Plant Intelligence
There is an interesting discussion over at metafilter about plant intelligence.
A Trip Down the Uncanny Valley
Why does it freak us out to see simulations of humans that fall just short of the real deal? When it’s stylized, just a collection of parts that attempt no more than to signify personhood, we’re fine. But when its real enough that we’re briefly fooled, that we feel the tug to fawn or lust, we’re repulsed.
We group the world in to things we judge to be conscious and things whose welfare we aren’t concerned with (NO! You took the batteries out of my gameboy in the prime of it’s life!). No one would wince if I stabbed an apple with a dagger. How about a drawn smiley face? A G.I. Joe? A tickle-me Elmo? The kind of stuffed baby doll a toddler would play with? How about one of these? A human fetus? An animal fetus? It might be a fun exercise to sound out our “gut” reactions, but after only a moments deliberation we can tell whether something is really worth feeling compassion for, right?
And yet, people disagree. In the United States most people are carnivores, only drawing th line at animals they might keep as pets; I don’t know anyone who eats dog regularly, but it’s commonplace elsewhere. I’ve stopped eating mammals from time to time, but I’ve never seriously considered abstaining from fish (for moral reasons, anyway) - I’ve spent a lot of time around them, and I have difficulty believing they’re conscious in any important way. At the end of the day, it seems like most people operate on the somewhat less than rigorous don’t-eat-anything-with-a-face rule.
Are individual ants conscious? I don’t feel too bad when I crush one. Is a colony of ants conscious? Harder to say. If they are, then what about colonies of people? Is there something-that-it-is-to-be-like the city of New York as a whole? Is there something-that-it-is-to-be-like the right half of my body (as distinct from the left, or whole)? How do you delineate something that experiences (an experience-er), when there aren’t any apparent lines in the sand where causal interactions stop? I find those examples far fetched (even though I fetched them), but if you’re not a mind-body dualist (and you shouldn’t be), it can feel like a slippery slope to something like panpsychism.
There is a great episode of RadioLab where researchers ask subjects two versions of the Trolley Problem while undergoing an MRI scan. The first formulation is a classic intuition pump for selling the virtues of moral consequentialism:
A trolley is running out of control down a track. In its path are 5 people who have been tied to the track. Fortunately, you can flip a switch, which will lead the trolley down a different track to safety. Unfortunately, there is a single person tied to that track. Should you flip the switch?
Almost all the subjects say yes. Then you ask…
As before, a trolley is hurtling down a track towards five people. You are on a bridge under which it will pass, and you can stop it by dropping a heavy weight in front of it. As it happens, there is a very fat man next to you - your only way to stop the trolley is to push him over the bridge and onto the track, killing him to save five. Should you proceed?
Very few people would push the fat man. Philosophically this can be justified (amongst other ways) by denying that consequences are all that matter in moral choices (perhaps acts are good or bad in themselves). But the researchers interviewed in the RadioLab episode aren’t philosophers - they’re just interested in what parts of the brain are being used while the subjects deliberate. When asked about the fat man, completely different regions light up then when given the first query. They speculate that in the lever example we’re able to be very Spock-like thanks to the involvement of machinery; the math is simple, maximize the number of lives. It’s not so easy with the mental image of pushing a fat man - perhaps it’s appealing to some much more basic part of our natures, not to hurt things that we empathize with, overriding our calculating selves.
Perhaps what is so disturbing about the uncanny valley is that its a perceptual illusion that suggests moral consequences. If I show you a penrose triangle you’ll probably find it an amusing curiosity, but hardly a challenge to your ability to function in the world. Now, a doll that you can’t tell from a person, however briefly - that calls in to question the epistemic grounds for your ethical decisions. If you can mistake something unconscious for something conscious, the much scarier reverse case could be true. A pop-sci article has been making the rounds that claims men’s brains respond to scantily clad women and tools the same way (supported by MRI scans, of which we should be skeptical). If our brains don’t recognize women, let alone ant colonies or trees, as things we have to treat morally right off the bat, our cherished moral intuitions are called in to question.
The documentary video links that follow explore the lives of people who come across as all too comfortable with this ambiguity. I recommend starting with the incredibly wittily titled Guys and Dolls, which is about the owners of high end mannequins designed for sexual gratification. If you can make it through that, try this feature from the BBC titled (matter-o-factly): My Fake Baby (Part One, Part Two). It’s about… Fake Babies. The documentaries aren’t just about the objects, but the people who relate to them, and the human drama alone makes them worth watching.
The Obama Innauguration, Balloon Photography
My family’s company, Digital Design and Imaging, will be providing video and still images for CNN and the AP for the hours leading up to, and shortly after, the oath swearing (for security reasons we’re grounded during the actual event). I’m told John King and his multi-touch screen will be presenting still imagery with Microsoft’s Photosynth, which is essentially a system for positioning many two dimensional photographs in 3d space through clever algorithmic inference based on shared regions of images. You can see a teaser of what we’ll be doing, shot several weeks ago, here (though be warned it requires a quick install and is limited to Windows operating systems). You should be able to watch imagery go up on that site in near real-time on MLK and Inauguration day. As more content becomes available, I’ll update this blog post, but generally the link for the teasers (a search for the username curtw) is where things should show up first. Stills may additionally be found in this gallery.
