dimanche 13 avril 2014

A consultation with a social librarian

    I came upon GoodReads again this morning, and decided to sign up and give it a goI've been yearning for some medium-light stuff in English to complement my worthy trawl through the French classics, currently La Peste.
    GoodReads is like Facebook for book lovers: 25 million users have shared their literary preferences, and, using the power of the social graph, once you do that too, those magical inklings about what you "didn't know you didn't know" become yours for the taking.
    This technique worked dramatically for me. Told I needed to rate 20 books to start receiving recommendations from the site, I set off, but was being presented with digital screeds of books I'd never heard of, and didn't even care that I hadn't. After scrolling through many pages of this, I finally lighted upon Richard Feynmann's Surely you're joking Mr Feymann?, gave it a five star rating (heartfelt: I even bought it for a nephew last Christmas), and from that moment on, my universe at the site changed completely, and I started to be presented with much more plausible choices of stuff that I might actually want to read. I was reminded that I have yet to tackle much of Dostoyevsky and Kafka for example.
    There was a slight dip when I came to rating my preferences in sport, where, as a simple matter of attentional survival (84kb PDF), I limit my reading strictly to cycling. Amidst a morass of recommendations treating on American-rules "football" (this though the ball is hardly ever kicked), baseball, and basketball, cycling was nowhere to be seen. This time I had to resort the the site's search tool: a search and a click on Tim Krabbé's The Rider sufficed to transform the site once more into friendly territory.
    It's a pretty awesome site, it has to be said. So now the problem is the more familiar "so many books, so little time." But my impression is that the site offers a powerful attentional tool, enabling ready surmise of populist preference, whilst easily forcing it to turn to the interest at hand (say, by specifying philosophical works). For an autodidact this is invaluable. For example, you'd seen references to the 19th century philosopher Schopenhauer, but didn't know where to begin; the GoodReads recommendation looks like a pretty palatable place to start. And unlike at Amazon, the GoodReads reader reviews generally seem non-trivial and sane.

mercredi 2 avril 2014

Was this URL censored on Twitter?

Try as I might (and I need my beauty sleep) I could not manage to post a tweet with this URL in it tonight: http://www.ouest-france.fr/louis-le-duff-sinterroge-sur-son-siege-social-rennes-2079853

The story there describes how a Mr Duff, head of his large eponymously-named company, only agreed to move his head office to Rennes from current locations in London and Paris, because he knew there was going to be a new airport at Notre-Dames-Des-Landes. Three hundred jobs are reportedly at stake.

The idiocy of this position can hardly be described. Why does he need a new airport 90km away, when he's already got one 10km away on the outskirts of Rennes? I crafted a tweet to point out this basic absurdity at 2330h on 2 April 2014:

 M. Duff retirerait 300 emplois si #NDDL n'aurait pas lieu: http://www.ouest-france.fr/louis-le-duff-sinterroge-sur-son-siege-social-rennes-2079853 Mais il déjà existe un aéroport à Rennes! #bizarre

[Mr Duff will pull 300 jobs if #NDDL isn't built: http://www.ouest-france.fr/louis-le-duff-sinterroge-sur-son-siege-social-rennes-2079853 But Rennes already has an airport! #bizarre]

 and thought I was about to head for bed, when it failed with the following error:

'Your Tweet was over 140 characters. You'll have to be more clever.'

Well, actually, it wasn't. It was 110 characters, plus a valid URL, which would normally be shortened by Twitter's URL-shortening service to 20 characters. This gives a total of 130 characters, well within the 140 character limit.

But try as I might, that Tweet wouldn't post. I redrafted it. It was very like Twitter was overwhelmed, as happens sometimes, so I thought I'd complain
and get off to bed. But when that tweet posted so easily and quickly (i.e. not at all like a general server fail that you sometimes see at busy times with Twitter) that my curiosity was piqued.

