The 'New' Media...

I have this odd tradition in that, whenever I'm flying, I read The Economist. I say that it's odd because I'd actually love to read the magazine weekly, but a) the subscription fee tends to be on the high side and b) I barely read through the couple of monthly mags that I now receive, let alone being able to find the time to digest a weekly.

My primary draw to the magazine, of all things, is the quality of its writing rather than the depth of its content (which I also enjoy, incidentally). As a scientist, I'm required to write quite a bit: obviously for publication, but also regularly in the form of professional emails and documents. As I and other more esteemed professional writers have long advocated, good writing comes both from practice1 and from comprehensive reading. Reading good writing begets improved ideas about proper style in one's one work.

It's interesting then that the current issue of The Economist (9-15 July 2011) has a special section about the future of journalism. Unless you've been living under a rock for the past couple of decades2, you're probably aware that the rapid growth of the internet in the West has put a massive squeeze on the circulation and revenues of print publications (interestingly, newspaper 'circ' numbers continue to rise in Asia). Honestly, it's difficult to compete: the net is a) 'free', b) fast, c) searchable, and d) global. Of course there's been talk of monetizing specific popular newspapers by placing them behind 'pay-walls', but I can't imagine how that will be successful in the long term.

To think that media on the net is simply a digital form of what was in print is a mistake, however. A large part of the very recent explosion of internet activity has come from participatory media (or Web 2.0 as people used to refer to it). This takes the form of both direct dispatches from 'the front' on sites like Twitter and Facebook (and what the hell, I'll add Google+), but also the ability to comment on particular news items and engage in discussions as the news is happening. As the article in The Economist notes, this 'democratizing' of the news is an obvious benefit to any liberal (small 'l' here for my fellow Canadians).

The author's discussion of the subject, however, notes 2 drawbacks to this new form of media: Firstly, that 'bloggers' and 'crowdsourced' journalism are difficult to hold accountable in the wake of allegations of corruption, libel, or crime. It's difficult to know who is legitimate (this also applies to traditional new outlets as well, obviously). Secondly, despite the breadth in available content, people tend to only seek out that content which echoes back their own beliefs. Internet news, they content, allows for a much deeper partisanship than a local newspaper that has to try to sell to everyone and thus be more balanced in its reporting (however, as the authors themselves note, there's a move towards strong partisanship in traditional American media as well).

I'd like to add a third drawback: The quality of online writing tends to suck... bad. The effort to capture the immediacy of current events without the ability to digest all of the information leads to a lot of poorly written, desperately-in-need-of-editing, garbage. This doesn't only apply to news - I've noticed the same trends on tech and enthusisast websites as well. I'm concerned that a generation of people who've only received their information from online sources won't even be able to recognize and appreciate what a well-written piece of work can do for overall comprehension of the particular subject matter.

Oh well. Some would argue that if that's what the market wants, who are we to criticize what it gets? Hopefully The Economist will never be written by the kind of people who write for Gawker media.   


1The primary reason to have such a blog as far as I'm concerned.

2Note to self: Tired cliche. (See what I mean!?)

Book Club: The Tipping Point...

I'd been meaning to read one of Malcolm Gladwell's books for a while now, ever since a person whose opinions I respect reccomended his work a while back. As far as I can tell Gladwell uses quirky, often idiosyncratic situations in order to draw attention to deeper underlying 'truths' about the strange intricacies of human cognition and how human social interactions work. In this way, his methods are somewhat analogous to the social sciences: coming up with a model of how or why a particular process works, and then determining how well the model fits the data.

Under such a rubric however, I found Gladwell's The Tipping Point (TP; 2000; Little Brown) to be more illustrative of some the pitfalls associated with attempting to apply overly vague and or complicated models to what are quite possibly largely stochastic processes, rather than a rigorous analysis. But I'll get back to that in a moment.

In TP, Gladwell attempts to explain the origin of social epidemics: why certain things become immensely popular while others fizzle out. This explanation involves three factors: 1) The 'law of the few', which says that epidemics are generated by rare people with special social gifts; 2) The 'stickiness factor', which determines whether particular ideas catch on and become memorable; and 3) The 'power of context', which says that epidemics are sensitive to the particular environmental (social, political, etc.) circumstances in which they begin.

