Search This Blog

22 November 2010

Babies and Cross-Promotion

Besides having a new baby - my other excuse for not posting is: working on a new band name generator.

So - as a shameless piece of cross-proomotion... and because I don't have time to post on any of the other (more impotant) issues I've been wanting to write about...

Here's "Banda", the band name generator:

From The Blurb

Banda is a band name generator.

It uses a database of common English words to randomly generate a new band name for you - every time you click "get a new one".

It also allows you to vote on your favourite band names - and to post any band name you really like to facebook, twitter... etc.

Banda - it's like Panda... with a 'B'!

14 October 2010

God Would Have Chosen Random Selection (or The Peter Principle Goes to Church)

New research, out in February - shows that the best process for selecting which people to promote, out of a group, is a random one.

It shows that one of 2 random processes, either:
  • selecting people completely at random, or;
  • selecting (at random) either the best of the worst candidate
produce more efficient outcomes, for any organisation, than any of the other strategies tested.

It is not lost, on the researchers, how counter-intuitive this is.

This research has now won one of the IG Nobel Prizes - the spoof prizes based on the Nobel Prizes.

The Point?

The missing realisation here is... it's a big one for the theists. Yes, that's right, the theists!

Because... it shows, without a doubt, that, if God exists - assuming that she is perfect, and therefore chooses the most mathmatically efficient process for any given outcome - she most certainly would have have chosen "random selection" as her preferred method for the development of her pet project (i.e. us).

[N.B. We will ignore, for the sake of argument, the fact that the theory of Evolution does not actually propose "random selection" as the selection method of choice - it proposes that individuals are "chosen" by survival of the fittest from a pool of randomly mutated options... but who said Evolution was perfect? Not the evolutionists, certainly... but God on the other hand... well...]

Achem... anyway. My point is a simple one.

No matter which way you look at it - the Creationists are barking up the wrong tree... even if God has designed it all - random selection is still the most efficient method of promotion... and so... it doesn't need a hands-on designer... any "intervention" that he/she/it performed was doomed to be less efficient than just letting it develop randomly... all God needed to do was push the "Go" button and sit back and watch her creation improve itself via the most efficient method...


Oh... but hold on! If you don't NEED a designer to find the best / most efficient solution then... oh dear...

Ed Note: Any implication that I may, or may not, have just ruined the Intelligent Designer's argument was completely unintentional - and was achieved purely through a random process of pulling ideas out of my head until something worked...

Quite efficient, I find.

REFERENCE: "The Peter Principle Revisited: A Computational Study," Alessandro Pluchino, Andrea Rapisarda, and Cesare Garofalo, Physica A, vol. 389, no. 3, February 2010, pp. 467-72.

In the late sixties the Canadian psychologist Laurence J. Peter advanced an apparently paradoxical principle, named since then after him, which can be summarized as follows: {\it 'Every new member in a hierarchical organization climbs the hierarchy until he/she reaches his/her level of maximum incompetence'}. Despite its apparent unreasonableness, such a principle would realistically act in any organization where the mechanism of promotion rewards the best members and where the mechanism at their new level in the hierarchical structure does not depend on the competence they had at the previous level, usually because the tasks of the levels are very different to each other. Here we show, by means of agent based simulations, that if the latter two features actually hold in a given model of an organization with a hierarchical structure, then not only is the Peter principle unavoidable, but also it yields in turn a significant reduction of the global efficiency of the organization. Within a game theory-like approach, we explore different promotion strategies and we find, counterintuitively, that in order to avoid such an effect the best ways for improving the efficiency of a given organization are either to promote each time an agent at random or to promote randomly the best and the worst members in terms of competence.

13 October 2010

Power Balance meets new ACCC powers - bring it on!

Back in April, the ACCC was given some powers that I had thought were a good idea, for some time...

See the article here for more details.

But now, Christopher Zinn believes he's found the perfect candidate for testing the new laws.

The Power Balance bracelet has been called a "blatant con" by the Australian consumer watchdog, Choice (read report).

So, watch this space, for more exciting updates. Could it be we're entering an age, finally, where consumers can hold manufacturers properly accountable for their claims?