So I thought I'd post another tweet, avoiding the hashtag #NDDL and the URL to the Ouest-France story, and see how I got on:
That posted very nice and smoothly. Hmm! This was looking, on the face of it, like a keyword censorship problem at Twitter. I used the is.gd and bit.ly URL shortening services. Those yielded 'internal server error(s).' It seemed to me that the powers-that-be had created some crummy propaganda (which really is of insultingly low quality) and didn't want sarky tweets flying around deconstructing it before it's had a chance to fool a few time-pressed loyalists.

So I alternated attempts to post the most refined version of the tweet (as follows):

M. Duff retirerait 300 emplois si #NDDL n'aurait pas lieu. http://www.ouest-france.fr/louis-le-duff-sinterroge-sur-son-siege-social-rennes-2079853 Et l'aéroport actuel de Rennes ne sert à rien?

with a few tweets on other subjects, as one might, while on general internet patrol.

The madness that is the public discourse over American healthcare never ceases to exercise a horrible fascination in me, for example:
Hmm. No problem posting that one either.

So what are they filtering on? Surely the #NDDL hashtag can't be entirely blocked, because that would be too obvious, and people would complain:

So #NDDL is alright. Must be that link. But I still have not succeeded in posting it in any way to Twitter from my location here in France. I complain

And start researching Twitter censorship. It seems that France is indeed the leader in this. On the spurious grounds of preventing the sale of Nazi memorabilia, the French government have demanded that internet players be able to filter content by national boundaries.

I, however, am not even trying to sell a Hugo Boss suit: I am legitimately participating in a public debate about whether to build a new local airport using my taxes! Leaving aside the environmental question for a moment, the proposed subsidy is €266 million euros (which is about €450 for every man, woman and child in Nantes)—for what? A useless white elephant further away from Nantes, with no public transport, when we're already proud owners of a prizewinning airport.

I regard this as prima facie evidence that tweets around a certain URL were blocked between 2330h and 0300h, when I stopped trying to tweet, and began to pull this post together. I really cannot believe that they are defending this completely idiotic story, which they obviously plan to circulate it as a major triumph in their new airport = jobs argument. Such a completely bogus argument. Mr Duff! Indeed!

Doubtless the filter will be switched off tomorrow and everyone can deny this ever happened.

In the meantime, I have engaged here:
But really! This is supposed to be the EU, not some banana republic.

vendredi 14 mars 2014

No more talks from Tony Benn

As is so often the way these days, I surmised the demise of a public figure by random occurrences in my Twitter timeline, such as this:
which is quite a nice way to get the sad news.

I did actually meet Tony Benn one day on Whitehall, sometime in 2003: opposing the invasion of Iraq, we were. He wasn't actually engaged on any more serious conversations at that moment, so I went up to him and shook his hand and told him how much I'd enjoyed a talk he'd given at Sheffield University Students' Union at the time of the miners' strike. He received the compliment without comment, smiled benignly and moved on. I was struck by how his dress sense resembled my father's: decent in a tweed jacket with leather elbow patches and brown trousers.
Anyway, here, in his memory, is the story he told, to that adoring student audience, back in 1983:

"So I was helping out on one of the miners' stalls outside a supermarket--we were gathering food, tins and whatnot, from the people coming out of the shop--when an old lady, obviously very deaf, came up and said [more than passable imitation of a Chesterfield accent] "What's going on 'ere then?"
"So, of course, I replied, we're collecting food for the miners, there's a strike on you know."
"And she replied, [more Chesterfield] Oh verra good. I'll 'ave a tin of beans then."
"And she took a tin of beans off the stall!"
[We all laughed with Tony, who of course had a wonderful beaming smile on his face as he told the story]
 "And of course, at the time I was flabbergasted, and I didn't know what to say. But I've thought about it, and you know, that old lady was right. Because that is what this strike is about: it's about defending the rights of ordinary people to live a decent life..."

Truly, a memorable homily. It was a big meeting: I'd say 1500 students had turned out to hear the great man, and we were not disappointed. He had that very rare talent, of great warmth and charm, and being able to put that over in a way that unites an audience.

