dimanche 22 octobre 2017

Transparency real and imagined

from the all very well in theory department

Back in August I came across this blogpost—a somewhat generic homily on the importance of openness and transparency in scientific communication. I share the author's optimism that open access to the scientific literature will provide better validation of published work than traditional systems of peer review. But unless this superior access is actually exploited by knowledgeable users, the potential gains may not be realized. Though a blinking cursor beckoned me below the line, the author has, in his or her wisdom, failed to publish my comment. So it must appear here:
Transparency is superior to trust—as long as some relevant person(s) actually exploit(s) the transparency. Look at how long that ssl flaw hung about in Debian, for example: https://pinboard.in/u:juliusbeezer/t:security/t:opensource/
That was all open code, utterly vital to the security of hordes of crucial servers run by the world's top-most geeks, and therefore, every internet user. But the problem sat there for two years, apparently.
That's an extreme example that did get fixed. Transparency is necessary yes, but unless it's actually backed by readers/critics/reviewers/coders/experts actually looking through the windowpane afforded by it, its value is only rhetorical.
It does mean that the guards can guard the guards and we can watch the guards guarding the guards though. Or maybe McGregor-Maywether.

jeudi 22 juin 2017

Another look at Dutch geography as it pertains to cyclists

from the polder layout department

The Netherlands are often held up as a model for good practice in cycle policy. I've made the point already that the country has a number of geographical peculiarities which mean that its methods should be interpreted with caution when considering generalizing them elsewhere. Of course it is possible, indeed more enjoyable, to ride a bicycle in more hilly terrain than any the Netherlands has to offer. But the techniques required by both the cyclist, and the urbanist seeking to encourage them, are self-evidently going to be different. This much is obvious.
But there are a couple of more subtle points to l'exception néerlandaise that have occurred to me since my pilgrimage to Groningen last year:

a map of 20th century Dutch polders
20th century polders, from Hoeksema (2007)
1) Fully 5% of the Netherlands' land surface area has been added to the country between 1930 and 1968. This often seems to have been landscaped à l'américaine, with spacious boulevards that admit complete modal segregation between facing frontages as much as 100m apart. Time pressures obliged me to admire the splendour of the Almere suburb from the window of a train rather than from the bike, but the desire to allocate huge reserves of space to the motorist there was plain enough, as one would expect from any bunch of 1970s planners anywhere.

2) In a country where for centuries the canals have been the main axis of transport between towns, it may be that it seems quite natural to have a "major" and "minor" towpath. I noticed on my trip last May that very often I received strong cues from the urbanists to travel on one side of a canal-line (which was often a mixed-use low-speed residential street), while across the water I could see cars moving quite fast on a higher speed street. I was cool with that as long as getting to the "right" side didn't take me too far out of my way. As a tourist, naturally I just went along with it, but I could imagine being a local resident with a bike and being daily pissed off by the tortuousness of the route I was obliged to follow. (Obligatory) cycle routes along major trunk roads often jink from one side of the motorway to other, adding significant travel distance perpendicular to the desired direction. This sucks.

We should be clear about all these distinctions when considering the applicability of Dutch cycle infrastructure design to other parts of Europe, where the width of roads and streets (their frontage-to-frontage distance) was determined in the era of horse-drawn vehicles. Horses go home at night, and the carts they were pulling too. So three cart-widths would be pretty much enough anywhere: one delivering, two passing in either direction. On-street parking is a twentieth century curse. Where such conditions exist historically in the Netherlands, they tend to mostly exclude motor traffic (as of course you should). Parallel universes of "protected cycleways" may be an option in new towns and suburbs, but in the vast majority of European cities the nettle of private car dependence must be grasped.

lundi 19 juin 2017

Is there "a Macron effect"?

