Pat Gunn (dachte) wrote,
Pat Gunn

Frozen in the Middle of Collapse

My new apartment is solidifying into a home, aided greatly by the fact that my DSL is, despite theoretically still having some appointments in order to be fully installed, unexpectedly working, and by my having assembled my futon today. I think that I'll actually start sleeping on it again -- it's actually hard enough to be comfortable for me. Perhaps my reluctance to do so before was because I didn't like going into the back parts of my old apartment. My new, smaller apartment doesn't really have anywhere like that. Feel-wise, it's more like my beloved 97A W. Northwood Apartment, which was one giant three-level room. The cats are comfortable here now too, and Tortfeasor has figured out not to jump down behind the stove. Picture, as promised:

I think, while in BioInformatics class today, despite my severe undersleeping, I easily made it through the class because they were using the side lighting on the room instead of the overhead lighting. For some reason, the torchlike side lighting does not tire me as much. It also helped that the prof mentioned, near the end of class, an idea that practically set my mind on fire -- Markov Random Fields. When I learned about Markov Chains as an undergrad, I thought they were clever, and tried to extend them to three dimensions, without success. Markov Random Fields are only kind-of a 2d extension, in that they ditch some of the context (sequentiality) that isn't very compatible with that expansion. I hope, when I get more time, to write some code to implement and play with the concept.

Part of my brain is still digesting the conversation last night. That's how one knows one had a really good conversation -- they inspire thought even when over.

Two people I don't know particularly well got married recently, and two others I used to know fairly well are getting married in a few months. Hmmmmm. Oh, and then of course there's another pair that I also know and see more often that are getting married in a few months.

In a wonderful welcome-to-the-future moment, I was trying to anonymise my brain data in order to give it to my BioInformatics prof (not that I mind my name being on there, but there's other info that didn't belong). I discovered that, much to my joy, a brain analysis tool I've been using for ages, XMedcon, has a badly broken anonymisation tool. This is broken in the sense that, despite in one of its modes even prompting to replace the data I wanted replaced and providing the old values as defaults, it completely fails to update the file (or even update the timestamp), no matter what arguments I give it. No errors or warnings, even in verbose mode, it simply doesn't do anything. Good job, guys! Having recently had success with it, I whipped out vim -b and edited binary data to anonymise it myself. If any of you want my brain data, it's available in Analyse format here.

Matlab/R/Maple/Octave/Mathematica are all very similar. It's hard for me to tell the difference apart from silly little naming and library differences. Meh.

Moving is really helping me throw stuff away that I don't care about. This is a good thing.

And now, off to Giant Eagle to get cat and person food... :)


  • Typing in Colours

    (Cross-posted to G+, but it's more of a definitive statement of views so it goes here too) A recent instance of 「Wasted Talent」: here I'm not…

  • Loyalty

    This is meant to address three ideas: Don't blame the victim If you care for me, you'd support me unconditionally Safe zonesAnd to be a topic in…

  • What Do We Owe Each Other?

    One of the central questions in political philosophy, or perhaps one of the most intuitive initial framings, is "what do we owe each other?". I…

  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded