Sunday, April 29, 2012

Interesting. ish. In a, cautious, hopeful sort of way.

the Queen of Thorns has been railing lately (and justly so) at the lack of vision present in the speeches of New Zealand politicians in general and in Labour politicians in particular. This speech from David Cunniliffe isn't, I suspect, going to make her any happier. It's not bad. Or at least, the structure is not bad. I've got no video of it so I have no idea what the delivery was like. The content on the other hand, I think is good. It suggests to me that there are some senior Labour politicians who can see the world clearly, that neo-liberal market strategies are not the solver of problems that their supporters like to present them as. It does leave me wondering why this sort of message doesn't come out of Labour more often or in the bigger headline speeches. He evens seems to get that it wasn't a whole bunch of people deciding to vote National that lost Labour the election, that rather, it was a whole bunch of people who decided to not vote all (specifically not voting for Labour). The sooner the Labour party strategists get that through their heads the better.

Friday, April 27, 2012

The modern expert.


Now this is something worth watching. Particularly if you still believe that the talking heads in the media know what they're talking about. It's lightweight, as in it doesn't go into to much depth. And personally, I don't think it makes enough of a distinction between the expert in the media and the experts who actually know what they are talking about. Who do exist. It is mainly, I think about the dilution of the meaning of the word expert. Which is, I believe a major problem for us as a society. It's not something you can regulate though so it's going to remain a problem. Still, watch this, I find that most things with Ben Goldacre in tend to be worth watching.
"The Trouble With Experts" - bengoldacre - secondary blog

Thursday, April 26, 2012

Misdirection from fear

Australia has it's share of politicians who are denying the science behind climate change. Via Open Parachute, Naomi Oreskes, whilst talking to one of them, one Nick Minchin (I wonder if he's any relation to Tim Minchin, I'm sure Tim would be horrified). It's an interesting view, namely that a decent chunk of the deniers are attacking the science because they don't want to confront the consequences of climate change.



Which to me sounds like fear. I don't think it's fear of the costs. A decent chunk of the numbers that I've seen suggest that a lot of the measures needed to control climate change either save money or are opportunities for new industries. Which suggests to me that it's fear of actually doing the work. Fear of getting stuck in and doing the work. The politicians - and I do think it's the politicians, I've seen sufficient arguments from various engineers who think that the technology to switch to a low carbon economy exists now and that it just a matter of scale that requires political will - are comfortable, content even with the status quo. They are afraid of facing up to the consequences of our actions.

Tuesday, April 24, 2012

Emergent systems.

One of the common stereotypes that scientists often get labelled with is that they reduce everything to dry boring nuts and bolts, taking the beauty out of everything. It's one of the reason I'm not particularly fond of stereotypes and broad generalisations - I know which ones can be applied to me because of my job title and how horrifically wrong they are. This one in particular, is complete bollocks. Or kinda nutty if you (like Richard Feynman) want to be polite.

Why is it bollocks? Personally, I think Feynman's reasoning is bang on. The beauty available to the eye is not the only beauty. Complexity can be a wondrous thing, messy or elegant, operating at many different levels. One of the biggest kicks I get, and I suspect, the main reason that I'm studying systems biology, is when I figure out how a system of parts combines to change something or create something new. In biology, especially at the cellular level, the number of things that contribute to the various processes is huge. Genes are important obviously, metabolites which can be affected by a variety of things such as your diet, proteins - dependant on which genes have been switched on or off, even the relative location and time of all these components can be important. And there's numerous systems affecting those systems as well - diet being one, the day/night cycle being another, all sorts. With all these systems interacting it gets horribly complex very very fast. Which is why I tend to bang on a bit about statistics sometimes, there's a chaotic element to the combination of all these systems that make it tricky to understand without a good understanding of the underlying statistics.
Yet out of these seemingly chaotic systems, working organisms rise. It is the large number of interactions that means most processes do occur when they they are meant to - the body responds the right way to different types of food or a plant responds the right way to an infection. From the combination of these various systems, coherent systems emerge.
Sorry if that's a little waffley, it's a heads up on what I'm going to be trying to work into my posts over the next few weeks, I hope it'll be interesting at the same time as I get better at explaining it.

Monday, April 23, 2012

Read (and listen) widely.

One of the things that I think is important, not just for scientists, but for everyone is that one should have interests outside of primary interest. Systems biology could, reasonably be called my core interest, it's what I do day to day and I do it because I enjoy it and I want to know more about how it works. Outside of that, I've got cooking, gaming, coffee, making beer and assorted spirits, a little bit of acrobatics and a wide variety of reading. The reading encompasses fiction, biographies, histories, philosophy, physics, computing a little bit of technology, law, planning, statistics, politics, maths, all sorts really. It is, I think good to be aware of what's going outside of your primary field, both because it gives you a better idea of how the world is operating around you and you never know when something from outside your filed is going to make you look at something inside your field differently.

So I'm quite happy with some of the talks going on in Auckland over the next month. We've got a talk about Bletchley Park. We've got one about Artificial Intelligence, the Robb Lectures gives us 3 talks about learning and development. We've got Lawrence Krauss giving two talks, one on God, one on physics. It almost makes up for missing NerdNite in Wellington tonight.

Saturday, April 21, 2012

Little things big things.

Hot on the tail of yesterday's cool stuff happening with very small stuff happening, I bring you cool stuff happening with very large stuff. there was talk a while back of the asteroid Apophis. It's a large asteroid that will be passing earth by sometime in 2029, close enough that it will be under the satellites that we have in geo-synchronous orbit. In space distances, that's very very close. It's unlikely but if it passes through a certain keyhole of space, then there's a good chance it'll be back to hit us in 2036.

Anyway, one of the suggested solutions for averting this was that we should actually try and steer it into orbit where we could mine the sod. Which, personally, I thought was a grand idea. Now there's people looking at steering other asteroids to earth orbit, bot for study and for mining purposes.What's even cooler is that essentially, the technology already exists. It's just a case of marshalling the resources, it's an engineering problem now, not a scientific one.

And even better, it would give us a much bigger presence in space, something I think we need to do, so we can learn more about what we need to do and give us the infrastructure to start planning manned missions to Mars and the rest of the solar system. Then all we need is a space elevator.

Friday, April 20, 2012

Cool stuff.


So it's Friday. I'm not at work (in Wellington having a cup tea and watch the sea from a deck halfway up a hillside actually). So I'm not thinking about life to seriously today. Funnily enough. So today I'm only going to link to this. It comes under the heading of seriously cool shit*. This act of spectacular-ness involves constructing  new forms of DNA/RNA. DNA and RNA are normally formed on a backbone of ribose. Apparently we've been able to put the 4 letter genetic alphabet on a whole bunch of different sugars for a while now. It still operates like RNA, in that it can attach to and bind other genetic sequences thus acting as an activator or a repressor of other genes. Depending on what sugar you're using you could potentially make it more resistant to the machinery in cells that break RNA down which could quite possibly lead to a whole bunch of better or even brand new treatments for a variety of illnesses or quite frankly, a spectacular tool for the lab to help us figure out what's going on.
Trouble has been creating them. So this bunch of clever people in Cambridge (in the UK, not our one) has jimmied the enzymes that create DNA into creating these new forms of DNA. Which means we've now got the machinery to produce these new forms in useful amounts.

Seriously, if you don't think the fact that we can even contemplate doing stuff like this is awesome then your world is lacking whole levels or marvel that mine currently has.


*I was thinking today, as the plane took off - I wish people from a couple of hundred years ago could see what we can do now.  It would seriously blow some minds. admittedly at the time I was only thinking of the moderately spectacular act of flying.


*I was thinking today, as the plane took off - I wish people from a couple of hundred years ago could see what we can do now.  It would seriously blow some minds. admittedly at the time I was only thinking of the moderately spectacular act of flying.

Thursday, April 19, 2012

Process and review - seriously important.

Okay, so this article has been on my radar (a tab on my browser) for several days now. It's a critique from Gordon Campbell of the public private partnership deals that the government has set up to run some trial schools in Birkenhead. I'm not going to go into the fact that when you take a good long hard look at the charter school process it ends up costing more than a normal school, that's been gone into many times before and should hopefully be obvious to everyone by now.

I'd actually like to try and relate this to one of the processes I use when I'm doing my research. And that is review. Campbell asks a number of questions, relevant questions, like "how will contractual compliance be monitored and enforced" and "What costs for this monitoring-of-compliance role have been incorporated in the contract?" or  "Who will fund the legal costs of pursuing non-compliance". It's that first one that gets me though. When I've got a new set of data (getting seriously sick of waiting for the sequencing data from my last experiment to come back) there are several things I have to do with it. Stitch it all together to make sensible mRNA, figure out which mRNA's are present in greater numbers than they would normally be or maybe figure out which mRNA's are possibly kicking of even greater expression of another gene. It's easy to get something wrong. I'm done it many times before - something seems like a new and interesting result. First thing to do is go back and check it. Then you move onto the next step in the process. Then go back and check it from the very beginning. Continual monitoring and enforcement so to speak. The number of times I've been two or three steps into a process and then figured out there's something wrong in the first step.
The thing is, if you're continually going back and checking, then you're a lot more likely to pick up your mistakes. The fact that charter schools overseas haven't appeared to actually work aside, there should be continual monitoring and enforcement of contracts, so mistakes and problems can be identified, as there should be with any large new venture of this sort.

And as an <indignant> aside, whenever people ask for details about what's being spent or what the contracts involve, I note they tend to get fobbed off with "sorry, that's commercially sensitive". Fine, it's commercially sensitive, but one of the parties in that commercially sensitive partnership is my government and by extension me. it should be a precondition of any commercial entity doing business with the government that they know all of details of the deal will be made public.

Wednesday, April 18, 2012

It's not just about mob rule.



Ngaire Woods: 'Democracy is not just about being elected' – video | Politics | guardian.co.uk

Also, I just watched this. it's a snippet out of a larger debate. Which I'm off to watch now. But please watch this. I don't know the speaker, Ngaire Woods, but the point she makes is an awesome one that I'm not sure a lot of people understand. I say that because of the number of times that I hear people seem to think that democracy is purely based on a majority vote. It's one of the things that I picked up in politics class years and years ago. If democracy is nothing more than a majority vote, then you have, in effect mob rule or a tyranny of the majority. A tyranny of the majority doesn't need to pay attention to it's own laws - it can overule them at any time.

A modicum of optimism

So I've been a little distracted today. I came across this yesterday. The site is called data driven journalism and I have to say, it causes me to feel some optimism. Not a huge amount, but enough to actually register. It's a set of tools, resources and teaching aids to help journalists make sense of the large amounts of data that governments are making available. Think about it for a moment. How good would that be? Journalists with the skills to actually make sense of and find good stories in the masses of data available.

I'm not over the moon, because I don't know of any New Zealand journalist participating in such endeavors. Having said that, it's opened a few doors for me, as in, if you dig for a bit there definitely New Zealand related data stores in there. I'm not entirely sure how deep I'll end up digging or even if I'll find anything interesting but at the least it will be providing me with a few evenings worth of entertainment.

Tuesday, April 17, 2012

It's not just for economics.

I suspect I've been rattling on a bit about why we should be using models of various systems lately. Certainly I think the use of models that are based on abstractions derived from observations of the real world are better than those based on abstractions derived from whatever the modeller thinks the world should be like. Primarily because people the former is likely to be a lot more useful. We know that already though. So today, I'd like to draw your attention to a couple of articles about people modelling pedestrian dynamics.  The Economist gives us the wisdom of crowds  and Slate gives us sidewalk science, both worth reading. especially the Economist article.

It nicely demonstrates the perils of not paying attention to your assumptions. A lot of the early attempts at modelling crowds of pedestrians assumed that each pedestrian moved as a individual particle. Except, if you think about it, that's not always the case - people moving in groups (like tourists, groups of friends or even couples out for a stroll) don't act independently, they maintain a group cohesion which behaves significantly different from a number of individual particles.
Sidewalk science on the other hand takes a more story orientated approach, but illustrates again why it's good to pay attention to the system you're trying to model - different countries, even different cities have different behaviour. In individual cities people tend to break either left or right when two pedestrians approach each other head on as well as having pedestrians that move at different speeds in different environments (Germans move slower than Indians when in a crowd apparently) - models aren't necessarily immediately transferable to all situations.