Here's hoping.

What other products do you think would be best for testing under the new laws? List them here in the comments - and I'll send a list of the best ideas on to the ACCC.

More references:

[Ed Note: for follow up, see post here]

30 September 2010

Pricing Carbon, Taxes and the ETS

OK - so I haven't posted for a VERY long time...

I HAVE been busy, producing a theatre show - but, excuses aside - here we go, finally:

For years I have been saying I wanted to write a post in defence of the ETS. Many people - many intelligent and knowledgeable people - have asked me why the Australian Government (and this goes back to the days of PM Howard here) supports an ETS over a carbon tax.

I actually think an ETS is a good idea. I think it's the only practical idea, in the long-term. I think it's the only effective idea, given the practicalities of government and I think it's the most efficient solution economically speaking. I'm THAT convinced.

But, as we all know, the Rudd government did an AWFUL job of explaining themselves.

We can excuse the Howard government for not explaining it properly, because, if you probed far enough, the truth was, they didn't want to do anything about the problem. They just proposed the ETS as their preferred option ... "if we have to choose something I guess we'd prefer an ETS".

But Rudd - that's where it really failed. He really wanted to do something (didn't he?).

In 2007/08 we had the Rudd-Slide - and I thought I'd missed my opportunity to explain why it was good idea, because it would just happen. Then in 2009 I thought I missed my chance to explain why it was a good idea because Copenhagen was going to ignite interest in the issue... then in 2010... well that’s when I should have got around to it - because it was damn clear the government wasn't going to explain their own policies to anyone.

So here we are, at the end of 2010 - and STILL no one is clearly explaining why an ETS would be a good idea. So I'm going to try, in my own little way, to do just that...

Before I try to lay it all out though... I must pre-empt the end of the story. Another idea has occurred to me, which I believe nullifies the only good argument I've ever heard against the ETS (that of the difficulty of getting people to report clearly in the lead-up years). But you'll have to read to the end, to follow the argument.

The Argument

Any possible solution for the problem of Global Warming (I'm taking that problem as a given, by the way... if you need to discuss that topic first, this probably isn't going to be a very useful discussion for you) should be judged on 3 basic measures:

  1. Effectiveness: It's ability to reduce carbon emissions
  2. Efficiency: It's cost to the overall economy
  3. Practicality: It's likelihood of succeeding in the given political system
In a broad sense there are 3 basic models put forward:
  1. Direct legislation to limit carbon emissions
  2. A carbon tax
  3. An ETS


So basically, the question is - how effectively does your plan reduce carbon emissions? How well does each given plan reach the particular goal of desired carbon output? And the answers are as follows:

  1. Direct legislation: quite effective, actually, but sometimes unknown or unpredictable (can go over the top and damage the industries involved more than necessary - but that's not what we're discussing at this point)

  2. Carbon Tax: completely unknown - depending on the level of the tax, it can completely kill the industries involved, have no effect on emissions or, if you get the value just right (and I mean "just right"), reduce emissions the amount you want.

  3. ETS: This is where the ETS really shines - you set the amount of maximum carbon output for the economy as a whole, and the market makes it happen by setting the right price for carbon. No more, no less - you get exactly the amount you said you wanted (ignoring illegal output, which applies to all 3 plans anyway).

For a really clear summary of the mechanisms involved, from some industry experts, listen to ABC Radio National - Australia Talks episode on carbon pricing. [I will add a link to the specific section I'm talking about, when I get time to edit the file]

The whole talk is quite interesting, in fact, and includes parts near the beginning where an expert from an energy production company actually makes an argument for putting a price on carbon! Who would've thought?


Here, the question is - how efficiently does the given plan reduce carbon emissions (to the desired amount)? How much does it cost, in total, for the economy to adjust to the changes and find a new equilibrium? And the answers are as follows:

  1. Direct legislation: There is some disagreement about this - but in essence the argument against it being efficient comes down to one idea: Government and Independent Bodies can't predict far enough in advance how best to reduce emissions. Don't legislate how it should be done... create incentives and let the cut-and-thrust of commerce and innovation work it out. Direct legislation seems, by all reports, the least efficient option.

  2. Carbon Tax: While a Carbon Tax is not the least efficient, it does seem to have one particular inefficiency. It allows the worst polluters (often the polluters with the highest profit margins) to keep paying to pollute. A carbon tax then, also, has the greatest financial impact on the (often lower-level) polluters who don't have the same high-margins. [N.B. While some people complain about an ETS that it allows the worst polluters to quickly make money on the cap-and-trade market by making reductions that were easy to make and should have been done already... this is exactly why it's more efficient than a carbon tax - it gets the attention of those who can most easily (for "easily", read "efficiently") reduce their carbon output.]

  3. ETS: By all reports, an ETS IS the most efficient way to reduce emissions across the whole economy. A similar scheme (to the one proposed for Australia) was introduced to reduce SO2 emissions in the US, in 1990. It was reasonably effective - and when the efficiency (total cost per reduction in emissions) is calculated it stands clearly above previous attempts to do similar things via other methods. For a more detailed summary of this program and the findings / lessons:


And finally, the last question is - how practical is the given plan? How likely is it to succeed (in the long-term) given the changeable nature of politics and the personalities and pressures involved? And the answers are as follows:

  1. Direct legislation: The problem for direct legislation, in the political arena, is that it's too easy to make a general argument that "that's not fair on me because X" - and it's too hard to balance out the differences for everyone involved. The moment you make a concession for one part of the economy, another will jump up and make a reasonable (sounding) argument for their own part of the economy. The only "fair" solution is one in which each area of the economy takes a chunk of the responsibility based on a range of sliding parameters. And politicians will never be able to wade through the sea of barriers that individual interest groups will put in their way (even if those politicians didn't have any self-interest involved, which of course they do).

  2. Carbon Tax: A carbon tax, by comparison, is simple and much more likely to get through, in the first place. But what about after that? What about in 5 years time, when, by some miracle, the politicians managed to set exactly the right tax level, and have reduced our emissions to 90% of 1990 levels, but now the emissions are creeping up again, because innovation means it's much cheaper to produce carbon, and carbon producers can afford to pay more tax to do so... You've got the remember, it's the total carbon output, not the total cost, or revenue we're interested in here... who's going to guarantee that the next government would make the argument to increase the tax by the required amount to keep us on track to a sustainable level of output? No one, that's who... we'd be in the same boat, all over again - right back where we started... fighting the same fights.

  3. ETS: In comparison, an ETS is a self-regulating system. Set-it-and-forget-it. Well, not quite, I hear you say... you still need to make the political argument to reduce the emission levels... Ah yes! But saying, to the voting populace, "we need to reduce emissions further, let's set the dial on emissions a little lower" is a MUCH easier political position than "let's increase taxes again". You tell me which you think is going to be more sustainable... Besides which, you could set-in a program of clearly defined year-by-year reductions, right now - and they are much less likely to need adjustment than tax levels... because when we set the level on an ETS we're saying exactly how much carbon we are going to allow, not guessing how much carbon a particular tax level will generate.

The Only Problem

The problem, with an ETS, that I have seen examples of, is this: how do you work out how many permits to create in the first place?

This problem was faced by the introduction of an ETS in Europe. And they failed to avoid it.

They asked people to report how much carbon they produced... and of course, they over reported, so that there would be too many permits - and permits would therefore be cheaper.

Conversely, a carbon tax has the opposite problem. It asks people to report - and they under report, in order to avoid tax.

And that's what brought me to the solution:

A carbon tax can be introduced much faster and more easily than an ETS... a carbon tax encourages under-reporting... an ETS encourages over-reporting.

  1. Introduce a carbon tax - in the next budget
    • A small introductory one, with indications that it will rise over the next 3 years.
  2. Ask people to report their emissions, as they would need to, in order to calculate a tax
    • Obviously some form of auditing would need to check-up on people's self-reporting - as is the case for any form of tax or trading system.
  3. BUT - and here's the important bit - make it very clear that there WILL be an ETS introduced at the end of that 3 year period
And, so what? Where does that leave you?

