Friday, December 17, 2004

Why don't *I* get credit for Blogging?

Students who register for "Internet and Society" at Northwestern University this winter are going to be required to maintain a blog. Over at Crooked Timber, Eszter Hargittai describes her ideas and motivation. The course will focus on "the social, political, economic and cultural aspects of the Internet." The first two lines of the course description: "What's it like to maintain a blog? How about if you are in Iran or China?"

I found it interesting that several of her commenters say that they have been using blogs for coursework for a while now, and that students seem to respond well. Extra credit for blogging might make me post more regularly. (Don't you blog enough already? You have a final in two hours! - ed. Yes, but I can't disappoint my adoring public. Derisive laughter from ye ed.. Ok, ok, point taken. One last comment and I'll go.)

There was a rumour floating around that Time magazine was considering naming 'The Blogger' its person of the year. (Unfortunately, it now appears that they've decided to go with George Bush.) What is the world coming to?

Monday, December 13, 2004

Soldiers accused of War Crimes

This Washington Post article is horrifying. I sincerely hope the soldiers accused are innocent of any misconduct; but it seems unlikely. As a longtime fan of the US military, I'm extremely depressed by this turn of events.

Interestingly enough, I was thinking about the U.S. Army a couple of days ago. I read an article which casually mentioned 'Col. McMaster' as commander of the 3rd Armored Cavalry Regiment. I was curious to see if this was (then Capt.) H. R. McMaster, who commanded Eagle Troop at the Battle of 73 Easting in the first Gulf War. He was then in the 2d ACR, and I distinctly remember that when I first read about the battle a few years ago, I wondered if he would ever command the regiment. I checked, and it's the same McMaster; I was nearly right, but he's commanding the 3rd ACR instead of the 2nd. For some strange reason, reading about his promotion made my day.

The mind is a remarkable thing; there was no reason that I should have remembered McMaster's name, still less that I should be so pleased about his promotion. What's even more strange is that I have a terrible memory for names in general, but I remember from an interview I read (years ago) that his wife's name is Katie! And yet I often struggle to remember the names of authors of papers I read months ago.

For everyone surprised by my interest in things military: Several years ago, I developed an interest in first small-unit tactics and then the intellectual challenge of developing strategy for larger units in wartime, when information is often unreliable or incomplete. In particular, Armored Cavalry fascinated me because of their emphasis on combined arms at all levels. I've always been impressed by U.S. equipment, training and doctrine.

And for the curious, the Battle of 73 Easting was one of the key actions in the Gulf War of 1991. Eagle Troop of the 2nd ACR, led by Capt. McMaster, made contact with the Iraqi forces and began an assault. With the support of two other troops, they destroyed an entire brigade of the Tawakalna division of the crack Republican Guards. U.S. losses to Iraqi fire amounted to one vehicle and one soldier. The battle is regarded as almost a textbook example of small-unit operations, and has been re-constructed in almost every detail by DARPA for use in simulations.

Saturday, December 11, 2004

Yushchenko Definitely Poisoned

I hadn't blogged about the Ukrainian presidential election results before, because I didn't really have anything to say. There's been lots of media coverage; Dan Drezner has a good roundup. In a nutshell, the official results declared a victory for the departing president's protege, Viktor F. Yanukovich. There were widespread allegations of fraud, and supporters of Viktor A. Yushchenko,the opposition candidate, paralysed Ukraine for several days. Given that most (Western-run) exit polls showed a double-digit lead for Yushchenko, most outside observers are convinced that the polls were rigged.

Just to make it more interesting, Viktor Yushchenko fell seriously ill during the campaign. The mystery illness changed his appearance completely; he looks haggard and worn now, while he had been known for almost-movie-star looks earlier. (CNN has good before/after pictures here.) Yushchenko alleged that he had been poisoned; his detractors claimed he had eaten bad sushi. Today, the Times is carrying a story from the International Herald Tribune confirming that Yuschenko had been poisoned. Tests conducted in Vienna indicate that this was a severe case of dioxin poisoning.

I'm not sure what this will mean for Ukraine; I hope it doesn't endanger the agreement to conduct new elections.