So? Well, town planning for one can use this sort of thing. Architects trying to design escape routes from buildings. Stadium planners. There's all sorts of uses for this sort of modelling. The actual models are getting better but will always be dependent on the validity of the assumptions used to make them.

Monday, April 16, 2012

A little confused.

Or at least, I'm beginning to be. I suspect I need to start compiling a spread sheet or some such. It may be personal bias or possibly I'm just reading the wrong media (yes, I read the herald I know I'm reading the wrong  media). A lot of the stories I recall reading though are like this. In as much as every time someone says we need to spend more on something, apparently the government has already increased spending on it over the last couple of years. Without raising taxes. Without major cuts (some planned but no major ones I can recall having taken effect yet). Without borrowing more.

Sunday, April 15, 2012

The bit that was missing.

You know that uneasy feeling you get when you read an argument that sounds coherent but just doesn't sit quite right? I had that last week when I was reading this piece in the Listener in the tea room last week.

It sounds reasonable. Makhlouf sounds like a reasonable chap, just wanting to fix our education system. The first thing that made me wonder, while I was at the table, was why do treasury think out system is broken? As far as I am aware, we have one of the best education systems in the world. Or at least in the OECD. This afternoon, I found a column by a chap called Gorden Dryden that pointed out a few more things. The comparisons to China were ... flawed at best. In the PISA comparisons used, Only two disproportionately affluent cities from China were used, combine that with China's one child policy and you end up with a very high ratio of students to teachers. So using that comparison to compare our education system for a wide variety of socioeconomic groups seems at best pointless. The PISA comparison we should be using, is against Finland - being the current all round number 1.
And the way I'm reading it, what he is suggesting is that the primary thing we should be doing is rewarding better teaching. which sounds awfully like performance based pay. Which, if Dryden is to believed (and from what little digging I've done, it seems to be true)  is the complete opposite of what the Fins are doing - they've gone the route of more teachers focusing more on students needs and less on national standardized tests.

Which makes it sound rather horribly like someone from Treasury who is cherry picking their data and subverting even that cherry picked data to suggest a course of action with no foundation in reality. The evidence suggests that the best course of action is the opposite of what is being suggested. Surely you can't get to head treasury without at least a basic understanding of statistics? A blinkered view of the world at best, though it sounds horribly disingenuous.

Friday, April 13, 2012

The Mathmatics of Occupy maybe?

Crowd sourcing. Kickstarter. We should all be familiar with these by now - getting funding for (usually) art projects directly from numerous small donations rather than from a single large funding source. Rockethub is a similar venture (though not the only one) that is focusing on science projects, small ones obviously, usually ones that allow one or two people to work on projects for a month or so.

There's some interesting projects on these sites, if I had some money (any money really) I'd be donating to them. To this project in particular. It appears to be in the same vein as behavioural economics, in as much as the chap, Lee Worden, from McMasters University in Canada, is looking to learn more about how Occupy makes it decisions.
working to develop quantitative models and simulations of consensus decision-making process. The purpose of this project is not to explain, capture, or predict the endless surprises and delights of human creativity and communication - that would be a fool's errand, and an insult to humans. Rather, it is to try to capture particular aspects of the process that seem important,
In one sense we already know how Occupy makes it bigger decisions, primarily through the general assemblies where ideas get debated and voted upon. Modelling this process isn't going to let us predict what decisions Occupy is going to make, what it could possibly do is inform Occupy as to how well the system is working. It could help in determining when tasks are best split off to smaller groups that then report back to the general assemblies. It could help in refining processes to prevent individuals from leading (in a particular direction) the general assemblies. It could help prevent minority opinions from being marginalized.

This is not to say that it will. It could though. And at the very least, we would have some models, based on real world assumptions of how direct participatory democracy works. Which could help us in all sorts of ways. Personally, I'm of the opinion that direct participatory democracy is only going to work up to a certain level of organisation, when the scale of the decisions will require some form of representative democracy (I could be wrong). Models like this informing us about how smaller groups make decisions could help us work this local connected form of democracy to a the larger scale/countrywide democracy. Useful. Makes me rue the fact that I have no pingas to donate to this project. There's nothing to stop you from donating it though.

