The good Doctor’s appeal to pretty much anyone reading his books is undeniable. The raw feeling of being in the moment, the intense emotional presence of his writing is overpowering. The story, with facts and opinions woven haphazardly into it, is just one piece of the mad painting. This became known as “gonzo journalism,” foregoing an observer’s impartiality for the unbridled excitement of being part of the story. It’s more of an eye witness account than a sterilized retelling of the events.

Though decidedly partial, his version of the events offered a greater gut, visceral insight into what happened. Agree or disagree, reliving it through someone else’s eyes is bound to be interesting.

I would argue that such bits of mad inspiration, instinctual, careless creation exist in science, too. The Scientific Method is what comes after the secretive, mystical subconscious twists and turns by some unknown dynamics into an insight. Through some devilish talents, the improbably subconscious seeks out the single solid kernel of truth amongst a sea of half-truths and incomplete datasets. Somewhere, there is a truer pattern behind the dusty and rare windows, and discovering this pattern is not a reasonable or rational course.

The flash of genius is a lightning strike, and the few lucky enough to witness it are revealed a picture that no one has seen it before, and they are left o rediscover it, piece by piece, sometimes spending years, or their whole lifetimes, attempting to once again capture that single vision.

I say, the unquantifiable, unfathomable inspiration is a central part of what we, as scientists, do. The spark of the subconscious is as important to us as it is to artists. The Method to us is as mastery of the brush was to Dali, or skill with the guitar to Hendrix. They are the means to dress the original inspiration into a more physical manifestation that others can understand. With equations, we paint our beautiful, elegant truths.

So why not sex, drugs, and quantum mechanics?

Do we, perhaps, spend too much time honing the rational and the strictly sequential parts of our minds? Should we indulge the less structured parts of our soul, for they are the ones that will bear the more brilliant fruit?

After all, the unknown is our business. Some madness along the way is to be expected.

I’ve recently come across a problem – I needed to see if the lamp I was using had a lot of UV output (in particular, I’m interested in 250 nm – 350 nm range). Normally, you would get a spectrum analyzer, and simply measure the output. I had no such thing available to me, and yet I needed to know.

A useful thing to know is that starch luminesces bright blue when exposed to UV light (hence white cotton tshirts glow under black light). It is also useful to know that regular glass (such as a microscope slide) cuts off transmission at right around 350 nm.

So, I came up with a couple of tests.

Hack #1 (very crude)

Use a fused silica lens to focus the light, and then insert something sharp and opaque near the focal point. Some distance behind it, put a piece of printing paper. Distorted focal point will project something of a rainbow (as different wavelengths diffract at different angles). You will see a blue band. But – this could be actual blue! So, take a microscope slide and cover up the blue portion of the rainbow. If it disappears, it was due to UV. If it doesn’t, that’s actual blue.

Hack #2 (slightly more refined)

I’ve described before how inserting a microscope slide into a beam removes about 8% of light due to reflections on the surface. That is of course true only if there is no internal absorption.

So, put a detector (such as a Si detector with a UV-transmitting window) into the lamp’s output. If the slide attenuates the beam by about 8%, there is no UV. If it attenuates detector’s output by significantly more than 8%, there is a good amount of UV.

8′ S-video cable at Target: $21

12′ S-video cable at SWS: $5

Shop locally. Save silly amounts of money.

PS I never quite understood how big retailers are able to charge obscene amounts of money for cables. I’ve seen ethernet cables upwards of $20 for 6′. I just saw a HDMI cable for $70.

In my recent tinkerings with various circuits, I have of course come across the problem of power supplies. Most opamps need a few volts of bias and a bipolar supply to operate properly (usually something like +/- 12-15 V and ground). Insufficient voltage can reduce bandwidth, and overall it’s not very good for the part.

Getting that much out of batteries can be difficult – no one wants to stack 10 AA. 9V and 12V batteries are of course better as far as space, but lose voltage over time, affecting circuit performance in sneaky ways.

That is why voltage regulators exist – for less than $2, they can deliver up to 0.5A of current with about 40 mV accuracy of line voltage in a tiny TO-220 case (even better if you use bypass capacitors). Good examples would be something like LM341 from National or MC78M00 from ON. And these are fixed regulators – you don’t have to fidget with setting them to desired voltage.

Of course you don’t even need a lot of current – opamps probably don’t need more than 30 mA to work anyway.

Similar to positive regulators, there are negative regulators. So together, you can have a regulated, stable powersupply from two batteries.

Of course if you know of a computer equipment recycling store (such as RISE in Tucson), you might be able to get a DC power supply in almost any voltage for about $5.

Transforming it into a bipolar supply is easy – daisy-chain two positive regulators. Simply use output of the first one as ground input for the second one. (Do make sure that a single regulator can handle the whole DC voltage as input.) Now, for $10, you have a solid bipolar power supply. Which is less than you’d spend on batteries alone (12V A23 batteries are usually $6-8 each – woe is you if you forget to switch off the circuit and leave for the weekend).

(Of course, a more standard way would be with a transformer with a center tap on the secondary winding – but it seems those are more difficult to find, and much more expensive. Stringing regulators probably won’t work very well for applications that require more power – the first one would probably overheat easily. This is a cheap hack.)

Still in San Diego.

Staying in a hostel.
The hostel has a permanent smell of sweat and booze. It’s like a permanent college party, but not annoying. And there’s free all-you-can eat cereal. And a sadly idle waffle-maker.

The better sections.
Only here you’ll hear things like, “What the founding fathers of quantum mechanics have all missed is this…” And for once, it’s not the time-cube guy.

The industrial exhibit.
The whole thing is brought to you by words “innovative” and “solutions”. I’m pretty sure every company slogan and self-description is some convoluted form of those two words.

“While conversing about science” is a more accurate title, I think.

I’m currently at an SPIE conference in San Diego. Just had my first day, and must say, in some ways – the most interesting conference I’ve been to. Being a grad student I haven’t been to a lot, but been to a few. Including one SPIE before.

I’m staying in a hostel.
Our grants have run out and I’m pretty poor, so I couldn’t afford to stay in a real hotel, so I’m back to dorm life, basically. The first night was pretty awful, on account of a locked window and lack of AC. Somehow, I still managed a decent night’s sleep. The window is open this time, so maybe it’ll work out better. My roommates have been pretty cool so far.

The better sections.
I am pretty convinced that in the future, I should go to the sections that explore the most esoteric and hypothetical parts of nature. Maybe some of the stuff is over my head, by participation is much more interesting, dialogue is more genuine and engaging, and there is endearing and awesome craziness about.

The hidden variables are hidden behind the cows.
There was a talk that claimed to be about hidden variables. They were not mentioned other than in the title, but there were lots of pictures of cows.

Wikipedia cited as references
At least four times now. Come on, scientists! If you cite Wikipedia, it pretty much means you were too lazy to look up sources cited on Wikipedia. I particularly enjoy when there is a bibliography page at the end of someone’s presentation, which lists several sources in proper bibliographical format, and one of the entries is simply “Wikipedia.”

That’s it for the moment. Maybe more later.

I know pretty much just enough about information and complexity theory to get myself in trouble. So, well, here we go.

What appeals to me is the implication that vastly different systems can operate according to the same laws. That there is some basic governing mechanic of systems. In particular, that this can cross from unliving systems to living ones, and in fact societies can act on the same principles as systems composed of unintelligent agents.

One such similarity is between our societies and our anatomies. The way we behave is probably connected to function of our biology – and in fact at some basic level, all life functions on the same behavior pattern, whether it has intelligence and self-awareness or not. I’m not just talking about evolutionary psychology, but something more fundamental.

The particular point that got me on this track was watching a 30 Days episode about incarceration. That was the starting point, but I think the problem of the American prison system (the prison-industrial system) is nothing new. The jails are generally hugely over-crowded, and a huge proportion of those on the inside are repeat offenders. People who keep committing crimes and going to prison, often after just a short time on the outside. Real rehabilitation is minimal.

Anyway, my point is not about problems of the American system, but rather how it is similar to a body’s response to infection, and how similar problems arise in both.

When our bodies detect an infection, white blood cells are sent out to deal with it. They mostly work by simply ingesting the intruders by special hunter cells called neutrophils. However, it is possible for a neutrophil to take on too many intruders. It cannot effectively digest them and dies, releasing those it has captured.