from the department of curious anthropology

My oldest buddy Dan Falchikov, sometime parliamentary assistant to Archy Kirkwood, and still, I believe, a partisan for the Liberal Democrats, asked me on Facebook what I made of "the Macron effect." I was happy to give him my view, and perhaps my remarks may be of wider interest. So here [lightly edited] they are:
[Hi Dougie, just wondering whether you have a take on the Macron effect? Obviously being up against a fascist meant he was always going to win the presidency but I'm more interested in how he emerged from nowhere to beat the more established candidates. Is there any decent analysis you're aware of?]
He [Macron] was pretty high profile in his role as economic minister in the Hollande administration, then resigned early enough to escape (somehow) the backwash from them being found red Tories in practice. Being President is his first elected position! A first round vote for Macron was basically a vote for the status quo, in terms of the relationship between business and the French social contract; Fillon's promise of deep cuts in public service employment was deeply unpopular, and his allegèd filching from the public pocket also played very badly. Mélenchon played a blinder, and came closer than any true left candidate since the 1930s; Hamon was a lowish-profile Hollande minister, and one of the PS rebels in the last parliament, which does not play well in French culture; so triple whammy for him (6%!). Now that my French is good, I can understand what Marine Le Pen is saying, and apart from her obviously distasteful world view, she is frankly someone who does not have the mental capacity to be a serious politician. The FN vote reflects, in my view, three tendencies: 1) people who are racist ultra-rightists (few, though there is plenty of casual racism in France); 2) people who want to smash the whole Paris political class (cf Brexit); 3) a survey result for the number of people in France who are… how can I be polite? … incapable of analysing competing paradigms.
Macron is a weak king: with a first round vote of only 24%, and a second round win against the FN (with near record abstention and 4.4 million spoiled ballots) he has little mandate for his promise to "govern by decree" if he doesn't get his majority in the the Assemblée (though it looks like he will). Personally I find this promise an objectionable rejection of the constitution, and believe a parliamentary committee should generally outperform a presidential decree. But hey! The fun will really start when he tries to pass his new employment laws, or if he tries to build the proposed airport at Notre Dame des Landes. Macron is detested by large numbers on the left, there will be some very heavy demos, and I will be dodging clouds of teargas here in Nantes. Ho hum!
As for your specific point, his rise is unprecedented, (though he has all the qualifications, did well at ENA etc) but he succeeded in rallying large numbers of young middle class people to get the vote out. There is no doubt he is highly intelligent and capable (and his English is remarkably good for a French person); also no doubt he is firmly in the pocket of the big business/lobbyist complex that really runs France. I guess I wish him fair wind, but the moment he does something really stupid my good will (and many others') will rapidly evaporate. I predict an ignominious decline in his popularity over the next 5 years (though I'd be happy to be proved wrong). Thanks for asking! 🙂
[Was there a particular pitch to the young middle classes, or was it more of a shared feeling/shared values?]
That is a good question, and I'm not sure I have a good answer. My impression is that enough of the French public are suckers for someone who is handsome and clever on the telly. There's no doubt there is frustration with the rigidity of French employment practices, and a feeling that France is getting left behind other countries that manage greater flexibility in the labour market. The difficulty is that in France *everything* hangs on your "statut social", which basically comes down to your employment status. (cf Americans and their pre-Obama health insurance). So any changes to the system provoke enormous reaction. The last round of attempts at reform (la loi El-Khomri, which was basically authored by Macron) caused several riots in the streets here, one of which I unexpectedly attended when I turned a corner on my bike on my way to pick up the boy from music! Eek! The political discussion then becomes what to do about the riots, not a reasonable discussion of the strengths and weaknesses of what's being proposed. (I was about 75% against/25% for El-Khomri myself: I don't see any sense in increasing retirement age when there is so much youth unemployment, for example, but I would be in favour of more targeted occupational health examinations). My impression is that people do follow the big issues along better than the UK public does, but there is also an important tribal element, and Macron benefited from young people sick of tribalism, and feeling stuck in the system. The system is very difficult to reform however, and when push comes to shove I'm not sure that Macron's vote will translate into people in the street backing him. But if he governs by decree and the riot police repress the demos against him, then there will be an awful lot of people on the street against him. Certainly here in Nantes, and probably Paris too.
Update 12h02, 19/06/2017. It looks as though M. Macron and La République En Marche have obtained a legislative majority. May the new Assemblée enjoy peace and good fortune in its deliberation of the public interest!

