18 September 2005

The Emperor's New Mind by Roger Penrose

*Spoiler Warning*

First, a warning of sorts: this book of 602 pages is not light reading. This strange hybrid of a philosophy treatise and a science text doesn't even have a very satisfying conclusion. For a book that sets out to answer, or at least attempt to answer, what consciousness is, the answer, when it came, lacked conviction. I'm sure that the author was convinced; the fault lies with the writing. Science communication (and I distinguish this quite clearly from communications between scientists: science communication to the lay-person is an art form, not an application of the scientific method) is not only about the transmission of information and knowledge: it is also about getting logical agreement and emotive buy-in from the reader. A popular science book that doesn't provide a good 'satisfying read' serves no one.

This is disappointing, for the background information Professor Penrose provided (which in a popular science book such as this one is essential to leading the non-expert to the the author's conclusion in an inevitable and logical way) is a good and solid introduction to the fundamentals of logic, artificial intelligence and modern (i.e.20th century) physics. It is unfortunate that the robust background information could not build up to a firm pronunciation (the author, was in the typical English gentleman way, far too apologetic) of Professor Penrose's belief that consciousness is non-algorthmic, (possibly) works because of non-local effects stemming from the Einstein-Poldolsky-Rosen paradox, does not cause vector-state reduction (refer to the 'Schrodinger's cat' thought experiment), and will require further improvements in our understanding of the Grand Theory of Everything (in which the marriage of quantum electrodynamics and general relativity is only the first step) before any description is possible.

Don't get me wrong: this book has its brilliant moments. The introduction to computing and Turing's Test is clear, and the arguments between a deterministic universe versus a computable universe compelling. The physics discourse is heartwarmingly nostalgic of my undergraduate course and actually taught me a new thing or two. The chapters on existing knowledge of the human brain and consciousness were well written and informative (though I wish there were more details). My favourite paragraph in the entire book is this:

"Many philosophers and psychologists seem to take the view that human consciousness is very much bound up with human language. Accordingly, it is only by virtue of our linguistic abilities that we can attain a subtlety of thinking that is the hallmark of our humanity - and the expression of our very souls. it is language, according to this view, that distinguishes us from other animals, and so provides us with our excuse for depriving them of their freedom and slaughtering them when we feel that such need arises. It is language that allows us to philosophize and to describe how we feel, so we may convince others that we possess awareness of the outside world and are aware of also ourselves."

This had me lean back on the chair to marvel at the idea that somehow, the very language that communicates this thought to me is the very embodiment of consciousness. This is a simple concept, yet it hides such complexity behind its 'obviousness'. What of dolphin consciousness/intelligence? What of two people speaking completely different languages? Surely there are ways of bridging the gaps! (Just imagine a first contact situation, where two aliens... sorry, beings.... are trying to learn from one another how each other counts.) And if so, does it not favour an objective universe, and not a subjective one? The sheer importance for the role of imagination in a subjective universe defies belief, and I'd go with Occam's Razor on this one.

In the fifteen years since this book was written, there has been progress in the field of super-string theory which goes some way to further explain the fundamental 'realities' of this universe we live in. My feel is that if Professor Penrose had these data back then, his arguments might have altered a little, but the fundamental arguments would still be the same. As our knowledge of clinical consciousness and the elementary (excuse the phrase... would 'fundamental' be better?) physics increase, how can we use science explain consciousness? With new, unknown science, it seems.

As you grow older...

... you may find several truths.

1. The weather is getting worse.
2. Children are getting ruder; people less caring and civilised.
3. Politicians are getting more self-serving and (gasp!) corrupt.
4. The world gets smaller thanks to improved transportation links, but 'progress' becomes doubtful.
5. It takes more to make you go 'wow'.

Sad, isn't it?

11 September 2005

Sound tests (not rigorous)

This week, Apple announced a new product called the iPod nano. This, of course, created a wave of iPod desire which simultaneously pleased many ('more toys!') and pissed off many others ('I've already have an iPod, grrr!!'). It's difficult to decide which category I fall into.

Now, the hypothesis is that the people at One Infinity Loop has been improving their product offering. To convince myself of this hypothesis, I did a back-of-the envelope, limited-to-ears/brain-interface, doesn't-conform-to-the-Scientific-Method experiment to test the audio output between a 2nd generation 20GB iPod (bought in the good old days of 2002 when iPod was still Mac-only and favourite green-eye monster bait for all those Windows people) and a 1GB iPod shuffle. It's a little like comparing apples with oranges, given that the two products have quite different designs and specifications. However, since they both play the same set of music, it's fair game (just like apples and oranges are both fruits: fair game), methinks.

1. Downloaded Star Wars Ep III soundtrack (160kbps) onto both devices.

2. Set iPod to no EQ setting. Locate experiment in quiet place. Standardise volume output as much as possible using ear/brain response to the first bar of the music.

3. Used old ear phones that came with iPod for general 'environment in crowded subway' testing

Results: Like many have noted elsewhere, I noted that the Shuffle gave a bass response which was both louder and clearer in comparison to the rest of the audio range. No obvious difference for middle and higher registers. Overall, the Shuffle gave a clearer, more crisp experience.

Note: Hints of strings being plucked strings could be discerned from the iPod; same set of strings being plucked are very clearly heard in Shuffle.

The same results are generally true for most other music types with well defined percussion rhythms.

To repeat the experiment:

1. Downloaded Mahler's Second Symphony, Fifth Movement (128kbps) onto both devices.

2. Set iPod to no EQ setting. Locate experiment in quiet place. Standardise volume output as much as possible using ear/brain response to the first bar of the music.

3. Used Shure's E2 in-ear ear phones (see also 'tinnitus in abnormally quiet environment')

Results: there's an soft but appreciable background hiss from the Shuffle, which wasn't there with the iPod. This is most likely due to impedance matching between the player and the ear phones, but mechanical insulation effects are also likely. Also, the solid state electronics give a very soft, high pitch whine with each press of the 'scroll wheel' button, which is cool from a geeky point of view, but highly annoying to a musician's ear. (One can argue weakly that it resembles analogue hissing from vinyl, but this hiss isn't 'vinyl warm'; it's more 'dentist's drill'.)

Despite the hissing (which your brain happily ignores once the music becomes loud), the Shuffle gave a better technical performance, especially with the bass response. The balance between the low, middle and higher registers was more pleasing than what the iPod gave. This was demonstrated quite clearly by the time the choir sings its second triple-piano entrance two thirds into the movement. (The first entrance is way too intense to be anything but a test of your hearing's lower threshold).

In terms of atmosphere though, iPod gave a better performance. In general, the iPod gave a better feel when it comes to classical music, and the Shuffle performed better for anything that isn't classical music.

Overall, the Shuffle is still the better of the two, given its size and form factor, and the kind of music one listens to in busy subways and commuter buses.

It'd would be interesting to find out if the iPod nano has the same hissing problem with 'sensitive' earphones, given its size and flash memory design. Maybe it's time to head down to the neighbourhood Apple shop.

Tacet! Tacet! Tacet!

John Cage's most infamous composition was his 4'33". Read a review of the performance here.

Only if there are more of such innovative music around. Where else in this busy, noisy world will we be able to enjoy the tinnitus of an abnormally quiet space?

So, What's the frequency of your tinnitus response?