Monday, March 26, 2012

Immunology under the microscope: knowing what is known

The previous post in this series can be found here.

As a passionate believer in science, I have had many debates over the dinner table or at the bar with those who consider scientific research either unimportant or ineffective. The argument often comes back to “Well, science doesn't know everything!” To say this inherently misunderstands science and scientific research. Firstly, science doesn’t know anything; science doesn’t think or know - it simply is. Science is just what exists, it is everything. Everything must be explicable otherwise it couldn’t exist, even if understanding it is beyond the capability of humanity or any other intelligence. If God exists, He presumably understands Himself and so is explicable, to Him at least. I suppose that when people make the claim that “science doesn’t know everything”, they really mean “scientists don’t know everything”.  This statement is obviously true, we don’t know everything, but it never fails to surprise me how little people often assume we do know. I don’t expect people to understand science in great detail, but it is a great shame that there is so much ignorance of the sheer amount of knowledge that we’ve gathered as a race. I have, in the past, been comprehensively informed that “we still don’t even know how genes work!”, which is very surprising news considering I thought I could explain in atomic detail how a gene is transcribed, translated and expressed as a protein – that is to say: how it works. An unfortunate mistake that many people often make is to confuse the statements “I don’t know this” and “nobody knows this”.

Perhaps this is to be expected; seen from the outside, science probably appears very sporadic: there are long periods where nothing is being discovered and then suddenly there’s a big finding and everyone gets very briefly excited before everything dies down again. This, however, is simply a product of the sensationalist way that it is often portrayed in the mainstream media, in which everything deemed too complicated or insignificant is not reported and things that do make the grade are often over-hyped. Following science in this way is a bit like following literature only by watching Hollywood adaptations of major novels. In fact, science progresses in tiny steps that can seem insignificant on their own but contribute to the field as a whole. Immunology is no different.

Friday, March 23, 2012

The scaling of time: ChronoZoom

One of the conversations that is starting to develop here at the blog relates to the concept of scale. In a recent comment, Michelle wrote about an application that aids with the visualisation of time-scales (although the application's designers see it as more of an interactive teaching tool).

Seeing as "Scale of the Universe" is so popular on the internet, I expect that this application, ChronoZoom, should become equally popular one day soon. Scale of the Universe allows you to zoom in and out of distance scales and it shows you the relevant physics, chemistry or biology at each scale. ChronoZoom kind of does the same for time. However it is quite a bit more detailed and much, much more ambitious as a project. Watch the video above, read a recent article on the project, and play with the application itself.

Michelle wrote in her comment that she finds visualising time-scales through this application more difficult than the equivalent for spatial-scales. This is not surprising. We encounter space with our eyes every day, so visualising space is already second nature. Zooming in and out is also something we do naturally every day. Extending this to a need to zoom in and out on a greater range of scales shouldn't be too hard. However, no matter what happens in life, we always proceed forwards in time at a rate of one second per second and we definitely don't "see" time. Despite all of that I think it is also true that the Scale of the Universe application was specifically designed with the aim to give a sense of perspective. Whereas, ChronoZoom has been designed more as a way of organising information. The sense of perspective it can give seems to be much more of a bonus than a feature.

One interesting observation I can make (and have already to a certain degree in this comment) is how in the early stages of the universe, time scales and length scales were intricately related. This relationship arises from two different mechanisms. The first, and more obvious, is the expansion of space with time. As time progresses, things in the universe get further apart. Therefore the relevant distance scales get larger and larger as well. The second mechanism is the growth of structures in the universe. Structures grow hierarchically. This means small things grow first, then larger, then larger. This is actually a distinct effect from the expansion of the universe, but it's aesthetic implication is the same; earlier times mean smaller scales.

Of course most of the ChronoZoom application is devoted to later times. Something that struck me quite a bit when I first realised it is that the Earth is about a third of the age of the universe. Whether this means that the Earth is very old, or the universe is very young depends on your perspective, but whichever perspective you hold, when it comes to time, the universe and the Earth exist on similar scales. You can see this really clearly in ChronoZoom. What is also quite striking, but already appreciated by many people is just how young humankind is. If you go to ChronoZoom and click on "humanity" right at the top of the page you are given quite a sense of perspective regarding how young we are as you watch it take forever to zoom in on our relevant time.

For both those old dudes, Earth and universe, we humans really are just a flicker.

Monday, March 19, 2012

Veritasium: another element in science videos

For those of you following us on facebook, you will already be aware of the YouTube channel Veritasium because I posted on facebook last week about this awesome video. For those of you who aren't following us on facebook, why not? You're missing out!