Wow. Not to be overly dismissive or anything, but can you imagine a social epidemic that simultaneously began among unconnected people, didn't involve anything particularly memorable or interesting, and would have been just as successful in any other place or time? Gladwell writes very well and certainly concocts an interesting narrative, but these obvious 'factors' are about as useful as saying that airplanes depend on the 'law of not touching the ground for a extended periods'. Though, to be fair, they shouldn't be dismissed if they have predictive power in terms of future epidemics.

This predictive power is where the book is both most interesting and yet also most frustrating. The majority of TP involves the author laying his 3 factor model on top of previous epidemics and trying to explain them in this context. The law of the few, for example, can be broken into three crucial groups of people: Connectors - people who link groups together and provide a conduit for information; Mavens - people who must be 'in the know' about particular subjects and are thus up on current trends before everyone else; and Salesmen - charismatic folks who sell those trends to the public. But this taxonomy is blurred in Gladwell's own examples, where people act as both Connectors and Salesmen (wouldn't charisma be a major asset to both?) - or even Mavens. Distinctions are only as useful as their explanatory power: groups can be binned into innumerable different categories depending on any arbitrary criteria - but do those separate bins have meaning?

Stickiness and context are even more vague. It's easy to provide a detailed explanation of why Sesame Street was so successful after the fact, but can it be used to predict the next giant children's educational property? According to Gladwell, the developers of Blue's Clues used their experience with the aforementioned show to do just that. However, the vast majority of the cases in the book are more of the 'golly gee, ain't that cool how they did that' variety, rather than legitimately instructive about how to create the next social epidemic.

Allow me to take a step back and set the context of my issues with The Tipping Point: Certainly, studying how New York dropped its crime rate during the late 80s, or how Airwalk created a branding phenomenon is instructive in terms of planning similar ventures. They also make for fascinating reads. But the focus of the book isn't on how people can learn from these ventures and apply them successfully; it's on how small decisions can lead to the 'next big thing'. In this respect, Gladwell does offer some useful, predictive perspectives, particularly in his 'law of the few'. Understanding where to start epidemics by understanding social connections is obviously crucial (anyone who's been exposed to viral marketing knows that this has been taken to heart). But I think that it would be much more significant to have at least similar understandings of what defines stickiness and context1. Some of this is attempted in the book, but in a method that I find rather unsatisfying.

As I re-read what I've written, I realize that I'm probably coming off as more harsh than I intended - The Tipping Point is an interesting read, and some of the analyses of the social epidemics discussed within are impressive. It's simply that there's an air of smugness that accompanies the kind of books you associate with the self-help section of the bookstore to the whole affair. Categorizing the factors that influence whether things become popular in the context of select examples is only really useful if those categories are meaningful in their definition2 - stickiness and context strike me as being difficult to pin down and implicitly mutually interrelated, no?

1Obviously, the power to predict these two factors is pretty poor, or else we'd see far fewer products and attempted trends flop. Gladwell does raise an interesting point about both stickiness and context, however, in his brief examination of how certain trends cause a desensitizing effect. For example, as the volume of overall advertising increases, people tend to work harder to ignore all advertising, and thus the returns garnered by any particular firm for dollars spent on advertising diminish in a value per dollar sense. This observation isn't particularly novel though, and is well known in my own field as 'negative frequency dependent selection', indicating that certain traits are more valuable if they are rare. 

2Jared Diamond's Collapse is an example of a book that uses case studies of failed societies in order to define a taxonomy of features that are responsible for said failures. However, his categories are better defined and measurable: the presence of hostile neighbors, climate change, or collapse of essential trading partners. It's fairly straightforward to see how the failure of the Mayan culture due to overpopulation and deforestation has clear parallels and implications for modern societies (whether or not the solutions to such problems are themselves straightforward or politically expedient). While societal collapses may be explicable by more clearly definable factors, it's arguable that many social epidemics like beanie babies or, that whole soother/pacifier craze among girls 15 years back, are based on the workings of unpredictable 'right-place-right-time' circumstances.

A Return to Form...

A long time ago in a graduate school far far away I used to loved immersing myself in science. Of course everyone is somewhat immersed in the vagaries of the particular projects in which they're involved, but I'm talking about much more than that; I mean thinking, reading, and talking about the broader themes of science. Where is the field going? What is the relationship between science and society? What is the role of individual scientists in the face of widespread superstition? You know, things like that. I read a ton of random science books and blogged regularly - in fact, some of the major themes of my Ph.D. research came out of random details that I read in older books that got me thinking about their modern applications.