vendredi 14 avril 2017

Air pollution sceptic refuses comment

from the start-with-cholera department

Euan Mearns is an honorary research fellow at Aberdeen University and a former oil industry consultant. His admirably active blog exerts a pitiless scrutiny on the renewable energy business, and I like to read it as part of a conscious effort of escape from my bubble.
He recently posted a long article which attempted to unpick a claim made by Channel 4 News that air pollution from diesel engines causes 40,000 deaths annually, following the publication of a Royal College of Physicians report. As Mearns points out, diesel engines are not the only source of air pollution, and so the headline figure is certainly open to scrutiny and debate. But this does not really entitle him (or the commenters he has attracted) to accuse the RCP of publishing "fake science" based on "green thinking." As is my wont, I composed a comment for his delectation, which he has chosen not to add to the other 69 he has published. So it is here instead:
Yep, the statistics of the effects on air pollution aren't easy to interpret. If you want to pursue this line a masters degree in epidemiology would provide a good foundation for your further studies e.g. http://www.lshtm.ac.uk/study/masters/mse.html#first

Look, it's quite clear fine particulates and gaseous oxides aren't good for your health. The evidence that I find most convincing is that all-cause mortality rises in pollution peaks: e.g. https://academic.oup.com/bmb/article/68/1/95/421216/Air-pollution-and-infection-in-respiratory-illness Note that the people who die acutely are those with the least cardiorespiratory reserve due to pre-existing illness, but the fact that this increase in mortality is observable suggests that a harmful effect is occurring in everyone's cardiorespiratory systems, and thus, long-term, risks diminishing everyone's cardiorespiratory reserve. Note also that the mechanism of fatal illness is not necessarily directly linked to the peak itself (e.g. someone with asthma suffering an acute attack and dying of respiratory failure immediately), but in increased rates of respiratory tract infection for several months after the peak.

Even if the epidemiologists' best estimates were out ten-fold either way (say 4000 dead/year—400,000 dead/year) either figure would still constitute ample evidence for tightening the regulatory screw on those who indulge in the antisocial practice of burning hydrocarbons in the atmosphere in the pursuit of an inefficient method of personal transportation.

So the issue must be addressed. The choice is not between petrol or diesel. The choice is between the private motor car and the other less polluting modes of transport. Although there is some justice in that motorists themselves receive the highest doses of air pollution (e.g. http://cycling.today/cyclists-exposed-to-five-times-less-air-pollution-than-those-in-cars/), everyone is affected. You're no longer allowed to blow smoke in my face in a bar—why should you be allowed to blow smoke in my face at the traffic lights?
 M. Mearns himself runs a 1.6L diesel-engined Volvo, and I suspect has classified me a "green troll." Ho hum!

Update 0820h, 16/4/2017. Following reorganisation, all participants in the interdisciplinary "Engineer in the clinic!" programme are being encouraged to attend Epidemiology 101.

mardi 21 février 2017

The unbearable asymmetry of bullshit

 from the department of anti-link-rot-action 

Somewhat at random it has been my custom to follow the blog of the Oxford Uehiro Centre for Practical Ethics in my RSS reader, and occasionally I would even feel moved by the spirit of enlightenment to offer up a comment to the august minds philosophizing there. Sadly the blog now seems to be in a state of permanent technical failure, effectively removing these comments from public view. My cacheing system has worked nicely though, so I guess it's up to me to take over the hosting from here on. I was particularly pleased with this comment, which pulls together my thinking on attention, agnotology, peer review and bullshit, written in response to this article by Brian Earp:

The key problem with bullshit that enjoys a certain "truthiness" is attentional: its presence is a time-wasting distraction from the good stuff. The agnotological processes whereby e.g. the tobacco companies, or climate change denialists, skew "debates" by creating false controversy is well-studied. In theory scholarly publishing should be relatively resistant to this process by the diligent application of peer review, but this merely displaces the reader's attentional process to a consideration of the judgement of the peer reviewers, potentially also prone to the retransmission of bullshit. The detail and verbosity of scholarly refutation amplifies the attentional problem.