Now, before I get on to the subject of this post, let me first say I really, really like Veritasium's videos. I thought minutephysics was the pinnacle of awesome when it came to internet science videos, but I have to admit that Veritasium challenges minutephysics for that crown.

Here is why. Minutephysics is awesome and entertaining (and I'm still a huge fan), but take a minute and think, how much does it really improve your understanding? It definitely entertains and therefore sparks an interest in the subject it is describing, but part of the limitations the author imposes upon himself restricts just how much understanding he can actually teach. I think (though James might wish to chime in with a different opinion) that people who watch Veritasium's videos will come away having actually learned more.

Here is why. Derek Muller, the man behind Veritasium, has actually done research (a doctorate, in fact) into what does and doesn't work when teaching science in videos.  Most importantly he found that if you first of all explain the common misconceptions and why they are wrong, that people learn better. He actually found in his research that if people with misconceptions watched a video that only explained the truth they often come away even more sure of their misconceptions. Surprising, I know. I really encourage everyone reading to watch the video where he explains all of this - for me at least it made for fascinating viewing.

Has Veritasium made a mistake?

Anyway, here is a quick quiz. I think Veritasium has made a mistake in the video above. Can anyone spot it? Am I just being pedantic about this particular "mistake"? (I don't think so, I think it is quite important) Or is it me making the mistake and everything in the video is correct?

I know when I've tried to provoke comments in my posts before nobody has replied, so my hopes aren't high, but I'd really like to know whether other people can spot what I think I'm spotting.

Twitter: @just_shaun

Friday, March 16, 2012

At the neutrino OPERA, the fat lady is warming up.

If you weren't living under a rock last year, you will remember the neutrinos that seemed to be going faster than the speed of light. Everybody and their cat wrote about it, including me (though my cat only had editorial supervision of the final manuscript).

Earlier this year, OPERA (the experiment that apparently saw faster than light neutrinos) updated their press release informing us that they had spotted a few mistakes in their experiment. The mistake was an unfortunate one, but wasn't in OPERA's actual analysis or method itself. The mistake was basically a piece of faulty hardware. Normally you don't need to know things at a neutrino detector to within nanoseconds, so this fault hadn't been noticed before.

I don't hold OPERA to be at fault for going public with their initial findings. They had checked everything they could think of and still had this seemingly crazy result. If the result stood up to scrutiny and was repeated by other experiments, then they'd made one of the biggest discoveries for a long time, if not ever. Once they had checked everything they could think of, any lingering mistake is far more likely to be found by a group trying to repeat their experiment, than by someone in their collaboration.

I didn't write about this update, mostly because I was travelling, but also because nothing conclusive had yet occurred. OPERA made a mistake, fine, but it wasn't proven yet that this mistake was responsible (of course everyone, myself included, strongly suspected that it was – but we all strongly expected that some mistake had been made, and said so, from the start!).

Today, something genuinely new can be added to the story, due to ICARUS, another neutrino experiment. Ironically, both ICARUS and OPERA are at the same laboratory, at Gran Sasso, and in fact measure exactly the same neutrino beam from CERN. The new piece of the story is that ICARUS has measured the speed of these neutrinos and found them to be consistent with the speed of light, and crucially, inconsistent with the results from OPERA.

Here is the crucial figure from their article:

The neutrinos' arrival at ICARUS and OPERA. δt=0 corresponds to when light would arrive.

The purple bars show the scatter in time when neutrinos were measured to arrive at ICARUS. δt=0 corresponds to when light would arrive. Clearly the ICARUS neutrinos are arriving at the same time as light would. The bars on the right are what OPERA's previous measurements indicated.

So, given that (a) a lot of other (measured!) things in physics would be very difficult to reconcile with something going faster than light (b) We know OPERA have made a mistake that they need to fix and (c) ICARUS is measuring the exact same neutrino beam at the exactly same place and finding the neutrinos to be travelling at light speed, it seems that, in this opera, "the fat lady" might not just be warming up, but could very well be in the final stages of her aria.

But, as Matt Strassler pointed out in his own blog, there is a silver lining to all of this. 
The experimental particle physics community has learned how to make long-range distance and timing measurements that are more precise and more accurate than were ever possible before. Don’t be surprised if this knowledge turns out to be useful, in some unexpected way, in future experiments.
In fact, I'm sure this precision timing will be extremely useful in increasing the bit-rate of the neutrinos that have very recently been used to send digital messages through the ground, something that is impossible with any other known form of communication.

As a final passing comment I want to add that I am still in awe of the fact that both ICARUS and OPERA could make such a precise measurement of the speed of neutrinos. OPERA's mistake was unfortunate, but it doesn't change how impressive the measurement itself was. It just changes how impressive the result is.

