Gentle readers,

I have moved my blog to mindlace.net where, in all likelyhood, it will reside for the rest of ever. My reasons for moving to WordPress:

• I am paying good money to have a whole machine on the internet.
• I hate the way lj handles formatting. I want markdown or one of its variants.
• I can't figure out LJ's templating mechanism, and I want to have a better look and feel for my blog.
• I want to (eventually) mix in non-bloggish content. WP is simply and cleanly written in php, and I can hack it trivially.
• Categories. Memories are kind of like it, but not the same.
• WP is standards compliant, using xhtml and css for structure and design, respectively, and doesn't muck up my posts with html.

So as a follow up on my little claymation snake, here's a snake capturing and then swallowing another snake, and subsequently regurgitating it.

so, in a programming language, the first thing you write is a "hello world".... well, here is my hello world clay animation. Also, a snake.

this image is the singularly most effective political statement I have seen in a long, long time.

For the last few weeks I've been particularly self-conscious. I feel like almost every interaction I end up in comes off awkward or poorly. The last two times I went out in public I felt this to the point of being seriously uncomfortable with the situation and with my interactions with people.

This isn't normal for me, at all. I think it partially has to do with the fact that Vika and I have made a number of commitments that require that I really don't slip up for the next month or so. I keep coming back to the point that I am what I am, regardless of my running narrative about who I am.

Doesn't make me any less anxious, or make me want to hang out with anyone more. If I've been at all difficult in your interaction with me over the last few weeks, I apologise.

Around The World In A Tea Daze - Tales Of The Inexpressible - Shpongle

Current Mood: pensive

I'm taking precalculus to prepare me for the two math classes required if I want to take most engineering courses - MTH 141 & 142 (calculus and analytical geometry I and II). Right now it's kicking my ass - I'm getting a low B, and after this second test, well, if I'm lucky I'm maintaining that.

Anyway, it's sure blown a hole in my prima-donna attitude. While I know this isn't representative of my undergraduate career, it's made me really reconsider the scale of my ambitions ... makes me think I'm going to have to pay more attention to social factors and whatnot to get where I'm hoping to end up.

Also, we had a housewarming party, that was most excellent. A couple of net.people came over that were slightly odd but mostly it was smooth and fun. We fed people all sorts of good hors d'oeuvres. There was some really good discussion about the nature of consciousness, with Andrew really pitching in - there's a philosophy group meeting on Friday, which hopefully will give me the chance to talk big talk w/o getting people annoyed. Our downstairs neighbors are cooler than I thought previously.

So, if I don't yap any more before the end of july, just blame precalc.

I've been wanting a new bike ever since my last one got stolen post-playa. Dahon makes folding bicycles for around the same price most places charge for non-folders. Definitely where my next biking-dollar investment will go.

I am reading an article from thirdspace called self-esteem theory and management. The rest of thirdspace looks interesting.

Radio Darvish - msng - msng

I moved today, and that is good, i suppose, but I'm too tired to feel it.
Here are some photos of the apartment. They don't really do it justice; I'll try again tomorrow.

It's still a lot of work between now and a nice place, but it has a lot of potential. If anyone reading this knows any dark ritual for removing paint, I would be eternally grateful for suggestions.

The refrigerator smelled nast and was old, even after the landlady made a pass at cleaning out the super funk that was in there (made me realise I am definitely going to want a quite separate lab.) So we went out on this odyssey to get a fucking refrigerator. We considered every model in existence, and were about to settle on a $650 22cf side-by-side model that consumes 725kw/yr. We ended up with a 5cf sanyo SR-4910M for $100 (consumes 316kWh/year) , a $130 FoodSaver Vac 800 and a Haier 5cf chest freezer for $180, which includes home delivery. The freezer consumes ~260kWh/year; I'll get back to you on that once it's delivered. So, we got 10cf for $280; that's $28 per cf as compared to $34 per cf for the larger model, and $0.48 spent per kW h/year vs. $0.89 per kWh/year; on the downside it's 39.4 kWh/cf vs 38.1 kWh/cf; however, I believe the slightly larger usable space on account of less gewgaws like an icemaker might tip this calculation in the favour of our arrangement. No fridge we looked at has an equal freezer/fridge ratio, and Vika and I can carry both of these. Finally, there's them 149kWh/year we're not spending. My guess is we're replacing a fridge (~19cf) that consumed around 2000kWh/year; so we're saving 1424kWh/year from what used to be spent in this apartment. We did get central air put in, so we're certainly going to blow that margin.

