If you’re a blogger and you venture into government whether in the bureaucracy proper or as a ‘staffer’ you’ve got a problem.  You can’t keep expressing yourself as candidly as you might wish for fear of breaching the relevant public service code of conduct, of having some perfectly sensible observation you make beaten up by the media or just getting others’ noses out of joint.  So often people down tools.  As Tim Watts did when he ceased writing his excellent blog Tree of Knowledge upon becoming a staffer first for Stephen Conroy and then for Sir John Brumby.  (OK John Brumby doesn’t have a knighthood, it just sounded good and I wanted to make sure you were paying attention).

Tim, being a man of substance and an irrepressible one to boot has resurfaced in a manner that finesses all these dilemmas while keeping his blogging instinct finely honed. He’s heading through his entire bookshelf, and reviewing all the books in it. Go and blog Tim’s bookshelf with him.

As I promised a few weeks ago, I have a brief report outlining the results from the survey I conducted comparing two different mortgage calculation procedures. For the truly curious, here it is.

Well bargain hunters fresh from your kills at Borders (they don’t stand a chance when you’ve got those Troppo coupons in your hand) have we got a deal for you?  The entire autobiography of Ben Franklin read by Ben himself. OK, well I lied about that last bit, it’s really Greg Houser but he sounds awfully like Ben must have sounded in his day. Anyway, it’s free!  Yes, that’s right free but there’s more!  You get a new exercise regime to work out your abs while you listen to Ben.  Actually that’s not true.  You don’t get one of those and it’s not Ben (as I’ve already conceded).  Anyway, it’s not an amateur with a squeaky voice and a lithp from Libra Vox, but the full quid from Learn out Loud.

The Harvard Open-Access Policies

The goal of university research is the creation, dissemination, and preservation of knowledge. We collectively take this to be a good. It is an essential part of our duties as faculty members to distribute the fruits of our scholarship as widely as possible.

A committee established by Provost Steven Hyman has proposed a set of measures that Harvard can undertake to promote open access and move toward a more sustainable publishing system that is at the same time more in keeping with the goals of the University. One of these measures is for faculty to grant the University permission to make scholarly articles openly available. On February 12, 2008, the Faculty of Arts and Sciences approved the first such policy, with several other schools following.

The policy works by setting up an automatic license to Harvard for scholarly articles authored by faculty members. This license is nonexclusive and enables open-access distribution, so long as the articles are not sold for a profit. Copyright remains with the author (until such time as the author may assign all or part to another entity). Harvard will use this license to enable it to distribute articles in an open-access repository whose contents will be searchable and available to other services such as web harvesters, Google Scholar, and the like. Contents of the repository will be maintained, archived, and preserved.

The license is transferable, so that the University can allow authors legally to distribute the articles on their own web sites if desired, and can allow educators here and elsewhere to provide the articles to students, for example, in course packs (not sold for a profit).

The policy is intended to serve the faculty’s interests by allowing articles to receive open distribution, simplifying authors’ retention of distribution rights, aiding preservation, and providing unified action to discourage publishers from rejecting articles because they will be available in open access. However, there may be individual cases in which the license works against the interest of a faculty member. In keeping with the spirit of the effort, the policy allows for waiver of the license or delay of distribution in such circumstances.

Additional resources are available at the links to the left, including text of the policies, further explanatory material, information about procedures, and a list of other resources.

Truly ruly.

Exciting stuff!

Infrastructure For A Learning Health Care System: CaBIG

In his proposal for a new cancer care policy in a data-rich future (Jan/Feb 09), Lynn Etheredge correctly notes that the National Cancer Institute (NCI) has built the requisite infrastructure for a learning health care system. Currently, the Cancer Biomedica lInformatics Grid (CaBIG) is connecting in a national networknot only sixty-plus NCI-designated Cancer Centers, but also NCI community cancer centers, where 85 percent of all cancer patients are treated. Moreover, CaBIG enables the seamless continuum that is at the heart of a learning health care system in which knowledge ofaggregated clinical outcomes drives next-generation research discoveries, which in turn are validated at the bedside for improved clinical outcomes, in a seamless “virtuous cycle.” In addition, CaBIG and the Food and Drug Administration (FDA) have collaborated for electronic coordination of clinical research information. CaBIG has also demonstrated its capacity to exchange information with the emerging Nationwide Health Information Network.

While these efforts to date have been focused on cancer, the NCI is in effect prototyping a twenty-first-century knowledge-based biomedical system. The “BIG” (Biomedical Informatics Grid) inCaBIG serves as a nationwide, interoperable, interconnected information technology platform that enables information sharing. A health care ecosystem then formsfor the first time everthat electronically links academic centers, caredelivery organizations, insurers, diagnostic and pharmaceutical product innovators, government research and care institutions, and all other players in the biomedical enterprise. As data are shared among those previously siloed entities, reunification occurs between the currently divided worlds of clinical care and research. Specifically, the availability of clinical encounter and molecular characterization information permits prequalification of participants and rapid assembly of study populations; research can be conducted without reestablishing duplicative tools andinfrastructure; redundancy of research activities is eliminated; real-time monitoring of safety occurs; and the development of new therapeutic inventions can be conducted faster and less expensively with near-term benefit to patients.

This prototype is already under way. We invite all sectors to participate (http://www.bighealthconsortium.org) to carry Etheredges bold concept even farther.

Take a look at the job advertisement below the fold. The pay is good, but not great by UK standards (though I guess you couldn’t complain at the top of the scale). They do seem to have a rather comprehensive set of requirements for the right applicant. Anyway if you were thinking this sounds like you, you’re a tad late, applications closed in March.

(Continued)

If you click through to the source, you’ll find twelve of these segments from a ‘master class’ of 2001.

And I’d never heard the song featured in this final segment.

Who doesn’t like awards?  When Alexander first went to school becoming Cool Kid of the Week was pretty much the major priority. After having earned the award a few times, resentment set in when Alex realised that the award seemed pretty randomly passed around and that in fact if you were naughty, that seemed to be a pretty good strategy for Cooldom, as  increasingly frustrated teachers tried with increasingly fractious kids to see if a bit of Cooldom would do the trick. Meanwhile one of the early pioneers of behavioural economics, Bruno Frey is trying to quantify how valuable awards are.  And this picture of the performance of call centre workers before and after either receiving or not receiving an award tells a thousand words.  In the words of the article (pdf)

Result 1. Awards increase the performance of recipients as compared to nonrecipients subsequent to winning.

Result 2. Receiving an award improves the performance of winners, whereas the performance of nonrecipients remains unaected.

But . . . the plot thickens. It turns out that the teachers’ strategy looks pretty on the money.

Another explanation for the observed increase in performance may be the increased visibility of the award winner in the month following the award. Recipients may feel a need to live up to the honor of having received an award for their voluntary work behaviors, and this may aect their core performances. This eect should be stronger for award winners whose core performance was below average prior to the award. The data allow us to test this hypothesis by separating the winners into two groups: those individuals who performed below average and those who performed above average.

Looking at how much performance increases between the month of the award and one month later, we nd that, on average, the rating of low performers increases by 0.58, whereas the performance of high performers decreases by 0.17. The one-sample t-test indicates that both coecients are highly signicantly dierent from zero. This differential impact of winning an award supports the notion that the increase in performance is caused by social pressure or the winners wanting to live up to the award with respect to core performance.