Whilst freedom of expression worthy of the name must surely include the freedom to bullshit, the solution surely lies in maintaining reputational systems that offer the user efficient filtering systems that enable the basic command "never show content by this author again." It would be an error to universalise such judgement, for we all have our own foibles and tolerances. For example the Facebook system assigns a single "interest score" to each post which is then used to rate each post for all users. Twitter's "follow/unfollow" mechanism, which delegates filtering to the judgement of each individual user is much closer to what is required. The Pirate Party's attempts to implement "liquid democracy," whereby rank and file members anoint experts as delegates on any particular issue is also worthy of study.
Drinking at the commenting firehose at a heavily-trafficked site can be made less overwhelming using ranking systems (see Slashdot). And so on.

Frankfurt's book is of course itself somewhat bullshitty, devoting as it does, a substantial part of its rather slender discussion to a lengthy argument establishing that bullshit is synonymous with humbug. (What is humbug?) But hey! It's a fun cite on any reference list.

Harry Frankfurt seems a most charming man. This interview is a better outline to his thought than the book. He has humbly acknowledged elsewhere his surprise that his university press wished to work up his essay into a slender book:

mercredi 8 février 2017

On Scout's pace

from the department of getting-around-easily

A friend's remark made me consider of which piece of writing I am proudest. Two things spring to mind: this thing I wrote for "The Listserv" about Scout's pace (more about the venue here):
Scout's pace means moving at a brisk pace by alternately walking and running. Baden-Powell himself recommended alternating every fifty paces, but I prefer twenties as it's easier to count. My perennial optimism about how long it takes to walk somewhere makes me do it, and I enjoy breaking records for regular trips. I like the control that comes from having a number of "gears" at my disposal--easy strolling, brisk marching, an unpressured trot, the truth of a life lived under my own steam, and feeling... fit and well.

Scout's pace is obviously only the English term: do it a bit harder and you're speaking Norwegian: a fartlek. As for what the technique's most legendary exponents--the nomads of the Kalahari--call it, I have !kno idea! It is only the blink of an evolutionary eye since we were all hunter-gatherers; and yet we have forgotten Scout's pace, with dreadful consequences. Many ills of the Western lifestyle--obesity, high blood pressure, heart disease, stress and depression--can be traced to our lack of exercise.

I, too lazy and too ambitious to walk all my life, have transferred this knowledge to the bicycle. I could have discussed the virtues of a 24:32 minimum development, or the necessity of mudguards, or the late 19th century campaign by cyclists for paved roads, that led, in time, to the lamentable car culture, but I wanted to stick to the basics.
And maybe the editorial I wrote for the BMJ before bike week one time, which its then editor was kind enough to describe as "a poem." (He rode a bike too). But it's hard to decide. The best paid thing I ever did had my name sawed off (thus the generous cheque). My online commenting has never earned me a thing, but the satisfaction that I might just, maybe, have changed one person's mind somewhere, just as they were ready to, is ample reward.

[ et en français ?  ]

lundi 30 janvier 2017

On the virtues of mechanical autonomy

from the keeping-it-on-the-road department

A dear Facebook chum wrote:
I love learning how to do new stuff on the bike and of course "having" to buy new tools....but it's just not a very good use of my time.
I replied:
You can't really ignore the technical aspects of cycling, and the fact that a bike is mostly fixable with only a few simple tools is part of its genius. You'll get quicker with practice. The availability of online video of expert mechanics performing just about any procedure is transformational.* It's well worth putting in a bit of effort, for your own comfort and efficiency, not to mention the manifold possibility of future gallantry. And even if you do decide to delegate to the bikeshop in future, you'll be a better informed client, which is a good thing both sides of the deal.
Cycling involves both a human and a machine. It's easy to feel developing mechanical skill is too difficult or complicated, but it really is all one: you'll be more confident out on the road, knowing you can handle any problem that arises—and you'll be riding on a bike that you know is set up just right for you, for maximum comfort and efficiency.

*This sorta thing:

samedi 28 janvier 2017