Geoff Coupe's Blog

Reflections on life at “De Witte Wand”…

Year: 2008

  • Windows Live Photo Gallery Revisited

    Last month I mentioned that I was trying out the latest release of Windows Live Photo Gallery, and that I’d run into a couple of bugs. Subsequent to that, I’ve been in communication with the WLPG team trying to sort out the bug whereby WLPG doesn’t seem to be tracking changes to IPTC/XMP keyword metadata.

    I had documented what I’d seen and sent it off to Michael Palermeti, a Program Manager on the WLPG team for investigation. He reported back that the developers could not find anything untoward with the test files and data that I’d provided. They were unable to reproduce the failure to track metadata changes. This struck me as being very odd, since I was clearly seeing the behaviour on two systems at home (my desktop and laptop PCs). So I went back to do more testing. And I think I’ve found out what’s going on.

    It is, as I suspected, associated with the fact that I’m using hierarchical keywords. A hierarchical keyword means that, for example, my keyword trees is actually the leaf node of a hierarchy:

    Objects/built environment/settlements and landscapes/landscapes/natural landscapes/vegetation/trees

    What’s going on is that when I assign a keyword to an image, I also explicitly assign the parent keywords to the image. So when trees is added as a keyword to an image, I’m also explicitly adding the additional strings:

    Objects/built environment/settlements and landscapes/landscapes/natural landscapes/vegetation
    Objects/built environment/settlements and landscapes/landscapes/natural landscapes
    Objects/built environment/settlements and landscapes/landscapes
    Objects/built environment/settlements and landscapes
    Objects/built environment/

    And there’s the problem: WLPG doesn’t seem to register these six instances of hierarchical keywords (trees, vegetation, natural landscapes, landscapes, settlements and landscapes, built environment) properly when they are assigned at the same time as a group. Worse, WLPG will not respond to subsequent deletions or additions to any keywords on the image. It’s almost as though a repetition of a part of a keyword hierarchy has the effect of locking the file as far as WLPG is concerned.

    When I change the method of assigning keywords to simply assigning only the full string once, then for the most part, WLPG appears to be happy. It will correctly register the keyword as a tag, and also track subsequent changes to the keywords.

    However, there is one crucial set of circumstances where WLPG still does not work correctly. That is where I assign keywords that share part of the same hierarchy to an image.

    Let’s take an example. I have a number of images with the keyword natural landscapes assigned (I.e. I have the hierarchical string: Objects/built environment/settlements and landscapes/landscapes/natural landscapes assigned to the images). Subsequently, I want to refine the metadata of some of these images by also assigning the keyword trees to them. So I go ahead and assign trees – in other words I’m adding an additional keyword string:

    Objects/built environment/settlements and landscapes/landscapes/natural landscapes/vegetation/trees to the selected images.

    The problem is that WLPG does not register this change – it continues to display only the natural landscapes tag associated with these images. Because the trees keyword is actually part of the same hierarchy as natural landscapes, WLPG fails to work properly and add this additional keyword string as an additional tag.

    What I actually have to do is explicitly delete the natural landscapes keyword from these images before I add the trees keyword to the images. This also has the effect of never being able (in WLPG) of having both the natural landscapes and trees tags associated with the same image because they are part of the same hierarchy. This strikes me as being somewhat of a limitation… In fact, I would say that it’s a bug, since now I cannot have certain combination of keywords associated with my files and have them tagged correctly in WLPG. Take house, castle, and dining room. They all share parent keywords in common in my hierarchy, so within WLPG I cannot tag images in such a way that will distinguish between a dining room in a castle and a dining room in a house. I suspect that this all goes back to the design decision that appears to have been made for WLPG that selecting multiple tags is an OR function and not an AND function. In my opinion, that was a very bad decision, but that’s another story…

    Anyway, to sum up, I’ve now found out that WLPG cannot cope with multiple instances of hierarchical keywords that share the same parents. I hope that it can be fixed in future versions.

  • Wakeup Call

    Today I noticed a number of blogs referring to Wassup 2008. I had no idea what on earth they meant until I watched the original advert followed by the current version here. Simply brilliant. An American version of Grosz.

  • Taking Pride in Ignorance – Part II

    Following hot on the heels of Tracy – here’s Sarah Palin demonstrating once again that she hasn’t a clue, and is proud of it. Scary stuff.