Of course, he is as easy to criticise in hindsight as any of his contemporaries, left or right, but I think he was a very decent man, and I mourn his passing. Plenty of YouTubery out there to enjoy, but @medialens picked this one out, which does capture his late era persona rather nicely.

dimanche 2 mars 2014

Shocking resort to science to resolve question of English language usage

Almost two years ago, David Crystal posted a short article on his blog musing about when to use "for instance" and when to use "for example." Such pedantry is perhaps not the most thrilling topic, but I guess métier oblige, and Mme Beezer and I count ourselves among his fans: she even buys his books.

As Crystal was requesting the intuitions of his readers, and as anyone had yet to post, I thought it a reasonable opportunity for increasing my notoriety, and quickly drafted a few remarks. These stated my intuition that "for instance" was perhaps more likely to be used in spoken English, than "for example," and showed that "for example" was more common, citing evidence from the Google Books corpus. I also remarked that my hunch about spoken vs written usage would be testable, given access to the right corpus. Happily I managed to express all this, and still get the first post, which I'm sure are more read than others 'below the line'.

Even more happily, I'm now five weeks into the most excellent Future Learn MOOC on corpus linguistics, which has given me access to the British National Corpus, and some good teaching about to how to use it. Anyway, as we all now know on the course, the BNC comprises corpora of both transcribed speech and written texts, enabling me to test some of the intuitions I offered Crystal two years ago against hard data. Here's a table showing the relative frequencies of both terms in the spoken and written sections of the BNC:
SpokenWrittenWritten/Spoken
For instance51.4976.61.49
For example106.25257.42.42
(freq/10^6 words)(freq/10^6 words)(ratio)
Data: British National Corpus hosted at http://bncweb.lancs.ac.uk/
As you can see, the outnumbering of "for instance" by "for example" witnessed in the Google books corpus is replicated. I'm no statistician, but it does look as though "for instance" is relatively more favoured in speech than in writing, when compared with "for example."
As some of Crystal's other commentators (Lucy, Rick Sprague) remarked two years ago, fr'instance does slip more felicitously off the tongue, or, if you like that kinda thing, "is preferred for phonotactic reasons."

lundi 24 février 2014

Don't want a new airport? Tear gas for you!

Obviously I went to the demonstration against the proposed construction of an airport at Notre-Dame-des-Landes (#NDDL). My pics are here. I have captioned them all, though the default Flickr landing page gives no clue of this: click through to the view which does present the captions alongside if you'd like to read them.
Happily, the excellent Hervé Kempf, formerly of Le Monde, has written a very fair report of events on the day, which I entirely endorse, saving me a job here.
As far as I know M. Kempf doesn't actually live in Nantes, so I'll just add that I share the view of some quoted in his article that the tactics of the authorities towards the demonstration on Saturday were:
a) unprecedented: every Nantes demo I have seen has gone up the cours de 50 Ôtages, so to place a great barrier and a water cannon to prevent this is an obvious provocation: it's like banning a big Edinburgh march from Princes Street, or Londoners from Trafalgar Square; and
b) entirely counterproductive to the maintenance of public order.  As I have observed in the comments to the YouTube video embedded below, which shows some time lapse footage of this point of confrontation between riot police and demonstrators, the authorities evidently wished to provoke this violence. Of course the threat of violence is an effective way of reducing the size of demonstrations (for example, though, like me, squarely against the proposed airport, Mme. Beezer et fils were understandably reluctant to turn out for their corporal punishment by the state, a reluctance I endorsed). It might be an effective tactic on the part of the authorities, but it's a grave stain on their purported democratic credentials.

This debacle has reportedly resulted in at least six injured police officers, and a young demonstrator losing an eye. I myself don't appreciate being teargassed as I peacefully express my political views on a matter of legitimate public concern.

