Pat Gunn (dachte) wrote,
Pat Gunn

To Preside

  • Tomorrow is 「Pretend to be a Time Traveller Day」.
  • A good presentation on Argument maps (I talk about how awesome they are every so often) as one approaches a mature product to handle them. We've reached the MS Word 6.0 level of them, where we basically know what's needed to do them well enough for broad deployment - the next big step in the topic is very difficult (perhaps impossible) - handling the reasoning based on the facts. Argument maps as they are now are great to help us discuss facts, and ok to help us manage certain kinds of arguments. generalising to other types of structured arguments and reasoning methods would be very ambitious and fairly difficult.
  • I got around to seeing Obamaのspeech_to_schoolkids. It was a kind of diluted-strength version of Bill Cosby's famous Pound Cake speech. It would be hard to dislike the content (Newt Gingrich called it a great speech) - apart from some elements in popular culture, it argued for what most people believe - that education and diligence are key to a successful life, and that a nation is successful to the extent that its people have these virtues. It's likely that Obama wrote most of it himself, and as a nation we're getting used to have a president who's a great speaker. I'm impressed as well, although I'm not sure it's the way I should be impressed by a national leader. I've commented before on the difference between the perception of politicians-as-representatives (chosen as blank slates upon which the people write) and politicians-as-chosen-statesmen (chosen by the people because they feel some kinship with the expressed ideas and style of thought). More importantly, this is a distinction between politicians as role models/cultural figures and as administrators. I lean fairly heavily towards thinking that politicians should be excellent administrators - I don't particularly need heroes or role models. Good government should be efficient, comprehensive, high-quality, and so noncumbersome that as a topic it slips readily from my mind. The mail arrives in a timely fashion, the roads are in a good condition, the educational system produces generations of educated, skeptical, well-read, curious people, the military is professional in their handling of our national security, etc. Do we need leaders who write great speeches, inspire us, help us see the world in a new light? Not so much. However,
    • I may be wrong about this - maybe individuals who can inspire dreams of a nation are necessary to help us "buy in" to society, government, and each other. When we stop caring about each other and become too selfish and cynical, a great speech might be able to pull people back together. I might not need it myself so much (I subscribe to a metaphor by Thomas Paine, 「The World is my country, all man and women are my brothers and sisters, and to do good is my religion」), but others may
    • I may also be wrong in another fashion - in a two-party democracy where there are people who define themselves by an intent to destroy the state and a belief that all societal institutions that are part of the state are inherently broken and not worth trying, speeches lay the ground on which political battles are fought. High political figures either are captive to the situation or take control of it, and so great speechwriting is a definite tactical asset even if it is not an intrinsic part of the job definition.
  • Faraday Cage - Faraday cages are pretty cool. In my prior job at CMU in Psychology, the neighbouring group built one for some neuroimaging experiments, using chickenwire-type enclosures. In this job, the ECE group we're affiliated had a large bankvault-style cage built for our wireless experiments (noise filters for the power lines and optical switches were installed). This is probably the nicest faraday cage I've seen (photos forthcoming). (P.S. for every possible topic, there exists a kook)
  • Working with this giant crazy codebase at work, I have come to the conclusion that the code is not all that terrible (even if many parts are pretty bad), the biggest problem is that there is close to zero systems-level documentation for it and it's far too big for anyone to keep most of it in their head. Some things that looked like facepalms seem a lot more reasonable now that I've spent the time to learn it. The problem is - one way or another I'm not going to be at this job forever, and neither will the individuals at University of Utah that cooked parts of it up. Anyone dying, changing jobs, retiring, etc. would cripple the project far beyond what's reasonable (I should note that people with the systems experience combined with the programming experience I have are not the most common people in the world, even if they're not super-rare).
  • There was a group meeting this Wednesday where we covered two papers, one was Baumann et al's 「The Multikernel, A new OS architecture for scalable multicore systems」, the other Porter et al's 「Operating System Transactions」. Both were pretty interesting (not as interesting as neuropsych - Oh how I miss group meetings talking about neuropsych topics in my old job!). Baumann's paper was about the future of hardware architectures and how to design OS's to best handle his predictions. The arguments for why were sloppy and made me want to throw things, the designs were pretty cool. Porter's paper was much better - it talked about desirability for introducing a transactional interface into generic operating systems at the syscall level, showing a Linux variant that did this (some syscalls are not transaction safe, including obviously the network ones, but disk and memory related syscalls are both supported). This is hellishly cool, and as the paper notes, offers strong protection from race conditions. There was another paper mentioned in our discussions that I should find relating to operating systems that do heavy speculative work as a basic service.

Given the awkwardness of self-massage, my neck and shoulders are sore enough that I am convinced my head would come clean off if it actually were massaged properly. I think my occasional jogs in the evening contribute to this. I would think that I could use a running partner, but my social life is too full of failure and I'm sure anyone serious enough for that would be faster than me by a good bit. For me, jogging is just something to do, not something to be competitive about.

I might check out Tazza d'Oreo tonight. It's quite a hike from where I live - I will probably bus there, but getting back might involve a walk depending on how the busses work. It looks like a hip, liberal place, and like the Beehive, it might add a regret to my not having a more timely way to get around town.

I've been thinking about the G20 demonstrations recently. Unlike a lot of other things that might be worth protesting (animal testing, chemical weapons labs, etc), we have a strange dependence on something we feel is inherently rotten (capitalism), and our protests thus lack the bite of something we could simply demand be shut down (I realise that banning most animal testing would have a societal cost, but for many kinds of testing, I hold that we should just bear that cost because our morals are more important (I hold this strongly enough that I approve in theory of ALF's actions to sabotage most animal testing labs)). We cannot simply demand that the economy stop, nor can we say that economic planning and coordination that happens at G20 is not necessary for our well-being in some ways. Even those of us who are committed to an alternative model (socialism, in my case) cannot demand an instantaneous change. This understood, what are the suitable goals for G20 demonstrations? Is direct action particularly undesirable for this event? Is there anything direct to protest at all?

I don't know if I'm going to the demonstrations - will have to think about this some more, and see if I'm feeling ok that day.

  • Still alive

    Been feeling a bit nostalgic. Not about to return to LiveJournal - their new ownership is unfortunate, but I wanted to briefly note what's been up…

  • Unplugging LJ

    It's about time I pulled the plug on the LJ version of my blog: 1) I'm much more active on G+ than I am with general blogging. I post many times a…

  • Mutual Trust

    I don't know which should be considered more remarkable: That a cat should trust a member of a far larger and stronger species that it can't…

  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

  • 1 comment