  • X-Rays and Brazilians

    Dangers lurk in the most surprising areas. Perhaps a razor would be safer. I simply recall the outings to the shoe-shop and the xray machine in the corner…

  • Rudolph

    One of the current crop of web sites that gives me guilty pleasure is Not Always Right – a web site devoted to the dark side of customer service. Very often, I find myself shuddering at how people treat people. But every now and then comes an example of how humanity, even at its less than attractive side, can bring a smile of schadenfreude to the jaded soul…

  • The Atheist Bus

    So Ariane Sherine had an idea to put an advertising slogan onto London buses. A slogan that I could well stand behind:
    "There’s probably no God. Now stop worrying and enjoy your life"  
    Simple enough, worthy enough, and almost certainly correct. Of course, it didn’t take long for the damning with faint praise to begin. And as a direct result of that barf-making guff from Mr. Barrow, I have contributed to the campaign. Thank you, Mr. Barrow. I see that the campaign is currently far beyond the original target of 5,500 pounds and now stands (at the time of writing) at over 31,000 pounds. Good show, people.  

  • Taking Pride in Ignorance

    Be afraid, be very afraid…
     
     

  • I’m Still Here

    I see that ten days have passed since I last wrote anything on the blog. I think that is probably the longest hiatus since the blog began back in February 2005. Nothing untoward, I’ve simply been busy with other things. Normal service should be resumed soon.

  • Libraries

    A while back, I mentioned a collection of photos of the world’s most beautiful libraries. It seems to me that Jay Walker’s personal library can be added to that list. Quite stunning.
     
    And on an associated note, I see that Andy Burnham, the UK’s Secretary of State for Culture, is on record as saying that libraries are out of touch. Frankly he comes across as a barbarian. I’d make some comment along the lines of it’s like killing the goose that lays the golden eggs, but I fear that there’s a good chance that he would miss the allusion…
     
    Fortunately, Ophelia and John “the Wife” are on hand to deliver the slaps of contempt that Burnham so richly deserves.

  • Turning a Blind Eye

    A couple of stories this week made me think of how privileged some of us are, and how we ride on the backs of others. First up was Carole Cadwalladr’s piece in last Sunday’s Observer on the social strata and tensions in Dubai. And lest we think that the society in Western Europe is far removed from building the economy on the backs of slaves, Johann Hari’s piece about Chinese workers in the UK contains similar themes. Pacts with the devil take many forms, and we all turn a blind eye more often that we care to admit.

  • Homemade Wireless Keyboard

    How not to make your own wireless keyboard. The scary thing is that it took twenty seconds for the penny to drop.

  • Good Science

    Having mentioned Ben Goldacre’s book Bad Science in my previous entry, I just wanted to emphasise how good his book is. Really, you owe it to yourself to get a copy of this book to buff up your bullshit detector. The blurb on the back of the book puts it well:
    Ben Goldacre masterfully dismantles the dodgy science behind some of the great drug trials, court cases and missed opportunities of our time, but he also goes further: out of the bullshit, he shows us the fascinating story of how we know what we know, and gives us the tools to uncover bad science for ourselves.
    And it’s funny, to boot.
     
    Goldacre also has his own blog Bad Science, which is well worth keeping an eye on. I see that over the past few years, I’ve referred to things on his blog over 1,000 times. He’s very good at what he does.

  • Statistics and Lies

    I see from today’s Volkskrant that Lucia de Berk has at last won the right to a retrial of her case. De Berk is a nurse who was convicted in 2003 for the supposed four murders and three attempted murders of patients in her care.

    The history of the case makes chilling reading, not because of anything that de Berk may have done, but because of the web of statistical “proof” that the prosecution used to put her behind bars. It is perfectly clear that the statistical evidence was deeply flawed from the start, but here we are in 2008, and she has spent almost six years in jail for “crimes” that never existed in the first case.

    The judgement against her was based largely on the claim (from the prosecution’s statistician) that the chances of so many people dying on the wards where she was on shift were “one in 342 million to one against”. But, as Ben Goldacre makes clear in his excellent book Bad Science, the fundamental flaw about this claim is twofold. First, the data was selected to make the hypothesis, and then the prosecution’s statistician made a simple, rudimentary error: he combined individual statistical tests by multiplying p-values (the mathematical description of chance, or statistical significance). As Goldacre points out in respect of the first part of the claim:

    A huge amount of corollary statistical information was almost completely ignored. In the three years before Lucia worked on the ward in question, there were seven deaths. In the three years that she did work on the ward, there were six deaths. Here’s a thought: it seems odd that the death rate should go down on a ward at the precise moment that a serial killer – on a killing spree – arrives. If Lucia killed them all, then there must have been no natural deaths on that ward at all in the whole of the three years that she worked there.

