Saturday, September 17, 2005


This guy transcribed two tracks off DJ Shadow's Endtroducing for the high school percussion ensemble he directs. They worked on them for many months and then put on a concert. You can see videos of their practices and the performance here. I would have joined the high school band if we had gotten to do this. Well, maybe.
(via Metafilter)

Wednesday, September 14, 2005

Roberts Reveals His Favorite Movies

(via Wonkette)

Sure, why not?

Now we have LibraryThing, which is "like Flickr for your books".

I'm not sure what the point is yet but that didn't stop me from cataloging the books on my shelf at work anyway. You can see them here.

I will say that the search engine you use to find the books you want to add (it queries Amazon and the Library of Congress) works remarkably well.

And yes, they let you use tags too.

Monday, September 12, 2005

Hopefully, the social music revolution will be a relatively bloodless one

Via a plugin for your music player, keeps track of the songs you have been listening to on your computer and uses that information to update a "profile page" that contains a list of the songs you have listened to most recently and many "charts" such as "Weekly Top Artists" and "Top Tracks -- Overall." It uses this information to find "musical neighbours," or people whose listening data are similar to yours, and create a "neighbour radio" station with tracks you haven't been listening to that it thinks you might want to listen to. Presumably the playlist for the neighbour radio is generated from the set difference between your neighbors' tracks and your tracks, probably ranked by an average relative frequency somehow or something.

Anyway, a kind of neat addtion to the "social everything" bandwagon (yes, they have tags too), and what better way to find out what songs you actually like than to keep track of what you listen to most often?

My profile page.

Stupid product alert: Verizon ringback tones

Verizon now offers what they call "Ringback Tones." This feature allows you to select a custom ring for your contacts to hear back over the line when they call you. For this service, they charge you $0.99 per month and $1.99 per year per ringtone. I know this isn't that expensive, but still, do people really think this is worth any money at all? Will your calling experience really be enhanced by a significant amount if you get to hear 1-4 band-limited two-second clips of the latest pop song before the person you are calling picks up?

We're all mentally ill in our own special way

According to this, I am both schizoid, which means I "genuinely prefer to be alone," and avoidant, which means I "yearn for social relations." So are two mutally contradictory personality disorders worse than one, or do they just cancel each other out?

Paranoid Personality Disorder:Low
Schizoid Personality Disorder:High
Schizotypal Personality Disorder:Moderate
Antisocial Personality Disorder:Low
Borderline Personality Disorder:Low
Histrionic Personality Disorder:Moderate
Narcissistic Personality Disorder:Low
Avoidant Personality Disorder:High
Dependent Personality Disorder:Low
Obsessive-Compulsive Disorder:Low

-- Take the Personality Disorder Test --
-- Personality Disorder Info --

I say this because, ...

... inspired by this guy, I've been thinking it might be useful to start a research blog. In getting started in my thesis lab I've found that many of the ideas I want to bounce off my PI really only crystallize fully when I am actually writing an email to him. Just going down the hall and letting ideas come out of my head for the first time via my mouth is often not a good strategy, as upon escaping to the outside world in that sudden manner they often tend to mostly engage in a lot of ineffectual flailing about.

No, for now at least ("for now" because this is probably an issue of practice, and if I keep working at it I might someday be able to work things out in my head just as well as on e-paper), I think it will be productive to write out potentially useful trains of thought fully when working. And since blogging software allows one to easily keep a journal on a computer, a blog seems like an obvious place to put research notes. (A wiki would be another option.) Plus, the exhibitionist thrill of having my thoughts in a format designed to be accessed by others could be an additional incentive to keep up with my writing (ok, so maybe that incentive hasn't done much to help me keep up this blog.)

An issue to consider would be whether I would be violating any "no prior publication" rules of any journals we might want to publish in (if, God willing, my projects are successful) by keeping a work diary online. It's somewhat sad to have to think about this, but the biological sciences really do seem to encourage secrecy before publication. In theory, one's productivity could be increased by making the whole process public, but that's just not the way our field works right now. (Exposing your entire process to the public could of course also increase the productivity of other researchers--and thus the field of the whole--for if everyone had access to everyone's ideas, duds as well as gems, one researcher might find in the idle, never to be published thoughts of another researcher a solution to one of his or her problems.)

Is there a snowball's chance in hell of ever getting neuroscience to be this open? Could we ever convince people to stop hiding things and start turning the web into a vast collective neuroscience thinking machine? Can we restructure the field such that everyone collaborates and shares with everyone every single day? (I mean, when you tour grad schools, everyone talks about their love of collaboration. But if they love collaboration so much why do they limit its scope to a handful of their peers?) Can we agree to start preserving all ideas, good and bad, in case the bad ones turn out to be good ones later?

Sigh. We're a long way away from neuroscience working like this. It seems like CS and physics communities do function more like this, and that's probably because a much larger percentage of the members of those communities are computer geeks who have internalized the various FOSS philosophies.

(This is a thought many people have had, but think of the labor that could be saved, too. As AM brought up yesterday, if you spend a couple months doing experiments to test an idea and they don't pan out, you don't publish it and move on to something else. This leaves the failure of that bad idea as an unpublished secret. Does anyone benefit from having people try the same bad idea over-and-over again?)

So the moral of the story is that if I start a research blog, I'll probably have to restrict the IP space from which it is accessible. That's not because I think I actually have great ideas, or that anyone would read a research blog about my obscure project, but merely because if I have to sign a form prior to publication swearing that I have not published my thoughts elsewhere, and that the internet counts as "elsewhere," I want to be able to sign it.

Sunday, September 11, 2005

What is needed:

An interface between Blogger, Moveable Type, whatever, and LaTeX. It would be nice to be able to throw down equations LaTeX style and have them typeset all pretty and either rendered in images that would be automatically uploaded and <IMG SRC>'d into your post or added as html through a link with latex2html. Then one could rapidly put equations in a blog post and have them look the right way--or close to the right way, if you go the latex2html route.