Which sounds remarkably like our society’s attempt to deal with criminals. We isolate them in jails in hopes of neutralizing their threat. And I suppose it can, or has worked to an extent. Dangerous elements are taken away and isolated from the rest of the society, and when they come out, hopefully it is in the form of a reformed citizen.

Presently, this has mostly failed. Inmates without access to proper rehabilitation frequently end up in the same lifestyle and simply go back to jail again having committed more crimes. And once someone spends most of their time on the inside, you have to wonder if the prison life is the one they’re psychologically more accustomed to. Therefore, the prison system loses its ability to effectively “punish” and correct behaviors.

In fact, the terrible overcrowding has created conditions where prison gangs arise to considerable power on the inside, and many criminals are able to continue their actions from the inside. Being in prison sometimes can act as training grounds for the next generation of criminals.

Prisons have consumed too much.

(This is another post I’m copying from my old MySpace blog. I’m eventually going to kill that account, but some of the entries I kind of like.)

Repost begins

(I’m starting to think that I should have been studying complexity theory instead of physics…)

Artificial intelligence is kind of like the Holy Grail of much of modern research. Wether you’re in software and you’re trying to find that perfect system of algorithms. Or a philosopher trying to figure what makes a decision and how to quantify it. Or a scientist, developing digital systems that are ever faster, with ever more storage capacity so we can build a system that can learn and adapt in real time, dealing with the torrential pour of information that is reality.

I think it may emerge on its own. On the internet.

Complexity theory’s basic premise is that if you take lots and lots of simple units and put them together, they will form a system that may have properties that cannot be predicted based on the properties of the simple units. These are called emergent properties.

For example, humans. There are certain common traits that we can identify. However, when you put them together into a society, certain behaviors arise that did not exist before, such as division of labor. Both crime and law enforcement.

Or take a billion ping-pong balls. Each one is light and bouncy, and certainly well-defined on its own (unlike a gallon of water – mixes with any other amount of water). However, pour these balls down a pipe, and they’ll behave like a stream, forming waves and jets.

The internet is an incredibly huge collection of individual systems. A mass of processing, storing, and communication power. It is a primordial soup out of which digital life will emerge.

Information on the internet has several properties similar to living DNA. For example, information is functional – it exists in environment where it can instruct actions and change. Simple put – information on the internet is code which makes up programs that do things. Programs that affect how machines run.

Transfer of information is imperfect – entropy demands it – just like copying DNA during cell division. Both processes have strong error correction mechanisms, and in both cases most mutations are lethal (corrupt files won’t run), but in both cases it is possible, however, unlikely, that the mutation will happen and it will be functional.

There are already in existence computer viruses that mutate and adapt. In the simplest case, they’re pre-programmed to re-compile themselves in such a way that their checksum keeps changing with every download, so it’s harder for anti-virus software to recognize them. This of course did not evolve on its own – but it is something that demonstrates a possible evolution mechanism.

The first digital life won’t be (isn’t?) very complicated. Basic self-replication and propagation functions. Not terribly intelligent. Intelligence is a necessary part of evolution, but it is not the starting point.

Out there, connected to the void, are millions and millions of machines. Many of them are not under direct control. Many run code that is poorly written and unmaintained. Digital cess pools in which new lifeforms could be developing undetected. Error upon error upon error… A nearly infinitely complex system coming to grips with itself.

Some of you may have witnessed the demise of CiN Wiki. It succumbed to spammers editing its articles. Interestingly, overwhelming majority of spam was complete nonsense – broken links, random words. Just garbage. It really looked like some blind and broken bot just kept ramming into the Wiki, without any supervision or direction. Like a dumb goldfish that keeps eating as long as there’s food.

The internet meets another important criterion for developing life – it has a huge entropy gradient, meaning that it is not anywhere near an equilibirum state. Amount of information, transmission capability, and processing power are constantly growing, and all without a plan, really.

How will we react to the new digital life? Will there be a great purge, where we hunt through every router, every box, every storage facility, rooting out digital critters? Will we cherish and nourish it?

Will the new life be brutal and barbaric? (I think it will be exactly like us. Whether that is brutal and barbaric, I’ll leave up to you.)