Well, it leaves you with companies, across Australia who all have a vested interest in reducing the reporting of their emissions, over the next 3 years, in order to avoid tax - but who are also clearly aware that any attempt to do so will mean not enough permits to go around and much more expensive permits, in 3 years time.

You have 3 year to prepare for the introduction of an ETS. The tax level can be adjusted in the s2n and 3rd years, as its effect becomes clearer over time.

You might still get some under reporting... and you might even get some over-reporting in the last year, on the calculation that paying a bit of extra tax, in the last year is worth getting more permits into the system eventually... but on balance I think most people will recognise that the most efficient way to report is honestly.

Thus, in my opinion, reducing one of the only major difficulties with the initiation of an ETS.

Until then, let's hope we can get SOME price on carbon soon - so that it starts getting factored in to future growth and planning.

It's sad to reflect that, after all this time... this blog post is still as relevant today as when I first started writing it in my head... more than 5 years ago...

Can we move on now?

Other References:

05 May 2010

How bad was the government's insulation scheme, really?

Some months ago, I read a piece in Crikey which analysed the numbers on the insulation scheme - and how much they increased risk for the people who had used it.

I found the numbers fascinating. Basically, they show that the risk of fire from insulation, per insulation, was actually reduced by the introduction of the scheme, and the extra regulation and safety measures it brought in.

I won't rehash all the details here. You can look them up in the original Crikey article.

But I did want to say that it's not entirely unbelievable. The insulation scheme actually introduced safety regulations to an industry that previously had absolutely none - so however much the opposition wants to claim that the safety measures "didn't go far enough"... at least they went somewhere.

What I am seriously disappointed by is: if these figures and analysis have any basis in reality, then the government has completely failed to explain this to anyone, and has, apparently, simply walked away from the issue because it was a "hot potato". These figures don't make what the government did "alright", they just make exactly what wrong they did a different issue. If the government has walked away from a policy that was, in essence, working - and avoided the hard job of explaining this to the populace, because... well, I don't know... because they thought we were too stupid to understand, then... IMHO that's even worse!

I wanted to try to bring attention to this matter again. As the election grows nearer, more and more references are going to be dragged up by the opposition in reference to this matter... but the real issue isn't that the government stuffed up their policy - it's that the government completely failed to explain how their policy, and implantation of it, was actually working.

Let's understand, and complain about, the real issue here.

23 April 2010

James Randi, Psychics and Homeopoathy

After some rights issues (as mentioned on the James Randi Education Foundation web site), TED have finally released a video, filmed last year, of James Randi's take down of phychic fraud and homeopathy.

For those of you who know Randi... it's all been said before - but he does it so well - with so much passion, and such fun:

Thanks Randi!

19 April 2010

Homeopathy wins!

I had to share this one...

I love the fact that one of the oldest Double-Blind Randomised Controlled Tests ever recorded was performed on homeopathy... but I won't take the credit for the research... let Joseph Albietz tell you all about it.

Oh, it failed the test by the way... but it won in another way. Homeopathy is possibly the first theory to ever be disproved by the implementation of the Double-Blind Randomised Controlled Test.

15 April 2010

Irreducible Complexity

For those of my friends (a couple at least) who believe that the Intelligent Design proponents might "have a point" - I would like to take a moment to post this piece... it's from The SGU 5x5 (The Skeptics Guide To The Universe, 5 by 5).

I found it one of the best, succinct, refutations of the "Irreducible Complexity" argument that is often put forward by people who support the idea of Intelligent Design.

On the same subject, essentially - I also found this, shorter, piece on Teleology (the generic argument behind Intelligent Design), its history and the reasons for being skeptical of its implementation.

Basically, it efficiently outlines the argument against the idea behind Intelligent Design.

Nice work from the SGU team.

So, "what's the harm?" you might ask

I really quite like this...

I found an interesting web site, today - dedicated to answering the question: "What's The Harm?"

It's a soft argument, often used by people of, what I would call, a non-skeptical nature... to refute the general argument that any false claim (by proponents of drugs that don't work or procedures that do little good) should be tracked down and advertised as non-scientific and unproven.

"What's the harm?" they say. By this, presumably, they mean "if it doesn't do any harm, then there's no problem with allowing people to keep believing it works, whether it does or not".

