Slashdot Mirror
Lies, Damned Lies, and the UK Copyright Industry
artg writes "Ben Goldacre writes about invalid and misleading 'science' in the Guardian. Here's
his report on the statistics behind a recent press story that reported illegal downloading to involve 120 billion pounds worth of material."
It's the scumbags like RIAA gives lies a bad name. Lies keep marriage in tact, family together, friendship, gov't, you name it. Along comes RIAA and ruins lie's good name. Shame.
Big surprise. Everything that has come from this industry has been at best broad guesstimates, at worst intentionally spread lies. Trying to explain the demise of an obsolete business model without taking the obvious into account is hard!
"There's someone in my head but it's not me." - Pink Floyd, Dark Side of the Moon
Full article is posted on Ben's blog at http://www.badscience.net/2009/06/home-taping-didnt-kill-music/ (sorry Ben for the slashdotting) - the guardian tends to remove bits of his writing in print/on their website (for space reasons I assume).
I just read TFA in the paper (yeah, i'll hand my geek card in on the way out...) and it struck me that the most important thing that he doesn't mention is that there's no evidence that anyone downloading a pirate copy of anything would actually buy it if they couldn't download it for free. Therefore nothing is actually lost.
My guess is that 99% of the stuff "illegally" downloaded would never actually be bought if it wasn't there to download.
People don't spend less money because they get something that they would have payed for for free, they just spend it on something else.
My sig doesn't address Anons, sigs aren't visible to them.
And there I was thinking it was the credit crunch that has caused our economic problems, it's obvious now that the real problem are the millions of teenage girls downloading britney spears albums (or who ever is in at the moment).
So they added up all the bittorent users, multiplied the figure by 25, and assumed that was the total cost to the economy.
I'm sure the Blender team would LOVE to receive 25 pounds ($40) for every download of each and every one of their movies. Ms. Boyle would doubtless be substantially richer if she were given the same for every person who had ever downloaded (or watched on YouTube) a clip of her singing. More members of Ubuntu might be able to play space tourist if each and every file (whether it be a CD, DVD or just a patch) resulted in a $40 donation. Radiohead and Nine Inch Nails would be over the moon if each individual song they've released for free got them that in checks received via fan mail.
I'm not saying that all the legit material added together make a substantial chunk of the corrected figure, but rather that the researchers never bothered to consider the fact that the material is not of equal value and that some items have a value of zero. They assumed everything was illegal and everything had identical worth.
That goes beyond Bad Science. How many of you, in elementary/primary school, got taught algebra by being given shopping lists? Pretty much everyone? Good. It would be a pointless exercise if apples and oranges had the same price ($40 each), so we can assume your class used different prices for different object, right? Right. So. Hands up who can tell me what you could do then that these researchers didn't do now?
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Intellectual Property Minister David Lammy said the report brought home the impact illegal downloads had on the UK economy as a whole. "If we take as read the music industry's assumption that every download is a lost sale, then billions of pounds are freed up for ordinary people to spend of things of actual economic substance to keep local businesses healthy, rather than chasing phantom pseudo-value from things that have an inherent cost of production of zero. This makes the whole economy more efficient and lets money go where it is actually useful, rather than to Bono's numbered account in the Virgin Islands."
The government says it will be hard to change attitudes to free downloading, particularly from the entrenched old media parasites. "Studies consistently show that downloaders buy more music. We have to stop this and get them downloading dodgy rips from BitTorrent, rather than official high-quality versions from iTunes."
The report also noted that new, faster broadband services could increase file-sharing, which was already more than half of net traffic in the UK. The ISPs modestly declined credit for their part in helping Britain's financial future, noting that it was their customers, the great British public, who had voted with their browsers to do the hard work of keeping the country afloat.
http://rocknerd.co.uk
The Guardian tends to edit his pieces a bit when they put them up. If you look at his blog post on badscience.net containing the original version you'll see that sentence links to another Guardian piece about a study showing that people who download more also buy more music - he's quoting from that rather than making it up...
http://www.badscience.net/2009/06/home-taping-didnt-kill-music/ has the original and you'll see it links to http://www.guardian.co.uk/music/2009/apr/21/study-finds-pirates-buy-more-music
Poor form of the Guardian to remove that link.
Assuming only 1k electrons in a bit, then a bit would be 9.10938215(45)x10^-28 kg.
If we then assume that every p2p user downloads at 100kbps, then in 1 second a single user would have acquired 9.10938215(45)x10^-23 kg Assuming there are only 1 billion users in one second 9.10938215(45)x10^-15 kg worth of data would be transfered After approximately 317 years only 9.10938215(45)x10^-5 kg worth of data would have been transfered
I do not see how even one pound of data has been lost.
Ben Goldacre also makes up some facts, like this one "...for example, people who download more also buy more music."
No, you're wrong.
http://news.bbc.co.uk/1/hi/technology/4718249.stm
People who illegally share music files online are also big spenders on legal music downloads, research suggests. Digital music research firm The Leading Question found that they spent four and a half times more on paid-for music downloads than average fans.
Sure, with today's coinage, a 1 pound coin doesn't weight that much. But, originally, it was based on the value of a pound (mass) of silver.
By reading this signature, you hereby agree with the content of the above comment.
Exactly - this is basically the parable of the broken window. Also see: http://notnews.today.com/2009/06/06/downloading-keeping-billions-inside-the-uk/ .
Of course, I'm not surprised that the RIAA twist the truth, but to hear Government advisers falling for the fallacy? Either they are ignorant of basic economics, or they are intentionally being deceitful on economic matters. Either way, it's no wonder the economy is going down the tubes.
Slightly offtopic, I admit it, but please read it regardless.
"Lies, damn lies and statistics". "Don't trust statistics you didn't forge yourself". "70% of statistics are made up, 80% of all people know that".
And so on.
There's a reason for those jokes, and it's shoddy statistics. Often, it's not even malice, it's simple inaptness. Ok, far too often it's also malice. Numbers are just too impressive, and they have authority. People believe them. They are regarded as "hard facts". They are not "a lot", they're not "a few", they are a million, a billion, and so on.
Funny about it is, though, that people believe those statistics. Not much differently than they believe the fuzzy "a few" and "a lot" statements. Because they're unable to test them. Even if it is as easy to throw the "numbers" out the window as in this example. 25 pounds "damage" per infringment. Nuts? 25 pounds ain't even what a current blockbuster costs when you buy it on DVD (legally, ok? Not talking about those flying Chinese traders where you know you're buying a bootleg copy). But did anyone care to check?
Probably no. It was numbers. It was hard facts. Hey, they wouldn't dare to release information like this if they didn't fact check, do they?
Heh. It was printed in the SUN. Dunno about you, but I've made up my mind about the fact checking abilities of their reporters...
Anyway. It does hurt to see my original trade being abused that way. I'm a statistician, at least according to my degree. I was, and still am, fascinated with the ability to aggregate a whole lot of samples into a simple, understandable statement. Statistics can serve a valuable purpose if, and only if, they are used sensibly and earnestly. And NOT "creatively".
So here's a little guide how to use statistics and how to gauge their credibility:
If you don't get to see the sample or don't get any information about how the sample was gathered, throw it to the dump. I can easily "prove" that every single listener to music buys it and that no copying is going on if I pick my sample "right". It's easy to "prove" every computer gamer is a potential addict if I only look at people playing 10+ hours a day. If you don't get told what's the source of the data and what data they worked with, chances are good that the whole deal is rigged.
If it's a "voluntary", "opt-in" sample, throw it out. All those statistics based on online questionaires where people can sign up and go to to fill out forms if they're "interested enough" are worthless. You'll get samples filled out by people who have a strong opinion about the subject already. When there is an online questionaire regarding "too much internet use", what kind of answers do you expect to get? Worse, what kind of people do you think will participate at all? It's a rigged sample from the start.
If you don't get to see the sample size, throw it out. The sample size gives you a fairly good idea how much of an error you may expect. 1/N^2 is a good rule of thumb (with N being the sample size) for the statistical error. That doesn't mean that a small sample automatically leads to a huge error margin, 200 samples may be already good enough if they are picked well, and if they're not "hand picked" (see above).
If you don't get to see a mean, a median and a standard deviation, throw it out. It's easy to prove that everyone's doing quite fine on average, even in this economy, because on average everyone has enough money to live well. The mean says so (the "average"). Without standard deviation, you won't get to see that the average is nothing but an artificial number that has no reflection in reality. It's not that everyone has the average, there's some who have a TON more and many that have a LOT less. The median would easily tell you so (that's the "middle number" of the sample). Comparing mean ("average") and median ("middle") tells you a lot about whether your sample was homogenous or whether you have a few VERY different bits in the sample (which should have been cut from the stati
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Kids these days. What do you learn in school?
http://en.wikipedia.org/wiki/Pound_sterling#Anglo-Saxon
No, the pound coin in its current form hasn't been around for more than a few decades. But the pound as a monetary unit is more than a thousand years old and did indeed represent the value of one pound of silver. The first coin to to be worth this much was, afaik, the Sovereign which was introduced in the late fifteenth century.
What's your definition of "very long"?
May we live long and die out