They're backing a loser though: I really don't know anyone here who's in favour of the proposed airport, and what poll data I've seen suggests that this is because very few people are. The so-called pro-airport demo a few months back was pitifully small, and all this "silent majority" stuff they're coming out with is frankly, nonsense. (That was a ruder word in the first draft). And don't forget that the natural pro-airport constituency (like, I regret to say, frequent flyer Mme Beezer) are pretty happy with the existing airport, which is actually part of Nantes, and easy to get to, like, if you want to go to and from Nantes (duh!), and would be even better (e.g. requisite 1.2k of tramway connected +/or direct rail) if it hadn't had the ridiculous 40 year planning blight hanging over the 30k-away-and-no-public-transport airport on offer following any "transfer" to NDDL.

The insane persistence of the authorities in the pursuit of this project, which if it ever made sense when first mooted in 1967, is, in 2014, quite clearly SHEER BLOODY FOLLY.  The chaos in the centre of Nantes this Saturday was the result of the ineptly provocative tactics on the part of the authorities and the man responsible, Prefet Christian de Lavernée, should resign. Must we really wait for the political figures who have most invested in the project to be sacked, retire or die before the plan is cancelled? Or will they see reason? I trust that comparisons with recent events in the Ukraine are inappropriate...
Update 14:45h Tue 25 Feb 2014: According to today's local papers (Presse Ocean, Ouest France) more than 100 people were injured, of whom ten police officers and twenty demonstrators required hospital admission. The most severely injured person was a 29 year old man who reportedly lost his left eye to a flashball fired by a police officer. The cost of the clean-up is estimated at €1million euros.
And while we're here, a wee note on the media coverage:
The establishment media seem to have dutifully devoted multipage spreads to the violence and damage to property, happily enabling them to avoid the rather awkward facts of the debate proper:
a) reputable surveys show that the majority is against the airport;
b) the enormous €266million subsidy the project requires;
c) the authorities' lawbreaking with respect to legislation on water, the environment, and endangered species;
d) the effect of aviation emissions on climate change targets, and future kerosene price trends.
Ho hum!
Update Mon 14 Apr 2014: The number of demonstrators losing an eye to a headshot from a flashball has now risen to three. As a James Bond villain once observed, "Once is happenstance, twice is coincidence, but three times is enemy action." In other words, one or two such injuries might be explained by bad luck or negligence, but the third is strongly suggestive of malice. Charges have, rightly, been filed. Ce n'est pas fini, as the French say.

vendredi 31 janvier 2014

HuffPo atheists: "Let (some) freedoms ring"

from the freedom-from-illegal-invasions-is-also-a-human-right department
   The Huffington Post irritates in so many ways—from its browser-jamming javascripts, to its hideous advertising, to its forcing its .fr content upon me, even if I specify a .com URL. Still, HuffPo's been hard to avoid on general internet patrol recently, after my impulsive decision to follow one of its Middle East specialists for a while.
   Thus it was that I came upon this conversation between two north American atheists, of Pakistani and Iraqi origin. Now I feel a certain extra responsibility and compassion for the people of Iraq—it was, after all, partly my taxes that paid for their country to be royally fucked over in recent years. Getting one's head round the situation there obliges a wider concern with the Middle East more generally.  But this is the kind of thing that I'd usually read quickly and move on: sadly, 'middle eastern atheists cop their share of oppression, and there's plenty to go around,' is not such a surprising message. But there was something about their mutual consensus that people in the West "take their rights for granted" while supporting, in the same paragraph, the illegal actions of Western governments towards their former countries, that I thought was maybe worth puncturing.
   Comments were apparently open at the foot of the article, but my attempt to post a comment (using my Twitter account) failed, I know I'll get spammed to death if I give them my email, and I don't particularly think my friends will be interested in a HuffPo Facebook app spewing all over their timelines. So I gave up. Someone remains wrong on the internet! But, following the strict artistic rule that it is the comments that are difficult to post in situ that are most worth publishing, here it is:
It is certainly a good point that many in the West take their rights for granted because the political struggle for them took place before they were born; but the invasion of Iraq was a great crime precisely because, in doing so, the US and the UK were thumbing their noses at the authority of the UN's Security Council, and by extension the post-WWII settlement. In his hurry to damn the American left, and roam the world removing dictators he doesn't like, the M. Al-Mutar seems to have forgotten this.

vendredi 15 novembre 2013

The job ain't finished till the paperwork's done?

