wisdom of markets these days."

    ... Just about the only ones who still believe in omniscient markets anymore are the think-tankers who are paid to believe in it.

Thursday, November 26, 2009

Links ...


  1. Janet Stemwedel at Adventures in Science and Ethics: My work has been plagiarized. Now what?. She offers some advice; and so do her readers.

  2. Spiegel interview with Umberto Eco: 'We Like Lists Because We Don't Want to Die'

    ...I realized ... that the [Louvre] exhibition [that I would curate] would focus on lists. Why am I so interested in the subject? I can't really say. I like lists for the same reason other people like football or pedophilia. People have their preferences.

    SPIEGEL: Still, you are famous for being able to explain your passions …

    Eco: … but not by talking about myself. Look, ever since the days of Aristotle, we have been trying to define things based on their essence. The definition of man? An animal that acts in a deliberate way. Now, it took naturalists 80 years to come up with a definition of a platypus. They found it endlessly difficult to describe the essence of this animal. It lives underwater and on land; it lays eggs, and yet it's a mammal. So what did that definition look like? It was a list, a list of characteristics.

  3. Larry Hardesty in MIT News: Explained: The Discrete Fourier Transform.

Chrome OS


Just a couple of links telling us this is a very big deal -- even though the first machines won't be available for another year or more.

  1. John Stokes in Ars Technica: Chrome OS: Internet failing at PC > PC failing at Internet:

    In 2009, it's better to be an Internet company that's taking slow, awkward first steps toward the PC, than a PC company that's still trying and failing to truly integrate with the Internet. Ars looks at what Chrome OS means for Google, Apple, Microsoft, the netbook, ARM, Intel, and the cloud. "Revolutionary" is a clichéd term, but Chrome OS is a good candidate for it.

  2. Robert Cringely at I, Cringely: Chrome and Chrome, What is Chrome?

    While we’re talking about operating systems here, Google’s real target is Microsoft Office. Redmond makes money from Windows but makes a lot more money from Office, its productivity app monopoly. Google already has its Google Apps pitted against Office, but Brin and Page know they won’t crack Office’s hold on corporate America without addressing the Windows flaws that effectively underlie both Office and Google Apps in their current incarnations. That’s where the Chrome OS comes in.

Tuesday, November 24, 2009

Autonomous Colleges


I want to return to the column by Tapan Raychaudhury who doesn't like the idea of converting some of the academically better equipped (and more accomplished) colleges into universities:

My particular concern here is with the new initiative to confer the status of universities on selected colleges. One assumption behind it seems to be that colleges that, perhaps after a glorious past, are now suffering in quality will regain their old excellence if turned into universities. The logic underlying this assumption is incredibly bizarre. Spelt out, it would imply that institutions which are mediocre or worse today will become centres of excellence tomorrow by virtue of having university status conferred on them. It is well to remember that in the golden tomorrow, the people running these institutions will continue to do so still. If they are sought to be replaced by allegedly abler people, the seat of learning will be converted into a battleground for power. If, on the other hand, the old guard are allowed to remain in power they will ensure that the newcomers do not excel in any way. Such, indeed, is the way of all flesh as is well-known to all but the most doggedly optimistic among us.

On the other hand, the logic behind conferring university status on a particular college may well be a recognition of its excellence, and making that excellence available for the service to a higher level of learning. If this is so, I suggest some very simple tests to ensure the validity of the judgment. First, since we are, these days, so enamoured of American academic practices, let us take anonymously the opinion of students about the quality of teaching and make a high mark a sine qua non of the relevant decision. Secondly, since these institutions will be expected to contribute to knowledge, let us have surveys of the amount of quality research they have produced in the last ten years — in terms of scholarly books (reviewed in authoritative journals), refereed articles and theses done under their supervision. Thirdly, a quiet survey of library books issued to students and teachers in an average year. Of course both may have borrowed or bought books to supplement what is available in their college libraries and an enquiry into this aspect of the pursuit of knowledge would be indeed worthwhile.

Clearly, Raychaudhury is pretty negative about converting colleges into universities. But I want to shift the focus to a related system: autonomous colleges.

In our hub-and-spoke system of higher education, academically better-positioned colleges could be given an "autonomous status" by their university (the hub). This system has been in place for at least three decades -- I still remember colleges like Loyola College and Madras Christian College flaunting their autonomous status in the 1980s. And this system appears -- going by this list -- implemented vigorously by the universities in Tamil Nadu.

As I recall, this autonomous college issue was not particularly controversial -- people just assumed that the better colleges would eventually get the autonomous status, and many did.

For all practical purposes, the autonomous college is a university -- it designs and implements its own curriculum and grading schemes, with the parent university's role being limited (largely) to issuing degree certificates. At least, that's the theory.

There's much going for this theory. As Pratap Bhanu Mehta says in a recent op-ed on the reforms at the Delhi University,

Ideally, a semester system allows you to achieve the following objectives. It can facilitate the creation of a credit system, and hence allow more choice and flexibility. In institutions where the semester system has real pedagogical bite, it is premised upon one important fact: that the teachers teaching particular classes evaluate their own students. [...]

A semester system works well when each individual faculty member has substantial freedom to innovate in course offering at his or her level. This is possible only where there is no disjunction between those who set the syllabus, those who teach and those who evaluate. The crisis of undergraduate education has its source, in part, in this disjunction.

The academic autonomy enjoyed by these elite colleges has all the ingredients identified and recommended by Mehta. And this system has been around for over 30 years now. Has there been a review of this system? Is it seen as a success?

A pessimistic take on higher education in India


Read this column by Tapan Raychaudhuri, former professor of modern Indian history at the University of Oxford, in The Telegraph:

... [T]he education of our MAs and honours graduates, except in the case of a small percentage of them belonging to some elite institutions, consists in memorizing lecture notes. The quality of the said notes determines the quality of our higher education. The truth or otherwise of this statement can be very easily tested by using the method of sample survey.

Assuming my hypothesis to be true, and I should be very happy if it turns out to be false, what exactly do we gain by multiplying further the number of universities at a very heavy cost to the nation? If, as I suggest, our institutions are spreading mainly non-knowledge (for how else would we describe education based almost exclusively on lecture notes?), is it really worthwhile to increase their number? If we want more people with degrees that are worth very little in terms of the knowledge acquired, this target could be more inexpensively attained through open universities and correspondence courses ...