Recently, I've been having more of those sorts of moments again. Conversations with fellow postdocs about subjects outside of the daily grind, big-picture speculations, even random pie-in-the-sky conversations about what we would do if we ever had our own labs. These moments may seem minor when you read about them now, but I must admit that they've had the effect of kindling a long-dormant 'fire' in my 'soul'1. I mean, think about any scientific research project: You spend a lot of time working away at answering a question, expecting that all of your work will eventually pay off. Some of these projects take years, and the way to keep the excitement up is to visualize the end goal. Can't it be the same on larger scales?

There's something to be said about appreciating journeys and not destinations, but ultimately there has to be a destination or else there can't really be a journey, can there? At that point it seems like just aimless wandering, and I think that I'm getting too old for that...

1My metaphorical 'soul' that is.   

Adventures in Economics...

Alas, life continues to be extremely busy. I'm still not sure where I'll end up in the fall yet, and I'm still looking into to opportunities. I'm also trying to finish up my work on a manuscript in my current lab so that I have at least 1 first authorship paper for the two years that I put in here (prospects are looking pretty good, so let's keep out fingers crossed).

Earlier this year I blogged about how I've been getting interested the field of economics. In some part this is because some understanding of economics is crucial to any appreciation of how the world 'works'. But I also have a more personal interest resulting from the close relationship between economics and evolutionary theory: both fields are about managing limited resources and risk (the latter being active in economics and passive in evolutionary biology). Furthermore, both fields employ many of the same modelling approaches, including game theory, benefit maximization, engineering compromises, etc.

However, in my limited understanding of the field, I've always been a bit puzzled by how insistent some of my more libertarian friends are on the applicability of rational choice theory - essentially that actors within an economic model behave as if they're making careful cost/benefit analyses every time they're allocating any type of resource. Personal experience (and a lot of cognitive research) seems to argue that we don't always make the best decisions for a variety of different reasons. Yet another issue that bothers me - likely because of my biological science roots - is the lack of attention paid to 'externalities' created by economic activity. Externalities are the costs of an activity that are not reflected in their price; for example, if a now-closed mine is found to be leaking toxic materials into the groundwater requiring 100 million dollar tax-payer funded clean-up, that's a negative externality.

Aside from personal and hard-wired cognitive biases leading to irrational decision making on the part of economic actors, another major cause of malfunction in the free market is incomplete information said decisions. The market only works properly (is 'efficient') if the cost of goods/wages/etc. predictably reflects the value of the things themselves. If economics all boils down to managing risk, then modelling such risk requires that the appropriate model inputs be known. As a specific example, a health insurance company can only properly determine the cost of premiums if it can accurately determine the risk associated with issuing a policy. If everyone lies about prior medical conditions and current health, then the company will underestimate risks and likely not be profitable1.

On the subject of these informational asymmetries, as they're known, I've been reading some books by Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel recipient Joseph Stiglitz, who seems to be one of the few 'mainstream' economists who isn't a Chicago-school libertarian. Specifically in Freefall: America, Free Markets, and the Sinking of the World Economy2, Stiglitz explains that he won his Nobel Prize for showing that both informational asymmetries and externalities are pervasive in modern economies. According to the author himself: 

Whenever there are “externalities”—where the actions of an individual have impacts on others for which they do not pay or for which they are not compensated—markets will not work well. But recent research has shown that these externalities are pervasive, whenever there is imperfect information or imperfect risk markets—that is always.

The real debate today is about finding the right balance between the market and government. Both are needed. They can each complement each other. This balance will differ from time to time and place to place.

I'm certain that such statements will grate on some people, but they seem, at least at face value, to match my expectations of the functioning of free-market competition based on another form of competition that I'm more familiar with - evolutionary theory. A few salient observations from biological competition may have implication in market models:

1. 'Survival of the fittest' does not always mean survival of the 'best' in some absolute sense (if such a sense may ever be defined). In fact, it often leads to survival of the 'luckiest'. The fittest organism may have one particular, a priori unpredictable trait that causes it to survive when other species fail. A company may succeed for all kinds of ephemeral reasons - not necessarily because they offer the best product/service/etc.

2. All species ultimately go extinct. This seems rather obvious except when it's used as an analogue to capitalism. There's a strange undercurrent to the whole market economy thing where people will value competition but not want to deal with the consequences of many, many inevitable failures.