Thursday, April 12, 2012

Reality based markets.

Here's a point of view that's illustrates nicely, I think, several of the points that I've been trying to make over the past few weeks - nicely applied to the real world. We have Professor of Marketing and Statistics at Penn State that is advocating the prying open of large businesses to get accurate information and the use that information to model some economics that actually work. He rightly points out that where we've had mathematicians and scientists making models before, but with companies sitting on their data as tightly as possible, these models have been flawed from the beginning because they haven't had accurate real world data on which to build. Economics is a horribly complex system (not as bad as the weather or biology, but still pretty bad) and we're a long way from being able to model it perfectly but with the right information used and shared properly we can get a pretty good idea about how to handle certain types of situations and avoid some of the bigger risks.
This is the thing about models. They're not perfect. If they were it wouldn't be a model, it would be a replica. they help us understand what is actually going on rather than what we want to think is happening though. And if they're not based on realistic assumptions, they will fail.

Wednesday, April 11, 2012

Things are looking up.

So occasionally I mention food. Normally when I'm horribly disappointed by it. Like most of the (few) times I've been to Kiwiana for fish and chips. Seriously, don't go there. However, in Kingsland town, there's a new kid on the block. We have a new Thai restaurant/takeaways, called Bangkok Cafe. It's worth it.
I'm aware that the days of good sub-$10 meals is long past. In Kingsland at least, there has been a dearth of even approaching cheap, good food though. Most of the restaurants round here are at the $20 for a mains mark. Which in the case of Canton, is fully justifiable, but for most of the others not so much. Handmade, I suppose does a decent burger, but for a burger, it's still in the pricey range. Kiwiana - cheapish, but not so good.
Bangkok though, there's a good selection, even for the wegetarians, it's $13 for mains with rice. $11.50 for the mains with noodles. And the food is nice. As in really nice. Could possibly do with a tad more chili, but they've got their flavors spot on quite frankly. Well worth a trip, I'd rather like people to wander along and try it. It's in a spot that hasn't traditionally done well for the various cafes that have been there previously and this is one food stop in Kingsland I don't want to see disappear.

Friday, April 6, 2012

oooo, look at that.

Well this is depressing. I'm writing a post about a bus shelter ad.
Just drove past a bus shelter with a jetstar ad, something like "350,000" domestic customers have flown into queenstown in some specific time period. 65% of those for $99 or less". Nice, he said, nice well constructed use of statistics. It has the base population delimited in magnitude and time. And a simple, relevant statistic that actually applies to the population. Whether it's true or not, I don't know, but it deserves minor plaudits for being coherent at least.
It's a bit sad, this time of day, going on about an ad of all things and it's laudable use of statistics. Though I do think it's probably sadder still that I find the ad refreshing.

Thursday, April 5, 2012

Old story

So it's a week or two old, but this story sort of relates to what I was talking about yesterday. If you will, remember back to Gerry Brownlee's recent attempt to make an ass of himself by having a crack at Finland. Michael Edmonds disassembles Brownlee's facts. Or a couple of them at least.
One of the things that I would very much like the people running this country (it is a reasonably important job after all) to have, is a grasp on the relevant facts. I don't care what party they're from. They're all entitled to their own opinions, but not their own facts. When politicians come out with random cracks* like Brownlee did, it brings the profession as a whole into disrepute. When you compare the facts presented with the actual facts and a little to often, the presented facts are found wanting - that erodes trust.

*No, I don't seriously give credence to the idea that he was trying to be funny. It's the same pathetic excuse that is used time and time again when someone puts their foot in it. i.e the "lighten up it's was just a joke" defence. While I suspect most politicians have a (alarmingly boring) sense of humour, they shouldn't be attempting to use it in the house.

Wednesday, April 4, 2012

Who to trust?

Which source to trust? Why one and not another?

