Tuesday, May 26, 2015

Computer Science, Women, NYTimes!


NYTimes has yet another story on how some college or the other is doing such a great job of attracting, and retaining, women in its computer science program. If it is the University of Washington this time, it was Harvey Mudd last year, and Carnegie Mellon back in 2007. These are the ones I have read; there may be others that I didn't even know about. [Update: And, oh, there's also this from 2011, though it is not quite about women in computer science].

It would have been interesting if the later stories showed some awareness of the earlier ones -- for example, if Carnegie Mellon did some great things get a lot more women into its computer science program back in 2007, how well is it doing now? Has it improved enrollment figures for women even further? Has it hit a wall? Has it let things slide?

But, but ... I'm just quibbling here. The most recent intervention by NYTimes in the Women-in-Computer-Science debate is quite good in its coverage of the kinds of experiments at different places (UWashington, Michigan State, Harvey Mudd, Harvard, ...), as well as of the kinds of curriculum-related debates within Computer Science.

Doing Science is, in fact, "Rocket Science"


Leonard Mlodinov has an op-ed in NYTimes arguing against the myth that science is just a series of flashes of intuitive insights that just hit people by accident.

Two thousand years ago, Aristotle’s “Physics” was a wide-ranging set of theories that were easy to state and understand. But his ideas were almost completely wrong. Newton’s “Principia” ushered in the age of modern science, but remains one of the most impenetrable books ever written. There is a reason: The truths of nature are subtle, and require deep and careful thought.

Over the past few centuries we have invested that level of thought, and so while in the 19th century the Reuters news service used carrier pigeons to fly stock prices between cities, today we have the Internet.

Even if we are not scientists, every day we are challenged to make judgments and decisions about technical matters like vaccinations, financial investments, diet supplements and, of course, global warming. If our discourse on such topics is to be intelligent and productive, we need to dip below the surface and grapple with the complex underlying issues. The myths can seduce one into believing there is an easier path, one that doesn’t require such hard work.

But even beyond issues of science, there is a broader lesson ... We all run into difficult problems in life, and we will be happier and more successful if we appreciate that the answers often aren’t quick, or easy.

Saturday, May 23, 2015

Bad Incentives are behind Big Science Frauds


Updated with links.

* * *

Adam Marcus and Ivan Oransky, the folks behind Retraction Watch have penned an op-ed in NYTimes: What’s Behind Big Science Frauds? They flag a circle of behaviours to support their thesis that "the incentives to publish today are corrupting the scientific literature"

  • Science fetishizes the published paper as the ultimate marker of individual productivity. And it doubles down on that bias with a concept called “impact factor” — how likely the studies in a given journal are to be referenced by subsequent articles.

  • Journals with higher impact factors retract papers more often than those with lower impact factors.

  • Scientists view high-profile journals as the pinnacle of success — and they’ll cut corners, or worse, for a shot at glory.

  • [The reviewers at top journals] seem to keep missing critical flaws that readers pick up days or even hours after publication.

  • [P]erhaps journals rush peer reviewers so that authors will want to publish their supposedly groundbreaking work with them.

The news that appears to have triggered this op-ed is a recent retraction of a high profile paper. Since the original paper was on a topic of wide interest, it got picked up by many news outlets. Now that many problems with the data presented in the paper have been uncovered, the senior author (Prof. Donald Green, at Columbia) has sought a retraction, the journal (Science) has responded with an expression of concern, and the junior author (Michael LaCour, a grad student at UCLA) says he stands by his study and promises a comphrehensive response by the 29th of May.

Read the whole thing at Retraction Watch: Author retracts study of changing minds on same-sex marriage after colleague admits data were faked.

Here's a round-up of how various news outlets which covered the original paper have responded to the retraction.

* * *

Update (26 May 2015): An Interview With Donald Green, the Co-Author of the Faked Gay-Marriage Study by Jesse Singal in New York Magazine.

How the Gay Rights Canvassing Study Fell Apart by Naomi Shavin in The New Republic.

Anil Kakodkar speaks out


In the NDTV interview by Shekhar Gupta, Dr. Kakodkar opens up on recent controversies surrounding selection of IIT directors, as well as the government's disgraceful treatment of Prof. Shevgaonkar, director of IIT-Delhi. Kakodkar chooses his words carefully (at one point he says he "stays within limits"), but the subtext is clear.

Here's the video -- the IIT-related discussion starts after 10 minutes or so. The interview is set to continue next week as well.

Saturday, May 16, 2015

The Ultimate Kill Bill - Volume II


Watch this interview of HRD Minister Smriti Irani by Rajdeep Sardesai:

* * *
[Self-plagiarism alert!]

In the (unlikely) event that you haven't seen the two Kill Bill volumes (errr, movies), here's the iconic scene from the first volume.

The Ultimate Kill Bill - Volume 1


Watch this interview of HRD Minister Smriti Irani by Arnab Goswami:

In the (unlikely) event that you haven't seen the two Kill Bill volumes (errr, movies), here's the iconic scene from the first volume.