It's been a fairly quiet week in the cabinet, and all my birthday books have arrived including the latest (2nd) edition of the European Science Editors' Handbook. It's a unique compendium of scientific and publishing information, so I've been doing a little reading in between some admin tasks, and pondering its various chapters on taxonomy, a subject that I haven't considered per se for, ooh, nigh on thirty years. So, when Ross Mounce's tweet bounced into my feed earlier in the week, my mind was particularly prepared to dive into what may be the last storm before the final triumph of the web.
Ross was drawing attention to a 94-page article in ZooTaxa from a group of eminent taxonomists whose point of view basically seems to be that if it's not published on paper, then it doesn't exist, at least for the purposes of establishing precedence. Convention required, until the rules were amended in 2012, that at least five paper copies of any article be lodged in five separate reference libraries. Now, on first reflection this position seems somewhat dinosaur-ish, but Dubois et al's meticulously documented critique of the slippery nature of online publishing should give web triumphalists pause for thought.

As I understand them, Bubois et al's view is that an insistence on a paper publication forces all concerned to fix upon a canonical version of an article that exists and is publicly available for sale or by request at a certain date. Precedence is sacrèd in taxonomy because the winner gets to name the newly discovered lifeform, immortalizing a hero along the way, and the discoverer is then cited as the 'authority' responsible thereafter. With the prize of immortal memory at stake--not quite as good as having an SI unit named for you, but close--a legalistic rule book has developed over the years to ensure justice is done.

Dubois et al. are particularly scathing of electronic publishing practices which change articles without updating the publication date and reference information. Taking the long view—many 18th century works of taxonomy are apparently still in current usage—they criticise the electronic publishing community in general and BioMedCentral in particular, for playing fast and loose with traditions and systems that have accreted over literally centuries. As the linkrot rate, even in scholarly publishing, reportedly approaches 40% about five years after publication, the traditionalists certainly have a point, and everyone concerned with scientific communication would benefit from reflecting on Dubois et al's justified criticisms, and considering what may be done to address them.

Paradoxically of course open access is part of the solution—lots of copies keeps stuff safe, and all that, but there is still, in my view, a need to force all concerned into producing a canonical version of an article, and the moment when a file is fixed for the presses to roll would be a pretty good time for that. Then there needs to be some sort of digital service that provides assurances that we're dealing with files that have integrity. All the technical problems for this are already solved in the software world, but it will be a while yet before the necessary technical changes have fully permeated into non-technical scholarly areas. So I'm sure there'll be some space for boutique paper publishers for some time to come.

Update 1830h: I had to rush out to pick up the boy from school moments after completing the first draft of this blog at 1630h, so it seems a fitting homage to Prof Dubois to tweak it a bit now I've got a moment, but leave all the publishing metadata essentially unchanged. (!)
Update 17/11/2013 1310h: 1) My own response to Ross' tl;dr plea, after the briefest of flicks through Dubois et al's PDF, was admirably on target, though I say it myself: 2) Dubois et al's article really should have been better edited. A document of 94 pages begins to demand either a table of contents, an index, or preferably both, for sparing yourself this effort wastes everybody else's time. The unwieldy nature of the article undoubtèdly undermines its rhetorical effectiveness. ("Sorry for this long article, I didn't have time to write you a short one."--> But then, if you're ranting for the canon of the centuries, perhaps this is not an excuse you can honourably pull. 3) Also, if it was proofread by a native English speaker, s/he missed a few Franglicisms, deep down the page, none grave, but still...