A big props out to Sean for helping us get moved and stuff cleaned, and as usual pulled out the nice tunes.

The elmwood community center is right across the street, so I should be able to both lift weights and volunteer to help teach people to read english; I love to read, and I would like to help pass on that bug.

The other night, Vika said to me that the system is a way of selling you on uncertainty. Our government preaches fear daily. Corporations expend exraordinary sums to figure out how to best inform you of an inadequacy you never knew you had, then pitch you the solution in the same breath. When Anthony Giddens talked about this in 1999 he used the phrase Manufactured Uncertainty.

There's still a huge chunk of people out there whose manufactured uncertainty is around matters of survival; uncertainty about food, shelter, clothing, safety from violence, treatments for disease.

I thought I had read some more formal study of the posessions of the very poor, and by memory these came up: a bowl with a utensil, an outfit, a knife maybe, and a handmade hut or shanty.

It's been this way for thousands of years, so let's not pretend it's going to change soon. I think what can change is the nature of those devices.

The bowl of the peasant of the future is a bowl that you can fill with biological waste, put the lid on, and have it power the hut. It does this by converting waste to methane, and the methane to electricity. There's an extraordinary dairy farmer in Marin County has built a 75kw methanogen-based power generation system: it runs on decaying cowshit.

So I like the sentiment, but why dick around with the cow when you can cut straight to the methanogens? (Some methanogens make methane with the energy from rocks. )

The same bowl that can break greens - stems, bark, leaves - into starches or sugars, good for those lean times before harvest comes. a bowl you can put the foulest of water in, and drink from in an hour.

A little 180˚ lens bindi that contains a supercomputer, can see in all the directions it can; communicates at around 10gb/sec any bindi within 100m via ultrawideband RF and at terabit speeds within eyesight via on-silicon quantum cascade lasers. The bindi does voice recognition and video projection, and has to be put into a bowl once a week to be charged.

A knife that narrows to a diamond edge made cheap and in bulk in diamond fabs.

The hut is made from bamboo that grows in a greenhouse (itself made of bamboo) in the local center; the bamboo that grows there is cured in a bacterial broth that exudes xylose polysaccharides that give it enhanced structural properties and dramatically extends its usable life.

Cellulose is basically paper. Sheets of this cellulose are used to fill in the spaces between bamboo stalks. Also in the greenhouse, Algae are used to produce acrylic acid - which, combined with the methane from the bowl, can be made into methyl and ethyl acrylatewhich the cellulose can be dipped in to make it transparent and plastic-like. Rayon is a fairly straightforward set of transformation from cellulose, one I'm sure you could get microbial assistance in.

The utensil is a spork.

There's been a lot of blather about how the abuse at Abu Gharib is "un-American".

Unfortunately, this isn't true. Abuse of prisoners is as American as apple pie. Moreover it is a natural consequence of the effect the prisoner-guard relationship has on people.

It was clear even before Gen. Taguba's report that torture - the kind of soft-shoe torture that leaves no physical marks - was an established and accepted operational tactic in post-9/11 prisons. This is not isolated behaviour; it started in Afghanistan, where aside from the thousands of people who died while being moved in shipping containers to prison, systemic abuse of prisoners was par for the course.

Even though the operational focus on military intelligence that gave rise to the specific abuses in the 9/11 prison system is novel, the abuse of prisoners by guards is not.

Aside from the inescapable fact that the prison system in America is deeply racist, in that a wildly disproportional number of minorities are incarcerated, abuse is also accepted and common practice in American prisons.

Furthermore, there seems to be very little that can be done to free prisons from abuse; abusive behaviour seems to be a natural consequence of the social dynamics of incarceration. The Stanford Prison Experiment is the canonical example of how dividing people arbitrarily into 'prisoners' and 'guards' rapidly degrades into abusive behaviour.

There is a way out, however, and we have excellent examples to follow. When Finland became free from the Soviet Union, it inherited a Soviet-style prison system, with near-US levels of incarceration, harsh, long-term sentences, and high rates of recidivism. Now they have one of the lowest rates of incarceration, lower rates of recidivism, and crime hasn't risen. While this matter is worth extensive study, an article in the guardian, and an excellent article in the NYTimes (sadly behind their paywall) both give an introduction.

While many Americans condone torture, it is widely held to be horrifyingly unethical, prohibited by UN convention, and ineffective in producing accurate intelligence. Furthermore, in Iraq, people have been living under a dictatorship for a long time; they know that narking on your colleague is the best way to show loyalty, so inventing something to tell your tormentors surely can't be that far from the mind of someone being sexually humiliated.