    And in respect of the second flaw, Goldacre points out:

    If you multiply p-values together, then harmless and probable incidents rapidly appear vanishingly unlikely. Let’s say you worked in twenty hospitals, each with a harmless incident pattern: say p=0.5. If you multiply those harmless p-values, of entirely chance findings, you end up with a final p-value of 0.5 to the power of twenty, which is p < 0.000001, which is extremely, very, highly, statistically significant. With this mathematical error, by his reasoning, if you change hospitals a lot, you automatically become a suspect. Have you worked in twenty hospitals? For God’s sake don’t tell the Dutch police if you have.

    It’s a very cautionary tale of statistics gone horribly wrong, and very reminiscent of the Sally Clark case in the UK (which Goldacre also dissects). Clark was put on trial in 1999, and convicted, for murdering her two babies. At the trial, child expert Professor Sir Roy Meadows stated that the chance of two children in the same family dying of Sudden Infant Death Syndrome (SIDS) was “one in seventy-three million”. It was another case of statistics wielded in error, and Clark spent three years in jail (where she was targeted by other prisoners as a supposed baby-murderer) before her conviction was quashed by the Court of Appeal. She emerged a broken woman and died in March 2007. I fervently hope that that will not be the fate of Lucia de Berk.

  • Pot, Kettle, Black..

    I think my irony meter has just exploded…

  • A Small Incentive

    Justin, over at Chicken Yoghurt, has come up with a rather whizzo scheme to drive the FTSE share index back up again. It has a rather Ballardian whiff about it…

  • Know Your Human Rights

  • A Really Bad Disney Movie…

    … but we seem to be edging dangerously close to Head of Skate becoming a disastrous reality.
    Oh, and this piece written by Matt Taibbi is like being in the front row at the Grand Guignol and being spattered by the blood and gore. The trouble is, there’s a good chance that we won’t be able to leave at the interval. A small sample:
    So, sure, Barack Obama might be every bit as much a slick piece of imageering as Sarah Palin. The difference is in what the image represents. The Obama image represents tolerance, intelligence, education, patience with the notion of compromise and negotiation, and a willingness to stare ugly facts right in the face, all qualities we’re actually going to need in government if we’re going to get out of this huge mess we’re in.
    Here’s what Sarah Palin represents: being a fat fucking pig who pins “Country First” buttons on his man titties and chants “U-S-A! U-S-A!” at the top of his lungs while his kids live off credit cards and Saudis buy up all the mortgages in Kansas.
    The truly disgusting thing about Sarah Palin isn’t that she’s totally unqualified, or a religious zealot, or married to a secessionist, or unable to educate her own daughter about sex, or a fake conservative who raised taxes and horked up earmark millions every chance she got. No, the most disgusting thing about her is what she says about us: that you can ram us in the ass for eight solid years, and we’ll not only thank you for your trouble, we’ll sign you up for eight more years, if only you promise to stroke us in the right spot for a few hours around election time.
    Go and read the whole thing – and weep, not just for America, but for the whole world.
    Update: The ever-dependable Jonathan Raban has written an equally good piece on Palin for this month’s London Review of Books. It delivers a cool, surgically-precise flensing of Palin in contrast to Taibbi’s hatchet job.
    Addendum July 2017: …and, wouldn’t you know it, Mr. Taibbi turns out to be a despicable human being himself.

  • The Placebo is God

    Following hot on the heels of that august organ of journalism, the Daily Mail, today’s Guardian also jumps on the bandwagon of the latest "let’s all misinterpret the science" story. Yes, it’s the "Religious belief can help relieve pain, say researchers". Well, well, what a surprise: it’s the placebo effect of course. Yet another pronouncement from the department of the bleeding obvious, I would have thought.
     
    People tend to underestimate the power of the placebo. As a cure for this debilitating condition, I recommend a simple remedy. Merely purchase a copy of Ben Goldacre’s Bad Science and read chapter 5: The Placebo Effect. Instant relief and the realisation that "We are human, we are irrational, we have foibles, and the power of the mind over the body is greater than anything you have previously imagined".

  • So van Gogh Was Killed…

    I don’t know who this Christopher Howse person is, but he strikes me as being either a) an idiot or b) will write any old tosh for money. Either way, his piece in today’s Telegraph leaves a particularly nasty taste in the mouth. He’s not alone, today’s Guardian has a letter from Dr. Charlie Gere informing us that there is no such thing as free speech. With friends like these, who needs enemies?
     
    Ophelia is on hand to dismantle their apologia with the contempt that they deserve.

  • A Fair(y) Tale

    Alright children, gather round and let uncle Geoff tell you all about the tale of Copyright and Fair Use. Once upon a time…
     
       
     
    (hat tip to Nina Paley). Oh, and may I just say that I was pleased to see Sleeping Beauty in there. It may not have been reckoned as one of the great Disney films, but for my money the medieval style of the backgrounds achieved by Eyvind Earle were one of the great examples of the animated film.