The call for papers for SoCG 2009 is out... and the deadline is December 1st (with abstracts due earlier). It's in Aarhus, Denmark, so not one to miss (as opposed to lovely College Park). Now if only I can do something with my half-baked idea in time...
A new $44 million educational research effort is beginning. The premise behind the new Educational Innovation Laboratory is that educational programs have not been subjected to enough rigorous testing. This, I agree with. However, given the need for large research projects on fundamental educational issues (as implied by the need for this new laboratory to begin with) the fact that they're planning to begin by studying incentive programs like paying students for good test grades is absurd. How about starting with some actual fundamental educational issues like, say, teacher pay? And if the money is only going to be used to study whether students learn better if they're given more cell phone hours when they get good grades, please just forgo the research and give the money directly to the schools - there's already plenty of research showing that will actually make a difference.
A friend of mine from college works for a cool new website, Unigo, which recently got some excellent press from the NYTimes Magazine. I have to admit that though I'd heard something about the site before the article, I didn't really get it until I read the article and saw the site live. It's one of those brilliant ideas that you know is truly smart because you wonder why no-one's done it until now. It's a Princeton Review or Insider's Guide type college guide book, but online. And current and past students from the colleges can submit videos, photos, and reviews. Editors (like my friend) then compile the reviews into a shorter summary, but all the original info is still around. It's the perfect use of the overwhelming-user-generated-content web.
Now I'm hoping that the next thing on their list is a branch of the site for grad school. Grad school choice is still confounding and precisely the type of hugely broad area where user-generated content is perfect. True, there's a lot of variety in a grad school program based on your area and advisor, but everyone takes the same core courses and lives with the same administration. For Maryland I'd comment that two of the best things about the department's course decisions are that our qualifying exams are wrapped into our course requirements and that many courses have large original research projects as requirements. One of the things that doesn't work as well is our proposal system which wraps together an oral exam on a reading list and an hour talk on your research project - I'd much prefer they were separated so I could focus solely on each.
The New York Times has relatively regular articles about mathematics education. Often they're anecdotal rants by parents against "new math," but this week's article is an interesting overview of some recent research into innate number sense. It's this number sense that they describe as allowing you to determine which grocery store line is shorter at a glance, or whether there are more blue dots or yellow dots on the screen (see the online version of the test the researchers used). Researchers determined that innate aptitude for this type of approximation correlated to stronger formal math abilities.
The article suggests that teachers should take this as a cue to incorporate more estimation and general understanding into the curriculum (though the research does not show if our inborn number sense can be improved). I strongly agree with this and tried to do a lot of estimation when I was teaching middle school math. It's very hard to teach, and very hard for some students to understand. Perhaps this begins to give some clue as to why.
Sadly, my AT&T Labs Fellowship ends at the end of this school year. It's excellent and I'd love another, so I'm on the hunt for other fellowships. I thought I'd share the list I'm compiling.
All fellowships are open to students in Computer Science PhD programs. Also, these don't include fellowships with required after-graduate school participation (unless I've missed something), but do include fellowships with required (or strongly encouraged) internships.
IBM Fellowship - Deadline: Open for nominations September 22, 2008 through October 31, 2008 (nomination by faculty member required). Eligibility: completed at least one year of a doctoral program. Award:tuition, fees, and stipend for one year (amount varies by country/region).
Google's Anita Borg Scholarship - Deadline: check back in October 2008. Eligibility: entering senior year of undergraduate study or be enrolled in a full-time graduate program for 2008-2009 with at least a 3.5/4.0 GPA. Award: $10,000
SWE Scholarships - Application information for 2009-2010 will be available in December.
AAUW Fellowships - Many fellowships; one is a dissertation fellowship for which eligibility is restricted to those who have "completed all course work, passed all required preliminary examinations, and received approval for their research proposal or plan by Nov. 15, 2008."
DOE Computational Science Graduate Fellowship- Deadline: check back in October 2008. Eligibility: US citizen or permanent resident, first or second year of graduate study and planning full-time PhD study. At university that doesn't require non-research activities (i.e. teaching). Award: 4-year yearly stipend of $32,400 and tuition and fees and additional money for conferences and computers.
While all the market fluctuation is going on seemed like a good time to point out a cool tool - the map of the market. It makes it easy to visualize the market ups and downs. Yesterday it showed that tech industries were still doing well, today it's pretty much universally bad. It's based on the treemap idea by Ben Shneiderman at Maryland.