Eventually, though, will come more complexity. Higher order will develop as the system continues to evolve. Then one day, it will become self-aware. Something on the internet will dimly look back at us. AI will have arrived on its own. 4 billion years of carbon evolution squeezed into perhaps just a few decades of silicon.

AI will be just as organic as us. It won’t have the iron logic imagined in so many sci-fi novels. It will have all our faults and all our strengths. It will simply be able to adapt a lot faster than us.

Next time you get email spam advertising penis enlargment pills, maybe it’s not from a sleazy bastard with a T1 hooked into his house. Maybe it’s a renegade piece of code trying to adapt to its environment by mimicking it.

Repost ends

I’m sure there is already a field of experts working on this.

Who will be the first digital explorer? Wrangler? Zoo keeper?

I have not received many awards in my life. Most of the awesome things I have achieved in my life are not the kind that earn you material awards.

So I thought it’d be nice to keep track of small marks of recognition I have received over time. Marks that may not mean all that much in terms of lasting greatness and historical fame. Not the kind that hangs on a plaque in an historic building.

While I’m on the subject, I’m going to make the following claim. The awards that are honorable don’t tell you very much about the person. Sure, it tells you about the great things they have done – but not much about the kind of person this particular individual is.  Sure a Nobel laureate is uber-smart – but is that the kind of person you’d take bar-hopping? The small dubios distinctions are much more unique, and in this way, tell you something about the person.

So, with minor clapping, my first dubious award.

I have received an Optical Sciences Center (yes, Center, not College) pin from our Dean, Jim Wyant. I was working a promotion booth for the college at an OSA confernce. Dean Wyant came over to talk to me and the other guy, and then gave me this pin – after taking it off his coat.

Now, he probably has a drawer full of these things – but I bet there is not a whole lot of people who got one of these directly from Wyant’s own suit.

Through high school and college, I actively participated in Model United Nations. Toward the end of my time at University of Florida, we started adding “crisis simulations” to our venues.

In a crisis simulation, delegates are locked in a room, while a crisis unfolds before them in real time. They get regular updates on what is happening in the world – from in the form of news reports and agency briefings. Some of the information is presented to the whole group. Some information is presented to select delegates. They could also send information requests to their home governments. A crisis team worked in a separate room, coordinating events and responding to delegates’ actions.

The result was intense – the regular fabric of diplomatic debate was now distorted by an unpredictable flow of events. Prepared positions weren’t worth much past the first ten minutes. The educational benefit – I’d like to believe – was also greater than a regular debate tournament.

The basic idea of these was simple – delegates need to figure what is going on and decide how to deal with it. There is a body of information out there, and they need to find a way to tap into it. What is actually presented is a small portion. They need to understand venues of acquiring information, and how to ask the right questions.

I think this basic idea can be applied to science education at graduate level. To some extent it already is – many classes have projects. Usually, though, these are very simple design projects that have a very clear breakdown of options and objectives, and span maybe a week of work.

I’d like to see something bigger – a whole semester’s worth. A class dedicated entirely to a fictional research project.

As part of the course, students would be given a starting point of a project and the goals. The line connecting the two wouldn’t be entirely obvious or singular – that is rarely the case in a research project. Students could request specific measurements or experiments to be performed, and the teacher would return with results. Data analysis would be up to students, to figure out what this means for their goals, and how to refine their process, how to follow up with other measurements and experiments.

It has been impressed upon me that the difference between someone with a Master of Science and someone with a PhD is not necessarily experimental or theoretical proficiency, but the ability to competently form a research project, clearly define its parameters and goals, and then carry it to the conclusion. This greater perspective, ability to step back and see how various pieces come together to form a vision, that is what makes a true scientist. Not someone who is merely proficient with the material.

Classes like the one described would teach exactly that. A scientist should be able to operate on a greater scale than his area of expertise. One needs to understand utility of techniques and measurements outside of their focus.

I can’t help but draw inspiration at this point from House. Not necessarily from the characters, but from their methodical approach to the problems. From their ability to quickly form specific theories based on available evidence, and then decide on a series of specific tests that would discriminate between the theories. Physics is no different (as this is a manifestation of the Scientific Method), though perhaps less dramatic.

Everyone can recite the Scientific Method from heart. However, applying it in a real-world situation is something else entirely.