Thursday, December 09, 2004

All Your Crustless (PB&J) Sandwich are Belong to Us

U.S. Patent No. 6004596 gives Smucker's broad protection on its "sealed crustless sandwich". In 2001, Albie's Food Inc., a small grocery company selling pastries and sandwiches, received a Cease and Desist letter which accused them of violating Smucker's Intellectual Property by selling crustless peanut butter and jelly sandwiches.

IEEE Spectrum is carrying an article which comments on the woeful state of the U.S.P.T.O and suggests fixes. Their ideas are reasonable and will probably improve the efficiency of the patent office while decreasing the number of frivolous patents issued. I suppose that means they have a snowball's chance in Hell of ever being implemented.

Why, yes, I'm feeling cynical today. However did you notice?

Tuesday, December 07, 2004

PISA 2003: Survey Results

The Organisation for Economic Co-operation and Development (OECD) has conducted it's second study of learning skills among 15-year olds. The study (called PISA, to be conducted every three years) was first performed in 2000, and then again last year. Results from PISA 2003 have just been released by OECD. The main focus in this study was on mathematics, while PISA 2000 was only a reading assessment. Over 250,000 students from 41 countries participated in the study; countries that were not OECD members could also choose to be included.

Finnish students were overall the most proficient, though four other countries (Korea, Hong Kong, Liechtenstein and Japan; all in the top 8 overall) had a greater percentage of students at the highest proficiency level. Incidentally, Finland had also led in the reading assessment of PISA 2000. The U.S. was 28th out of 40 countries, just behind Latvia and ahead of Portugal and the Russian Federation. Interestingly, Canada has the third highest performance overall.

The full report is available as a PDF from the OECD website. So far I've only managed to skim through the entire 471-page report; I would read it through except for the fact that I was meant to be working on a unit project for the last 2 hours. They have some obvious conclusions ("Both students and schools perform best in a climate characterised by high expectations [and] supported through strong teacher-student relations"; "Students whose parents have better-paid jobs, are better educated ... perform on average significantly better in all countries than those without such advantages") and some interesting ones.

From the PDF or OECD article, some things that struck me were:
1. Australia, Canada, Finland and Japan stand out for high standards of both quality and equity, with above-average mathematics performance and below-average impact of socio-economic background on student performance.
2. Poland had a dramatic variation of performance between schools in the original study; this has shrunk drastically after the school system was integrated in the intervening period.
3. Most countries have more boys than girls among top mathematics performers, resulting in a slight overall advantage for boys in average terms. On the other hand, boys and girls tend to be equally represented among the low-performers. It's interesting to note that girls attend the higher performing, academically oriented tracks and schools at a higher rate than boys but, within schools, girls often perform significantly below boys. Girls also consistently report lower interest in and enjoyment of mathematics than boys.
4. Better performance was often related to an enjoyment of mathematics, but perception of ability was not as strongly correlated to ability as one might expect. About a third of US children (ranked 28th overall) did not feel as though they were good at math, but nearly two-thirds of Koreans (ranked 2nd) felt the same. (Editorializing: Has worrying about students self-esteem hurt academic performance? Increasingly, schools are refusing to differentiate between students for fear of hurting the image students have of themselves. Readers Digest recently ran a story on how schools are refusing to recognise true merit; some high schools had 50 to 100 valedictorians! In Nashville, one high school principal was told he couldn't release the names of high scorers at basketball games. Another school couldn't announce the winner of the spelling bee! But I digress... I'll leave the subject for a future post)
5. The US also seems to have the poorest outcomes per dollar spent on education. The performance in reading (18th out of 40) is better than that in math, though. "While spending on educational institutions is important," the report says, "it is not sufficient to achive high levels of outcomes." (I wish more administrators would realize that throwing money at the problem wouldn't work!)

Unfortunately, India isn't rated. I'm curious about how Indian students would have performed. I suspect that the reading assessment results would have been abysmal, but the mathematics results very good. This is entirely due to the fact that only students still in school at the age of 15 participate in the survey; the majority of Indian children have dropped out of school by this time, if they ever attended. (It's technically illegal in most states to not send children to school, but this is rarely, if ever, enforced.) Glancing through the questions, I would be shocked if the average 15-year old Indian still in school couldn't solve at least half of the problems listed as most difficult (Level 6). (See my previous post on high-school math curricula.)