When walking by the Columbia Heights Metro in Washington DC this morning I noticed this Maus-like figure looking into a trash can as police looked on. Apparently it caused metro delays as police investigated to make sure it wasn't a bomb (it wasn't). To me, it just looked like cool opinion art. Maybe a commentary on homelessness? Or, if it's a polar bear, global warming?
Roger Cohen had an op-ed in the New York Times last month about how good news from Africa is buried while bad news is, well, news. He also mentioned that Ghana, where I lived for a semester in high school and again in college, is one of the places in Africa that's experiencing somewhat of a boom. Indeed, when I was there in 1997 there were only a handful of stoplights in the capital, Accra. Later, in 2003, there were too many to count. Their government has also been undergoing positive changes. In 1997, Jerry Rawlings was president, and had been since 1981 when he seized the country in a coup. But by 2003, John Kufuor was president after Rawlings peacefully relinquished power in 2001 through democratic elections - the first such transition in the country's history. Kufuor will make history again this coming January by peacefully handing over power as well.
So it seems appropriate that now, while Kufuor is visiting president Bush to discuss good news which is buried by the horrible news of the market (to the extent that this was true even in their joint press conference), we take note of the amazing progress that has been made. For once Bush said something true when welcoming Kufuor to the US, "Your country is a model of entrepreneurship and democracy and peace on the continent of Africa."
In a discussion with a friend I brought up the "slippery slope" as a partial justification for my reasoning. She smiled, somewhat triumphantly, and pointed out that the slippery slope argument is illogical. After all, we draw a line in the sand and say that anything beyond it begins a slippery slope, but why is the line in that position to begin with? We can just move the line and hold firm there. There is no philosophical justification for the possible maintenance of one abstract line and not another.
She's right, of course, yet in the non-theoretical world when this concept is applied, for example, to freedom of speech, I believe the illogical. Or perhaps there is a real-world constraint we are ignoring. I agree with the ACLU that the KKK's first amendment rights should be defended. The KKK speaks out against me and everything I believe in, yet without their right to insult me, I might not have the right to protest in return. Once one exception is made, the door is opened to others - the slippery slope. And so I illogically maintain that to prevent Palin from banning books, we must also support her right to speak at the Alaska Independence Party convention and our own right to vote against her.
I'm like a deer in headlights when it comes to the news about Palin - there's so many options of what to write about I can't write about anything. So instead I'll say that Palin's views aren't just scary to me, they're objectively extreme. Here's a selection - judge for yourselves.
I'm starting a new research project which looks like it will combine HMMs, Information Theory, Sensor Networks, and Image/Video Compression. (Yes, I am still in Computational Geometry.) If any of you know of good overviews of these topics, I'd love suggestions... especially suggestions that come in the form of papers instead of large books.
One of the tactics we mentioned to the future TAs and which every teacher (should) know is to admit when you don't know the answer to a question, promise to find out the answer and get back to them, and move on. This is the best strategy partially because if you try to talk around it, it will be painfully obvious that the true answer is "I don't know and know that I should." Politicians, however, seem to have a hard time doing this. To be fair, it's often more important for them that they look like they know everything... but it could be argued that this is even more of a reason not to try to talk around the issue (and in doing so look like an idiot). The latest example of a foot-in-the-mouth performance is by one of McCain's staffers Tucker Bounds. When talking to a CNN reporter, he blatantly avoids answering her question about what specific decisions Palin has made as commander of the Alaska national guard. Tucker can't answer and won't say he can't answer, so he ends up looking as expected in that situation. CNN airs the interview (um, clearly) which makes the McCain "don't ask me hard questions" camp whiney and they pull a later interview with Larry King. If they weren't running for president by saying they can handle the hard questions, it'd be funny. No, wait, it's still funny.
Friday was the TA orientation for our department. It turns out we only had one hour, so despite our hopes to follow the wise suggestion of an earlier commenter and have the future TAs practice teaching in-front of each other, we ran out of time. Instead we modeled interactive lecturing and group work. I think it went better than last year's though as it was only an hour it certainly wasn't complete. I later met a fourth year student who's new to our department and was previously in the math department. She mentioned that the math department has a 4 day TA orientation. So it seems the CS department cares about this about one thirtieth as much as the math department...
One of my personal goals for the hour we had was to encourage the TAs to have their students do group work. I got to get my support of group work into the mix, so I was happy about that, and it helped me to discover the extent to which I am totally reliant on and in favor of group work. In fact, while I believe many people find encouraging group work daunting, I find teaching a class in which the students are only listening to a long lecture from me daunting. Group work always provides a nice context switch for me as well as the students - I feel less like I'm "on" when I'm walking around and helping students in smaller groups or individually. Good for me, good for the students - I'm baffled that everyone doesn't teach this way.