The somewhat strange, paradoxical way of thinking characterized in point #2 is often referred to as 'private gains, socialized losses', and manifests itself as government bailouts of insolvent corporations in order to save jobs (or the broader economy). One of the major points in Stiglitz's and other books that I've read is that it generates an unreasonable amount of 'moral hazard' to have a system wherein the government actively seeks to deregulate the financial sector, for example, while being unwilling to bear the consequences of said deregulation.

This brings us back to Stiglitz's comments above and the argument that there is no utopian system of free market efficiency nor of perfect governmental control. In addition to being inefficient and prone to falling prey to perverse incentives, it is doubtful that the private sector can be expected to supply many necessary 'public goods'. On the flipside, it's impossible for a government controlled economy to obtain all of the information it requires to properly allocate resources: communism sucks. Free market economies are the best way to perform said allocation, despite our knowledge of the inherent capacities for market failure. According to economists like Stiglitz, the role of government should then be to make laws minimizing the likelihood of market failure and inefficient allocation (e.g., monopolies) - as well as providing public goods.

The question as to whether governments are doing this efficiently is a completely different question, however. There's a certain amount of 'package deal' mentality that comes with many liberal or conservative economic opinions - and poor decisions are inevitably made on both sides. From my perspective, it would be nice to think that more of our political decisions could be made based on empirical evidence regarding what works and how people actually behave. Given the complexities of running an actual country, I'll be the first to admit that perhaps I'm sounding naïve... Oh well.       

1As a somewhat random aside, there's a really interesting classical paper called 'The Market for Lemons: Quality Uncertainty and the Market Mechanism' that's really worth a look if you want to understand risks associated with insurance. It also goes a long way to explaining why it's very difficult to have a non-universal government-operated health-care system (this is useful if, like me, you're often puzzled about America's reticence to such a system). 

2The actual subject of Stiglitz book is the recent housing market crash and the ensuing recession. I covered that topic a bit in a previous post, and figured it would be more interesting to blog about something marginally different here.

The 'lost month' and a big change...

Apologies for the lack of updates around here lately, but in all honesty, the past few months have been some of the most difficult periods I've had in recent memory. It's unlikely that anyone would argue that postdocing isn't a tough business. Many of the postdocs that I know tend to work hours that probably stretch the limit of what reasonable people would consider having any kind of life beyond work. Consequently when, as a postdoc, I say that I've been busy, I mean that I've been really busy.

The major cause of my busyness is that I agreed to contribute a short chapter to an upcoming collected volume on various evolutionary topics. It's obviously an excellent career-building opportunity as well as a chance to produce a professional piece of writing as a sole author. However, it was not part of my 'primary research', so I worked on it during my free time. I'm a little worried that given little time I've had to devote to it lately, it's not going to be up to my typical standards. At least I'll get some reviewer feedback along with the opportunity to make modifications before the final copy goes through.

Another component to my general busyness is a bit more major: After two years of working my ass off, I've decided to leave my current postdoc and continue my career-building elsewhere. Obviously the specific details as to why I've come to my decision are best kept between myself, my PI, and potential future employers, but I can say that this decision wasn't reached lightly. I struggled with whether my current lab was suiting my objectives, and after much deliberation and discussion, I came to the conclusion that it was not. This is a really tough, often stressful, and underpaid job - the absolute minimum one can expect is to actually enjoy what you're doing.

Thus, I am now on the market for a new position. 

One of the major 'stressors' for me in terms of reaching this decision had to do with the strange predicament in which we postdocs find ourselves. There are many of us, many of which are very, very talented. Every time there's an opening for an academic position somewhere, the number of applicants can often be staggering. It's tough enough to compete for a job in the field without the stain of having left a postdoc - especially while you were externally funded.

Honestly, I was so discouraged by the situation that I figured that my only recourse would be to 'bow out' of academia and look for a career in industry. There's nothing wrong with working in industry, of course, but I'd be lying if I said that my dream wasn't to have my own lab some day. Thankfully(?) some encouragement from former mentors and advisors have led me to believe that my chances at having a successful career have not been obliterated by this unfortunate event. My only hope is that I leave my lab on amicable terms, which seems to be the case, for which I'm grateful. 

Oh well, I'm extremely excited by the prospect of moving to a new lab where I can work on a project of my own in a field that I love so stay tuned1.

1My plan is to keep up with regular blogging moving forward, no matter what the future holds!

P.S. As a sign, I saw an awesome 'double rainbow' last week: Perhaps fate (which I don't believe in) is telling me to 'hang in there!'