The University of Everywhere. Audrey Watters takes a scalpel blade to The End of College. I am not entirely convinced by the ‘more research’ mantra. Much professional education research is RCT: randomised, confounded and trivial. Improving education is about design, with large search spaces, where judgement, integrity and a genuine attempt to makes things better counts for nearly all.

The Economist has a feature on Universities (as in, outsource, disrupt, maximise shareholder value, and create more state welfare for corporations to enclose the commons….). Must have been bad news week, as they have little new to say. It is not penned by Pearson (their owner), but parts of it read as though it were. Nor do they a good job of scraping below the surface as to why university fees have gone up so much, nor why teaching is undervalued. As usual, part of the subtext is that taxpayers need to subsidise the corporate sector ever more, because firms cannot be bothered to pay the cost of finding nor training their employees. I am not being too generous: there are many truths here. And plenty of blame to go around.

The tragedy of the Germanwings flight. People often compare doctors and airline pilots, and safety in medicine with safety in the aviation industry (usually unfavourably with regard to medicine). There are some interesting posts here, here and here. No simple message, except that the systems you place professionals in may have big effects. And these systems themselves are increasingly designed by people with little first-hand knowledge of the tasks involved. It is why some NHS hospital chief executives (and medical directors, too) often avoid walking around hospitals.

I am not being very generous today, it seems.

 

Obituary here

This was something I posted elsewhere awhile back, but if anything it seems even more appropriate now. From a  Nature editorial. [see also here]

Perhaps the most extraordinary thing about The Feynman Lectures on Physics, the book in question, is that it was nearly strangled at birth. Robert Leighton, chair of a committee tasked with spicing up the physics teaching at the California Institute of Technology in Pasadena in the early 1960s, did not think that Richard Feynman was the right man for the job. “That’s not a good idea,” was his original response. “Feynman has never taught an undergraduate course. He wouldn’t know how to speak to freshmen, or what they could learn.” (At around the same time, incidentally, an official at Decca Records decided that “The Beatles have no future in show business”.)

And what is less well known:

The popularity of the lectures and the enduring appeal of the books that grew from them are often attributed to the individual and spontaneous genius of Feynman. But they were painstakingly prepared and practised, and had generous financial backing. (The lectures were part of broader changes to the teaching at Caltech’s physics department funded with some US$1 million from the Ford Foundation.)
This is a lesson that university officials would do well to remember as funding is cut and pressure placed on faculty members to cram more into their timetables. Those who can, teach, but they need support.

BTW,  Jim Watson’s books (which I have studied) also had a major educational impact. I still think this speaks to the critical importance of domain expertise, and having teachers with imagination, and time to imagine. Domain expertise is necessary, but not sufficient. Which seems a nice way of reminding you of George Steiner’s words in ‘Lessons of the Masters’.

“Bad teaching is, almost literally, murderous and, metaphorically, a sin.  It diminishes the student, it reduces to gray inanity the subject being presented.  It drips into the child’s or the adult’s sensibility that most corrosive of acids, boredom, the marsh gas of ennui. Millions have had mathematics, poetry, logical thinking, killed for them by dead teaching, by the perhaps subconsciously vengeful mediocrity of frustrated pedagogues.

 

“After earning his medical degree in 1951 he trained in hospitals in Montreal. “To my surprise I also found I enjoyed clinical medicine,” he wrote in his Nobel prize biography. Then he quipped, “It took three years of hospital training after graduation, a year of internship and two of residency in neurology, before that interest finally wore off.”” From an obit of David Hubel. [direct link to this quote]

From an article in press, based on a talk I gave last year, in which I mull over some of the problems of modern medicine: a dysfunctional research system; a failure to consider what types of rationality underpin medicine; and a broadside on medical education .[Yes, modesty doesn’t figure]. Here is a flavour:

There is a phrase used by engineers, that goes roughly as follows: physics was made by God, but to engineer is human. Engineers have of course to be cognisant of the laws of nature, but that is only part of what they need to know in order to build artefacts. This applies to medicine too. Science may tell us about how the natural world works, but we want more than that, because we also have to know how science and society work. We want to intervene, and the test of our knowledge is how well we can do this in the social and economic systems we inhabit. UVR causes most skin cancer, but if the patient in front of us already has a thick nodular melanoma, we have to ask: was it inevitable that they would present when they did, and what can we do for them now? But the answers to these questions do not just depend on biology. Medical engineers not only have to talk the language of molecules, but also the language of health insurers. Furthermore, medical engineers also have to know how to engineer their own cultures of discovery, along the way.

Children are not to talk about the exam. Pearson again, and the business of testing for profit. I think there is moral hazard in spades here. As one commenter says: Whatever happened to Freedom of Speech? Oh, wait, it was sold to Pearson. I seem to remember being told that some publishers paid MRCP candidates if they could remember exam questions, so that they could be included in books they published. If you use MCQs, they will always leak. You need tens of thousands, and then you can flood the world with them. If you know the answer to say 10,000 dermatology undergraduate MCQs, you know dermatology. (Apart from that which you can’t test in an MCQ format.)

A sensible letter on MCQs in medicine (and other subjects). It is not what we do. Am I missing something?

Goodbye, SAT: How online courses will change college admissions. Not certain, but ideas welcome.

I suffer from the disease of being addicted to discussions about statistical inference. Probably because it is a part of the problem of why so much medical science is unreliable, and does not provide a good guide to action. But, it consumes so much CPU time that I need to get therapy. Great example here. After the first life time it may become second nature. Otherwise, by advice to the young is: start young, and clear your desk. Reading Stephen Senn is not compatible with a family life. I wish I had stopped with Gigerenzer and Royall. Having said that, as far as the clinic goes, I am all for the Adaptive Toolbox, which is how we practice, rather than profess, I think. Remember each day: I must wean myself off…..

Career advice: If something that seems like work to other people doesn’t seem like work to you, that’s something you’re well suited for. Paul Graham

Well yes, I am still moaning about having to spend all that time on a compulsory and badly designed health and safety e-learning package (why e-learning, anyway?). Well, here is nice quote to cheer me up:

From a review by Gillian Tett of the ‘Utopia of Rules’  by David Graeber.

Americans do not like to think of themselves as a nation of bureaucrats — quite the opposite actually,” Graeber observes. “[But] the final victory over the Soviet Union did not really lead to the domination of the ‘market’ [but] simply cemented the dominance of fundamentally conservative managerial elites . . . no population in the history of the world has spent nearly so much time engaged in paperwork.”

And I have (literally) just had to spend almost 30 minutes filling in a terminally tedious Health and Safety online form about how to position my chair. This has less to do with staff health (let alone sanity) but rather more with protecting the institution from perceived legal challenge. Of course, what is most annoying is how badly designed the elearning module is. It is not meant to inform, but to punish.

But even if you disagree with his politics, Graeber’s book should offer a challenge to us all. Should we just accept this bureaucracy as inevitable? Or is there a way to get rid of all those hours spent listening to bad call-centre music? Do policemen, academics, teachers and doctors really need to spend half their time filling in forms? Or can we imagine another world?

There are no easy answers. But the next time you see a bureaucratic form — and I have several sitting in my inbox right now — it is worth asking who really benefits from it? And, more importantly, who would suffer if we were to all suddenly rip them up? It is, perhaps, one of the more subtly revolutionary ideas of our age. 

 

Abelson: So the department went through, “Let’s invent the core curriculum.” It might have been ’78 or ’79. Then Gerry Sussman and I got the ability to do the new course, which was going to be called 6.001. We basically did it without telling people what we were doing, which is why they let us.

Seibel: What do you think they would have objected to?

Abelson: Oh, people object to whatever you do.

One of the reasons you need skunkworks. An interview with the legendary Hal Abelson, in Code Quarterly, The Hackademic journal