I've been thinking about the best way to illustrate this since yesterdays post. it might take most of the rest of the week.I figure the best place to start is with two different sources, one of which I trust, the other I don't.
For the first one, I'm going to use Nobel Intent from Ars Technica. This is not a day to day news site, it doesn't tell you about what is going on in your local corner of the world. It's a science and technology site that brings together interesting stories from science, technology, business, law, politics and any other topic that the editors thing is useful or interesting to their readership. why do I trust it? (this isn't an exhaustive list by the way)
  • Ars technica have an advantage straight off the bat, namely that they occasionally publish in areas that I am familiar with. So I can tell if they've got something horrendously wrong in those areas. The fact they they very rarely get anything wrong indicates to me that their editors are hiring good people and their writers know the subjects they write about, taking pains to get things right. This immediately means that I extend a certain amount of trust in the areas that they write about that I'm not an expert in.
  • They are pretty upfront in where their biases are as evidenced by their stance on SOPA and PIPA. That they acknowledge those biases is also a point in their favor as it allows me to read their articles with those biases in mind.
  • They also publish their sources. take a look down the bottom of this article on the findings a probe that was sent to Mercury. There's line that looks like "Sciencexpress, 2012. DOIs: 10.1126/science.1218809, 10.1126/science.1218805  (About DOIs)." This is a link to the source paper that the article is about. If I think something sounds a little odd, I go go and check over the orginal source to see if what the source and the journalist are saying matches up. I have done this in the past, not with every article, usually just the interesting ones, and the journalists usually get it pretty bang on. This also creates trust.
This is not to say that I'm going to continue to trust them for ever and ever. I'll continue occasionally checking up on them, comparing them to original content or other reliable sources. If they start to waiver, I'll lose my trust in them. What about untrustworthy sources? Lets try the NZ herald. It's a different type of publication, one that is attempting to bring me political, social and business news from my local area, my country and the world. Why don't I trust this source (again, not an exhaustive list, I could go on. and on. and on. but I won't).
  • I happen to be reasonably familiar with the use of statistics. It is a rare event to be able to read the herald and not find the reporting of at least one story to be skewed or completely due to an abuse of statistics.
  • In some cases stories are made up with little idea of what is being said. The "it's all in the stars" story last week about how you're more likely to win lotto if you are a taurus is a fine example. It's not a news story, it's not statistically sound. At best it is a fluffy piece of nonsense entertainment. And when entertainment is mixed with news, rather than entertainment becoming news, the news becomes suspect.
  • The herald claims to be neutral. However when you go and dig out the original sources (they are never linked to), what the article and the source say, quite often differs. This makes articles seem like editorials with unacknowledged biases - something that does not engender trust. 
The salient points here are track record, acknowledgement of bias, soundness of reasoning and how well the stories are reported compared to the original items. One doesn't check the original source for everything, but it never hurts to take a sample of what you read each day or each week and see how well it matches up with reality. One of these sources does all that. One only claims to, but when you check, it fails miserably. If they get it wrong in areas with which you are familiar how can you trust them in those areas with which you are not familiar?

Tuesday, April 3, 2012

Knowing where to look.

One of the assumptions that I suspect gets made about scientists in general is that we're really clever. I say this because over there years I've had occasion to have to convince several quite intelligent people that I'm not particularly smarter than they are. There is a modicum of smarts in the industry yes, and there's a number of really really smart people floating around, that I will concede, but the major difference between myself and large numbers of the general public is that I know how and where to look for information. I'm not talking about how to critically take in that information, (though that is part of the training you receive eventually acquire in the sciences) that's the identification of cognitive biases I talked about last week.
I'm talking here primarily about becoming familiar with various sources of information. it springs to mind from this post at WEIT. As part of a larger (inane, imho) spiel, some chap at the NYT is having a go at Richard Dawkins saying that you "actually cite chapter and verse" for global warming studies. Which immediately brought to mind the kerfluffle when Dawkins didn't get the full name of Origin of the Species word perfect. It might seem like a tenuous link to this train of thought, though I think it's valid - obviously. Maybe not 30 years ago, but today definitely - one of the things that scientists have going for them is that we know where to look for information. We don't need to remember each and every fact we've read, we just need to know how to find it again. Over a long period of time we build up a library of sources that we trust and we figure out how to evaluate new sources of information. Very little that I do, I suspect, couldn't be done by most other people who are sufficiently interested, with sufficient training.