I’m presently in the computer lab of the University of Kentucky’s business building. I arrived in Lexington from DC via plane about 72 hours ago. At around 7pm this evening I hope to be en route to DC, one of about 2 million. It’s roughly an 8 and a half hour drive, so I should arrive at my parent’s house at 3am. My compatriots and hope to return by 10am Wednesday morning and in doing so make it to class. Some have expressed the desire to live vicariously through me, so I’ll post periodic updates throughout the weekend. It’s going to be a hell of a trip.
Pop Music
The KLF were two British chaps who exposed the mass sphexishness of the music world, fired machine gun blanks at award show audiences, made and set alight a million quid, and wrote a wonderful book to help you follow in their footsteps.
Do not try and sit down and write a complete song. Songs that have been written in such a way and reached Number One can only be done by the true song writing genius and be delivered by artists with such forceful convincing passion that the world HAS TO listen. You know the sort of thing, “Sailing” by Rod Stewart, “Without You” by Nilsson What the Golden Rules can provide you with is a framework that you can slot the component parts into.
Firstly, it has to have a dance groove that will run all the way through the record and that the current 7″ buying generation will find irresistible. Secondly, it must be no longer than three minutes and thirty seconds (just under 3′20 is preferable). If they are any longer Radio One daytime DJs will start fading early or talking over the end, when the chorus is finally being hammered home - the most important part of any record. Thirdly, it must consist of an intro, a verse, a chorus, second verse, a second chorus, a breakdown section, back into a double length chorus and outro
It is going to be a construction job, fitting bits together. You will have to find the Frankenstein in you to make it work. Your magpie instincts must come to the fore. If you think this just sounds like a recipe for some horrific monster, be reassured by us, all music can only be the sum or part total of what has gone before. Every Number One song ever written is only made up from bits from other songs. There is no lost chord. No changes untried. No extra notes to the scale or hidden beats to the bar. There is no point in searching for originality. In the past, most writers of songs spent months in their lonely rooms strumming their guitars or bands in rehearsals have ground their way through endless riffs before arriving at the song that takes them to the very top. Of course, most of them would be mortally upset to be told that all they were doing was leaving it to chance before they stumbled across the tried and tested. They have to believe it is through this sojourn they arrive at the grail; the great and original song that the world will be unable to resist.
So why don’t all songs sound the same? Why are some artists great, write dozens of classics that move you to tears, say it like it’s never been said before, make you laugh, dance, blow your mind, fall in love, take to the streets and riot? Well, it’s because although the chords, notes, harmonies, beats and words have all been used before their own soul shines through; their personality demands attention. This doesn’t just come via the great vocalist or virtuoso instrumentalist. The Techno sound of Detroit, the most totally linear programmed music ever, lacking any human musicianship in its execution reeks of sweat, sex and desire. The creators of that music just press a few buttons and out comes - a million years of pain and lust.
We await the day with relish that somebody dares to make a dance record that consists of nothing more than an electronically programmed bass drum beat that continues playing the fours monotonously for eight minutes. Then, when somebody else brings one out using exactly the same bass drum sound and at the same beats per minute (B.P.M.), we will all be able to tell which is the best, which inspires the dance floor to fill the fastest, which has the most sex and the most soul.
Maybe we shouldn’t be so hard on them. As Spider Robinson points out…
There are eighty-eight notes. One hundred and seventy-six, if your ear is good enough to pick out quarter tones. Add in rests and so forth, different time signatures. Pick a figure for maximum number of notes a melody can contain. I do not know the figure for the maximum possible number of melodies–too many variables–but I am sure it is quite high.
…
For one thing, a great many of those possible arrays of eighty-eight notes will not be perceived as music, as melody, by the human ear. Perhaps more than half. They will not be hummable, whistleable, listenable–some will be actively unpleasant to hear. Another large fraction will be so similar to each other as to be effectively identical: if you change three notes of the Moonlight Sonata, you have not created something new.
To be fair, there are worse progressions to rip off.
Electronic Goban
Went to HacDC and became a charter member tonight. Looks like a pretty groovy group and a very nice space with the requisite piles of technological ephemera.
Anyway, I’m pitching the idea of developing an electronic goban, which I’ve described on the HacDC wiki.
Many clients exist to play Go on computers/handheld devices. These are wonderful in that they can record your game in the SGF format, which is (near) universally used by players to review their games (an important part of getting better, often done with a teacher or opponent). They’re also handy when you’re moving around, and can’t set up a bunch of slick stones on a slab of wood (trains, planes, automobiles, bicycles). For casual games, it’s nice to be able to stop and resume at any time.
The big downside is face to face interaction. It’s simply nowhere near as enjoyable to play by passing a handheld back and forth or staring at a computer, the latter not being particularly portable most of the time anyway.
So the plan would be to build something that’s as close to the form factor of a real goban, but shrunk a bit for portability, and completely electronic. I’d say about a 30cm square (~1 foot for the unit challenged) and no thicker than a paperback book would be a good size. Something you can throw in a bag without hesitation. It should have as intuitive an interface as possible, the ability to automatically score the game, and best of all store games and export them to a computer somehow.