"There's a bad moon on the rise"

In 3 days time, NASA's Lunar Reconnaissance Orbiter will mark its 1000th day in orbit. To toast this milestone, NASA has put together a short video depicting what we think has been going on for the last 4.5 billion years or so since the Moon first formed. It show very clearly how violent have been the processes that formed our largest satellite as we know and love it today!

Monday, March 12, 2012

Art and Elsewheres

Hiroshi Sugimoto, Surface of Revolution with Constant Negative Curvature, 2008

I’m interested in abstraction as a real issue for creative collaborations between art and science. As I wrote in my last post, science habitually turns to orders of abstraction and mind-bendingly variable magnitudes of scale when it turns to mathematics; but ‘abstraction’ also has a very specific meaning for art, and especially modern painting. So what happens when you put these two kinds of abstraction — mathematical and aesthetic — together?

Mathematics: A Beautiful Elsewhere was an exhibition at the Fondation Cartier pour l’art contemporain, Paris late last year that put leading mathematicians and artists into contact with a open collaborative brief. The curator quotes the mathematician Alexandre Grothendieck to describe the exhibition’s aim as offering visitors “a sudden change of scenery.” The show was unusual enough to elicit some thoughtful response, not least a long interview with David Lynch, who contributed a film. "If you think of cosmology, you picture colourful nebulae; with neurology, intricate brain scans," notes the New Scientist. "But what does mathematics look like?" The answer is, implicitly, abstraction. The Nature review was a bit more curmudgeonly, accusing both the artists and the mathematicians of insufficient or obtuse communication. Other pieces in the show included included a live video feed from the CMS and ATLAS experiments at the Large Hadron Collider in CERN, Geneva: a groundbreaking-everyday kind of experiment which this reviewer — an astronomer — describes as "the mathematical word made flesh." For another commentator, the dominant theme was mystery: 
Mystery is perpetuated not only within the pieces and puzzles themselves, many of which are initially impenetrable to the mathematically untrained. It also lies more fundamentally within the very essence of the subject under examination. The message runs clear throughout: mathematics is in itself a mystery, the truth of which may never be attained. 
Sea Waves (wave equation) image from the film Mathematical Paradises, 2011.


Talking with Shaun recently about fractals and the questions they raise — are they art? are they science?  Does the mathematical basis and quality of endlessness make this an 'aesthetic' or a 'natural' object? — reminded me of the persistent problem of vocabulary for the desire to talk across these particular disciplinary lines. In a book titled What is Philosophy?, a collaboration between a philosopher and a practicing psychoanalyst, Giles Deleuze and Felix Guattari express the problem in this way:
When philosophy compares itself to science it sometimes puts forward a simplistic image of the latter, which makes scientists laugh. … Paul Klee’s vision was certainly more sound when he said that mathematics and physics, in addressing themselves to the functional, take not the completed form but formation itself as their object. … If philosophy has a fundamental need for the science that is contemporary with it, this is because science constantly intersects with the possibility of concepts and because concepts necessarily involve allusions to science that are neither examples nor applications, nor even reflections.
(Deleuze and Guattari, What is Philosophy? 154-5, 162)
Ulam’s Spiral, in The Room of the Four Mysteries, part of a film by Beatriz Milhazes.

It strikes me that Mathematics: A Beautiful Elsewhere is an attempt in the right direction at a real space of collaboration — the kind of thing that Deleuze and Guattari would call taking "not the completed form but formation itself as [its] object." This is for a couple of reasons. One, the sheer range of works and collaborations in the show. Each is a sincere attempt to bridge the gap through multiple types of visual languages, and in a range of different media, and no individual piece can claim to individually crack the problem. This variety is an important part of making the viewer aware of the potential heterogeneity of the art/science matrix. By comparison, the CERN residency picked only a single artist following a wide call for proposals, which is a relatively pre-determined approach that is more likely to fall into the trap of fixating on artistic personality.

Second, Michel Cassé — an astrophysicist and a co-curator of the Mathematics show — put a range of experienced specialists together, gave them an open brief, but required the outcome to take the form of an art exhibition. This is a useful constraint for a topic so predisposed to wander into the literally ineffable. And so, for all the occasional accusations of impenetrability by the critics, this is an approach to disciplinary collaboration that I hope we see more of in contemporary art: process-oriented, carefully conceptualized collaborations that are not just flashy "allusions to science" but that prioritize the actual objects, methods, and incipient discourses that might fail, may communicate only partly, but begin to construct the common vocabulary.

Live video feeds from experiments at the Large Hadron Collider in CERN, Geneva