Since Americans are willing to dehumanise, incarcerate, and abuse vast swaths of their own population, there is little hope that whatever 'reforms' are made in the supranational prison complex operated by the US military will resemble anything like the success of Finland. It is clear that what is revealed in Abu Gharib is not the exceptional behaviour of debased individuals, but a mere pustule on the festering canker that is the 5.6 million inmate strong American prison complex.

And still we retain the arrogance to give people freedom from the barrel of our gun.

updates!

Slate calls the prisoner experiment allusion insufficient. Actually, in reading the article, it mostly says that the Abu stuff is worse. Also, The New Republic has a good article on how them merikans were just serving up a heapin' helpin' of homemade prison abuse.

I just signed up for Interpersonal Communication and Precalculus, to to take over the summer. These are the first college-level courses I've ever taken; hopefully I'll do OK in them. This fall I hope to take two more classes - probably Bio 101 and Portuguese 101, and then in the spring I should/could be full time.

As you are probably aware, the WTO has ruled that the US government must end an export subsidy mostly for Boeing and Microsoft. The bill that was supposed to achieve this has turned into a nightmare of pork featuring $170bn in tax cuts for corporations.

As you are no doubt also aware, corporations don't pay much taxes in this country; you and I do. They've never paid their fair share, but these days they pay an astonishingly low percentage of taxes.

Please consider calling your senators and telling them that you don't support tax cuts for corporations, that you don't support this bill, and that you certainly don't support any extension of the tax-cuts in this bill that are going to sunset.

first draft, comprehensibility dubious

Every work has its own language, and no a priori encoding scheme can be applied to it without the encoding running roughshod over the encoded.

There is a further critique that would suggest that the specific constraints of semantic encoding in XML imposes constraints on the structure of the document that are already a dire simplification of whatever 'structure of meaning' the document may have.

Some of the structural limitations can be worked around; xlink with xpath lets you escape the hierarchy, but there's no xml serialization of an xrange.

So within these already stark limitations, what can be said about the meaning in the document? Here are two pieces of encoding, one using the example in David Mertz' TEI Initiative overview and the other using a 'naive' encoding conforming to no DTD.

Both of these encodings have merit, but the questions they allow you to ask with XPath and XSLT differ; the TEI lets you ask questions about how many lines each character has, who spoke what lines, and who spoke adjacent to each other. There are ways to encode lower level in the TEI (more information about the quality of the lines; whom is the subject and object of the sentences, etc) but, from what I can tell with my admittedly somewhat limited stance, not 'higher level' concepts.

The naive encoding, on the other hand, does not allow you to find out much lexically, but does allow you to ask questions like "what weapons appear in King Lear". Naive encoding captures more than just higher level semantics already present in the piece; it also serves to illuminate the critical process of the encoder. The perceived downside to this is that while in some imagined future I could find all speakers of all plays helpfully enclosed in <speaker> tags, another naive encoder may consider weapons to be a primary feature of a play and thus worthy of its own tag, while this example relegates it to a type of accessory.

From the perspective of someone using markup as a sort of literary qualitative analysis, the Naive approach is a clear winner. Once a series of source documents have been marked up in a manner closely paralleling the meaning the analyst finds in it, broader questions can be asked. Any irregularities in encoding without a guiding DTD or schema can be cleaned up through the process of querying the data.

In short, I believe that the naive approach to encoding escapes the empty syllogisms Shirky warns of, and provides immediate utility to the encoder. As such, I think naive encoding will be (is) the dominant form of xml encoding, and (outside of highly normalized realms, like the xml surrounding an exchange of securities) the basic stuff machines will need to ruminate on.

Tonight our hairs changed colour and I have been working on installing syncato, to provide the back-end for the next generation of Roland^HT; this has to be done by the beginning of May so that we can have a poster ready. A brief quote about the project from one of our abstract submissions:

The roland corpus itself is being marked up in a dialect of XML that is specific to expressing the semantic components and interrelations of the corpus objects. The marked up objects are placed into an open revision control system called subversion that can be accessed using the WebDAV standard, though limits on existing clients generally require the use of a freely available client. Once in a revision control system the objects are imported into an XML database like Sleepycat software's XML database. At this point different views of the corpus may be generated through XPath and XQuery driving XSL transformations of the native XML dialect into formats suited to the final audience, with the first target being an XHTML interface suited to modern browsers.