Blink Test
So there isn’t much I can do on the VU meter project until I get a hold of an oscilloscope. The next logical step would be to design the envelope follower that the comparator will receive input from, but doing this “blind” would be difficult. In the mean time I thought I’d check that the relay I’m using can actually switch as fast as I need it to (without burning up) and get a feeling for what rapidly flashed incandescent bulbs look like.
I wired up a simple astable multivibrator 555 timer circuit (which outputs a series of 5 volt pulses) whose frequency I could control with a potentiometer. I got a 6ft lamp extension cord, cleaved the two wires in two and cut one to put the relay in line.
The primary reason to do the test was to see how much latency the 100 watt incandescent bulbs display at various rates. The bulbs take a little while to get to full brightness after current starts to flow and even longer to become totally dark after it stops. A little bit of this is okay, and the trailing edge will actually produce a neat ghosting effect as the meter falls off. Too much latency though, especially in reaching brightness, will mean a dim top edge to the meter and a disappointing disconnect between sound and visual.
By playing around with rapid pulsing and watching it closely I don’t think it will be a problem. It’s difficult to tell without seeing several of these in a row though, so I’m going to try to measure the brightness curve from the following video and see if I can write a quick simulation (using processing?) of what the completed meter would look like with these bulbs. The camera’s auto-exposure function sensor latency will skew the data a bit, but it should be enough for a rough idea. Alternatively, when I get an oscilloscope (what can’t they do!), I could use a photoresistor to measure the brightness curve more precisely.
I’ve started a photoset for this project here.
A Vast VU Meter
This is Patterson Office Tower (POT) on The University of Kentucky campus. It’s home to a small cafe/mezzanine, an obscure mathematics library, and the offices of all the professors and bureaucrats who don’t fit in the endearingly inefficient buildings elsewhere on campus. If, as civil engineers like to say, architecture is the art of wasting space, it’s oft-scorned brutalist style seems to be a good example of what would happen if we eschewed the discipline altogether.
On top of this mercilessly orthogonal concrete form is the broadcast tower of Radio Free Lexington. Despite it’s “dim bulb” 250 watt transmission power, the station currently reaches most of Lexington’s 275,000 or so residents via terrestrial broadcast thanks to the height of the tower and the hill it sits on. The FCC has approved WRFL for an upgrade to 7900 watts, which will let us reach a much larger swath of the central bluegrass.
So I have this idea for an installation I think would work well to draw attention to the station either in celebration of the upgrade or as a means to gather support for it. Following in the steps of the green building hack at MIT and the Blinkenlights Project I’d like to turn POT in to a giant real time volume display of the station’s broadcast.
To be more specific, I want to take a row of windows (or more, as money allows) in the building and place lamps in them to form a vertical bar VU meter. The overall effect will look like the video below, but 16 stories tall and visible from miles away. You could tune in to the station during the evening and watch this huge column of light on the face of POT leap up and down in synchrony with the music. I’m grinning just thinking about it.
How could that be done? Well the Blinkenlights and Green Building projects were both able to run significant amounts of cabling from large power supplies to relays across several stories/rooms. POT is in constant use so cost and difficulty considerations aside we can rule out any scheme involving huge cable lays.
Happily, the signal we want to represent is already available as radio broadcast on every floor. So instead of switching a huge number of lights from a central system, we can put a radio on each floor and have each set of lights switched autonomously in response to the signal it receives.
So the basic plan is to have a simple lamp with, lets say, three 100 watt bulbs in each window. You have a relay (actually, this relay) that can switch the bulbs on and off quickly in response to a low voltage signal. This signal is provided by a simple circuit that detects how loud the signal coming from the radio is, and while it’s above a certain threshold, activates the relay to turn on the lamps. By setting that threshold slightly higher on each successive floor, you’ll get that VU meter effect.
That circuit to sit between the radio and the relay will be an envelope follower and a comparator (probably an LM339). The envelope follower may be as simple as a diode to rectify the signal and a RC low pass filter. The whole apparatus (lamps included) should cost less than $40 (that’s a high estimate) per floor. Assuming we do 16 floors, and a few red ones on the top floor for a cool peaking effect, that’s around $800. Probably less. But let’s say $1000.
I’m currently working on a prototype of a single unit. I’ve already got one of the relays to play around with. The problems remaining are…
- Finding some radios that can run off a wall power supply, yet are cheap and can be had by the dozen
- Designing and testing the envelope follower with an oscilloscope. The big question here is what kind of time constant should be used to prevent flickering but not take too long to fall off.
- Finding good potentiometers for adjusting the threshold of the comparator.
- Packaging all of this, wiring everything safely, and in a duplicable way.
Then it’s just a matter of finding funding and convincing the powers that be (who do you even ask? the building super? each individual office holder?) to let us set it up. It could easily be reused or expanded (add some band pass filters for a frequency analyzer), and different locations provide for some interesting opportunities. For live events, you could use a low power FM transmitter to drive them instead.
Powered by WordPress using the DePo Skinny Theme.