Archive for October, 2004
I heard from squeakywheel about this event. Apparently, someone made some political comments that could vaguely be construed as threatening to the president, and an LJ user reported them to the Secret Service.
In objection to this, I ask that someone report me to the secret service. If people are being treated unfairly (by being hassled and frightened) by the government for their political views, I would like to do my part to make it difficult for the government to hassle and annoy everyone who is opposed to the current administration. So, please, someone report me to the secret service for the following paragraph.
I am going on the record as saying the following: George W. Bush is an awful president, and the world would be a better place if he had never existed. We would be better off without him. I want him out of office, and sooner is definitely best.
My server is actually getting hits!
I got an email from a random person asking about my dragon mask. Apparently, her son saw it online, and she wanted to know how to make it. I referred her to Elephant Pharmacy and offered to take pictures if she really wanted to make it. She said she did, so I took some pictures of the mask and put them on my site. I know it seems silly, but I’m glad to have an actual presence on the internet(s).
I also posted some of Louis’s noises to the guinea_pigs community. People are actually listening! I’ve only gotten a few comments, but people actually have listened to his little squeaks, and hopefully when they get home tonight, more people will listen and say, “No, that’s not cooing” or whatever is appropriate.
Kate and I had a tasty salmon dinner tonight, and I documented it all with photographs. You can see the whole series at my photoalbum, but I thought I’d include the basics here.
First, I bought a huge quantity (2 pounds) of salmon.
We then ate salmon
I can’t really believe that we’re eating salmon every week; it’s always been a rarity for me (we would have it only at Grandma’s). But it’s $5/pound, less than $1/pound more than chicken, and occasional fish is supposed to be very healthy. It’s not hard to cook, either.
A fark.com story used the tag for a jury that refused to indict a man who shot and killed someone trying to steal his car at night. There is discussion of it here (if you decide to look, brace yourself for an unthreaded discussion; it’s impossible to tell who’s responding to what). I’m upset because people are defending this guy and saying he had a right to do so.
I could go into the argument, and try to show what I think is right, but I don’t really have time to do that. But dammit, when people say it’s okay to kill someone for breaking the law, I want them to die the next time they break the speed limit. It would be poetic justice. The punishment should match the crime. I think it makes sense to be able to defend yourself, but killing people over property is not fair, right, or just.
I would love to rant for a while more, but I ought to go do productive things, like practice guitar or work on my class project.
I wrote recently about how my class project for Statistical Natural Language Processing is a hassle. I had a good idea, which I explained in my recent post, but Jerry Feldman (my possible advisor) basically nixed it. He favors a topic which is less interesting, and I don’t know whether I can contribute much to it.
So, today, I went and discussed this with Dan Klein, the professor to the class. He still thinks the project I proposed is fine; the issue is that it’s not 100% language-based, but he doesn’t care that much. As long as I have a good statistical model, it is relevant to the class. I’ll think about what the model should be, and I’ll meet with someone to discuss the other (less interesting) topic, and I’ll try to reach a decision. I’ll probably talk with Jerry Feldman one more time before I reach any conclusions, but I want to get everything straight before I go see him again. Anyway, I feel less depressed about it than I did the other day. It’ll work out.
If I am successful today, I will have read 5 papers: I started off my day reading a paper for a research group. I have two papers to read for class tomorrow. Then, I need to read a paper about some light research I’m working on with Ben and Ryan (which could turn into a Machine Learning class project, which would be good, since it hasn’t been terribly hard). Finally, I should read some stuff for my meeting tomorrow morning with Ryan and my Systems professor so that we can discuss that project. Oh yeah, and I should go over another paper that I need to meet about on Friday. So, if I’m really good, I’ll read 6. Since I’m posting here on LiveJournal, I’m probably not going to finish them :), but I’ll try.
Kate always has better, funnier posts than I do, and she always mentions new events before I get around to it. Of course, that has something to do with the differring purposes of our journals: mine is mostly a journal, while Kate’s is actually meant to be read by other people. I only have one LiveJournal friend, and I get to make her laugh at other times, so why work hard on being funny here? Still, I’m jealous.
Edit: I forgot to mention that she’s also more on-topic than I am. She told her readers about how we’ve actually set a wedding date, but I didn’t mention it at all. My family has been very excited, but I haven’t gotten around to actually writing it down here. It sounds fancy when you say that we’re going to be married at the St. Louis World’s Fair Pavilion, but it’s just a pretty place in a park. I think we’ve chosen a good location and a good date, though. I hope everything else works out, too.
I had what I considered to be a good Natural Language project idea, inspired by my lengthy debate with ilovethecure (which he has not been getting back to me about, lately; has he given up?). The idea was that I would analyze threaded comment schemes like LiveJournal’s, trying to extract information such as whether a comment is supporting or argumentative and whether its tone is polite or not. A major goal would also be to automatically segment the group into two classes by seeing who argued with whom and who supported whom.
So, I took this idea to my Natural Language Processing teacher. He thought it sounded interesting, certainly more so than the one recommended to me by Jerry Feldman, who I am considering as a possible advisor. Then, I took the idea to Jerry. He questioned the feasibility of the project, on the following grounds: Argument statements have complicated structure, so people use a variety of words, whether they are supporting, disputing, or just commenting. As a result, it would be very hard to extract meaningful data from such posts, so anything I could do in the month that I have would be trivial.
The problem is that the project Jerry Feldman recommended is also awfully hard, and I don’t think I could make much of an impact. I really don’t have any idea of where to start. He did find someone particularly knowledgeable in the field (a visiting professor who has worked on the problem before) for me to talk to, though, so there’s hope that I’ll find a part of the problem to address. I guess I should talk to him before I make any final decisions, but I am left feeling pretty hopeless about the whole thing. The project that I really liked just got, well, dismissed, and the project that remains looks hard and boring. It’s not that the dismissal was wrong–there are serious questions to ask about its feasibility–but that I don’t feel like there’s a good alternative. Besides, my project idea looked like fun.
I’ve been getting along okay this semester, but now I have to really start reading. Reading is the key to having a good research project, and I’ve got two of those to work on. Reading is the key to doing well on my prelim examination next fall. I know that seems a long way away, but I need to read two textbooks and many papers before then. Reading is also important for the research group meetings I’ve been attempting to attend. Those meetings could help me find a long-term advisor or thesis topic. Plus, there’s plain old class reading. This emphasis on reading is definitely something I didn’t expect coming here. My woes are exacerbated by the fact that research papers don’t make easy reading; there’s so much technical information that it requires careful reading and a lot of thought.
I also have made the mistake of getting myself embroiled in an argument with ilovethecure here. He’s not terribly good at debating, but he’s incredibly tenacious. He will not let anyone have the last word, and he seems to have more free time than anyone with any kind of work would have. Despite that, he’s not too keen on reading what I have to say (see here). It’s really annoying, and it’s not productive. As much as he writes, he doesn’t make particularly strong points, and I’m never going to convince him of anything.
Oh. I recently saw the music video for “Weapon of Choice” again, and it really amuses me. The music is good, and Christopher Walken’s dancing is particularly amusing (he even does the “T-Rex”). I’m embarassingly amused, to tell you the truth.
On Sunday night, Kate and I watched the original Superman. I asked her if Christopher Reeve was still alive, and she assured me that he was. Then, on Monday, she saw a news article saying that he died on Sunday from an infection. Apparently I’m an evil genius, killing Superman with a single sentence. Or maybe, like fairies die when you say you don’t believe in them, Superman dies when you ask if he’s dead.