Current Mood: ecstatic

I got my financial aid awards from Evergreen - no grants and $14,700 in loans and workstudy. I bill at more than I would get from workstudy, so 3,400 of that doesn't count that much. Tuition is $13k and change, since I'm not a resident of Washington. Additionally, Evergreen has no microbiology major, nor do they have Português classes.

So that's right out.

Instead, I'm going to go to URI as a full-time student in the spring. In the meantime I'm going to take some CLEPs for credit and take some classes in the summer session and in the fall. This should let me get to only a few credit-hours less than I would have if I'd started in the fall; if I take a summer session in 2005 I'll catch up entirely.

The other advantage of spring is that I'll be even poorer this year than I was in 2003, so I should get a pell grant in addition to local grants. Further, having taken CLEPs and a few classes I would be in a great position to attempt to apply to all the decent schools in New England - there's a "New England" residency where tuition is 1.5x the in-state rate.

URI does give merit grants, so my second year + should be less expensive for me.

URI has a laboratory of soil ecology and microbiology. As you may or may not be aware, I'm fascinated by decay. Microbes will eat your shit, no problem, and they're researching how to accelerate the process. A sufficiently accelerated and versatile decay engine would allow a biologically based artefactual space to be completely recycled.

URI is also involved in exploring microbial ecosystems of marine sediments, and they mention deep-sea sampling. I have a possibly naïve intuition that the microbes that live around heat vents will prove remarkably useful for biosynthesis, as they've got the run of an extraordinary temperature gradient, so I'd love to see and perhaps participate in any sea microbe action.

Exciting stuff; I'm looking forward to it.

So it's essentially official; all finalists have been contacted, and I haven't. Why they have to wait another month to send me an official decline letter, I don't know. I'm going to look into URI as a possible in-the-area alternative, and if that doesn't fly then it's off to Evergreen this fall.

At least 16 organisms from a diverse array of evolutionary lineages deviate from nature's standard code in the amino acid "meaning" they assign to specific codons.
evolution encoded, sidebar

I'm sure this is only of interest to cryptographers, some scientists, and those that have a vaguely theological interest in 'pure randomness'. Which really makes it no less general interest than most of my posts :)

I got blogged by boingboing today, for my little pedantic feedback on true random number generation. The point, which I should have expanded upon, is that hotbits uses the decay of Krypton-85 to generate its random numbers. This is a perfectly unpredictable event and as such, just as random as the photon shot through half-mirrored glass technique that random numbers uses.

However, hotbits measures four individual quantum events, divides them into pairs, and if the interval between the first pair is less than the interval between the second, generates a 1; otherwise a 0. This, I was told by someone more math-zen than I, reduces their statistical independence, though if it does it does so only very subtly.

This may very well not reduce their randomness; on the other hand, sufficient knowledge of the initial conditions could give you some predictability of output. The half-silvered mirror approach that appears to be used by randomnumbers.info means that every quantum event generates a bit, no analysis added.

Plus, you can have as many random numbers as you want as fast as the refractory period of the photo detectors allows; much faster than the 30 bits per second of hotbits.

So ok, the whack gig with the brown RUE program is that they changed their application procedures this last December. Now, instead of interviewing everyone in March and giving you notice by May 1, they interview only "finalists" in March and give you notice on May 1.

As you may have noted, March is kind of drawing to a close. So I called Thursday to ask if they have notified interviewees yet. She said she'd get back to me. No response, so today I called them again. A woman answered, after putting me on hold, variously that some interviewees had been called and that interview notices were in the process of going out.

So: possibilities include a.) Everyone has been notified, I'm not a finalist, but she's not gonna say anything in advance of the Official Letter of Dis that's still over a month away; b.) Phone interviews of remote people have been done and letters sent to locals; c.) they're conducting interviews but have so many finalists and so few interviewees that the process is just taking time.

I hope for b or c, obviously. However, since there's nothing I can do about this, I had to do something, so I called Evergreen and got my residency situation sorted out (not a resident) and figured out roughly what my fees would be. It looks like 5-10k/yr loans, about what I expect to go into debt for Brown, really. And, most importantly, my first quarter (tuition + housing + misc fees) should be payable entirely from loans and grants, so I don't have to stress too much about paying for anything other than my usual cross-continental relocation expenses.

All of which reassured me that if Brown follows up its institutional indifference with a rejection letter, I will still have a fine education awaiting me in September. Being away from Providence/Boston will kinda suck, but gods knows I'm likely to be back for graduate school, and my love is done with this joint next year.