Now, besides the fact that I would disagree with this basic argument, on principal (more about that some other time, maybe)... even accepting the argument as a valid moral position - the truth of the matter is that many ineffective drugs, non-scientific procedures and false claims do do harm. I have heard many such examples over the years, usually while watching interesting talks from well known skeptics such as James Randi and Simon Singh (who won his libel case today - yay!).

The problem is, I always forget the details, and can't draw them up from my memory banks in a convincing enough way, when presented with the need to do so in argument...

And here's the answer: a web site devoted to recording, documenting and measuring the harm done by fallacious claims and bad (or non-existent) science.

What's The Harm

So bookmark it... and the next time someone says to you "but what's the harm" pull out your smartphone (no product bias here, thanks), load up this page - follow the link to the topic of the hour, and read out a few examples...

If the person you're talking with doesn't become violent with rage over what a smartypants you are, you might just have managed to make a well deserved point...

Have a quick browse through some of the topics covered. I found it quite interesting to see how much detail, and how many examples they've managed to collect.

Simon Singh wins libel case

The British Chiropractic Association, today, dropped its libel case against Simon Singh, the science writer.

In what must be a great relief to many active and outspoken skeptics, across the UK, and around the world - the court of appeal in the UK overturned the previous judgment, that Singh's piece was "not comment"... that he would need to prove the objective truth of what he wrote.

Interestingly, the previous judgment basically said "you need to prove that Chiropractic doesn't work, or we can sue you for saying it"... which of course would lead to all sorts of possible libel cases regarding people speaking our against unproven claims.

Imagine if pharmaceutical companies could say "'this product cures cancer' - and we can claim it does, on the bottle AND you can't print a word saying it doesn't, until you can prove that it doesn't". The onus of proof would then be on the consumer, or the skeptic, to prove the non-efficacy of a product before anyone would be allowed to say it didn't work.

Anyway, the point is, the decision was overturned - and the BCA have dropped their case.

As Rebecca Watson from Skepchick points out, though, Simon Singh may have a hell of time recouping his costs - and he may still have made some significant financial sacrifices in order to see this case through, and not simply settle "out of court".

More links on the story:
Telegraph UK
Times Online

[Ed: Latest update. This from "The Millenium Project" - Simon Singh again]

Sarah Mahew - grab on to what is true

I was quite excited (and, kind of touched) by this video:

But, I am confused as to what "Nokia Responsiveness" is.

I'm happy to see the inclusion of this story here - but what exactly is the point of Nokia Responsiveness and its attempt at civic "conversation starting"? Can anyone enlighten me?

But - actually, just replay the video... and listen to that inspiring question at the end... "do we have the courage to let go of our beliefs in order to grab on to what is true?"

01 March 2010

Conroy already filtering his own site

Would I sound paranoid if I said Stephen Conroy's website is deliberately concealing users' searches for "ISP Filtering"?

Well the evidence is on the page itself:

//Customise the tag-cloud to display what shows up
if (unique[i] == "ISP Filtering")
Basically, this piece of code simply says "if the value in the List you are displaying is 'ISP Filtering' leave it out". It's there in plain code, in the HTML of the page you download from his site. However many searches anyone makes on "ISP Filetering" it will never be displayed in the list of users' searches - therefore giving a false impression of what people are actually searching for.

It claims to give you information on what users are interested in then specifically alters that information for, what can only been assumed to be, Senator Conroy's own purposes.

It is also a very blunt solution that obviously wouldn't catch values such as "Web Filtering" or "ISP Censorship". Not only is it surprising behaviour - it's also an ineffective, amateurish and clumsy solution to a problem (that he shouldn't have been trying to solve in the first place).

The worst thing about this is - if Senator Conroy doesn't understand what the difference between this and properly removing the results (on the server side) is, then he is an embarrassment to his portfolio and doesn't deserve the role... if he does understand the difference and can't be bothered fixing it "properly" (so you and I can't simply see it ourselves) then he simply doesn't think this is an embarrassing thing to do.

I think the former is more likely - but either way - he just doesn't get it.

More details - article on the subject.