From the section titled "The PISA approach to Assessing Mathematics Performance":
PISA therefore presents students with problems mainly set in real-world situations. These are crafted in such a way that aspects of mathematics would be of genuine benefit in solving the problem. The objective of the PISA assessment is to obtain measures of the extent to which students presented with these problems can activate their mathematical knowledge and competencies to solve such problems successfully.
This approach to mathematics contrasts with a traditional understanding of school mathematics which is often narrower. In schools, mathematical content is often taught and assessed in ways that are removed from authentic contexts – e.g., students are taught the techniques of arithmetic, then given an arithmetic computation to complete; they are shown how to solve particular types of equations, then given further similar equations to solve; they are taught about geometric properties and relationships, then given a theorem to prove. Having learned the relevant concepts, skills and techniques, students are typically given contrived mathematical problems that call for the application of that knowledge.

Indian (and Asian, in general) schools usually use the latter (traditional/narrower) approach, in contrast to many Western schools which teach mathematics the way PISA tests it. So I find it strange that Korea, Hong Jong, Macao and Japan would be in the top 8, and I believe India would be up there with them. (Assuming, of course, that questions were appropriately translated into the local language and cultural context; I can see the average Indian student having difficulty on the question about skateboarding.) Perhaps the PISA test doesn't measure what they think it does, then.

The New York Times story had some interesting insights; they've obviously spent a lot more time analysing it than I have... With luck I'll get to it after Finals week. (Don't forget all the exams you'll have to grade! - ed. Siiigh - you had to remind me, didn't you?)

Sunday, December 05, 2004

Word-of-Mouth Advertising

I was taking a break from an assignment and saw this article on Slashdot, based on a New York Times magazine story. (Warning: 9 pages!)

Convinced that everyday conversation is the most powerful medium for consumer seduction, companies have begun organizing advertising campaigns based solely on word-of-mouth publicity, and results have been fairly good. Whether you think this is downright sleazy or the coolest thing since the GPL (Slashdot is having a hard time making up its collective mind), the NY Times article is well worth the read.

Sunday, November 21, 2004

Give Thanks!

I suspect that the thing I'm most grateful for this year is the fact that we get a week off for Thanksgiving; I need the break! I'm looking forward to seeing Boston and New York and meeting family. Photographs will be forthcoming.

Oh, and I've always been curious about this: What's the origin of "the Bean and Cod"? I googled it, but was surprised to find no useful links except for a poem by William Corbett. I'm too lazy to look it up, so I'd be grateful for a reference.

Updates from Iraq

The Iraqi Electoral commission has confirmed that national elections will be held on Jan 30th. A commission spokesman, Farid Ayar, said that violence-prone areas like Falluja and Mosul will have elections at the same time.
"No Iraqi province will be excluded because the law considers Iraq as one constituency, and therefore it is not legal to exclude any province," he said.

In other news, the Post reports that American forces have found the houses where several hostages were tortured and killed in Falluja. The city appears to be slowly coming under American/Iraqi control, but an increasing number of resistance fighters are using white flags to pose as civilians and then attack deceived soldiers.

One can only hope that elections bring some peace to the country, but it seems doubtful.

Tuesday, November 09, 2004

And finally...

Firefox 1.0 has been launched. Download it today!

Sunday, November 07, 2004

Fighting in Fallujah

Not that it wasn't expected, but Fallujah is under attack. What's most disturbing is that Iraqi Prime Minister Ayad Allawi has declared a state of emergency, not exactly the best start to democracy. Quoting the Times and Post:
Hours earlier, Prime Minister Ayad Allawi, faced with an expanding outbreak of insurgent violence across the country, formally proclaimed a state of emergency for 60 days across most of Iraq. The proclamation gave him broad powers that allow him to impose curfews, order house-to-house searches and detain suspected criminals and insurgents. The order will run for 60 days but could be extended through elections planned for January.

An Eternal Golden Braid

I'm reading (for the first time, much as it pains me to admit it!) Godel, Escher, Bach: an Eternal Golden Braid by Douglas R. Hofstadter. I had meant to read it on many previous occasions, but never got around to it; now, that seems like criminal procrastination. So far, it's a superb book; the author's idea to intersperse dialogues with chapters was nothing less than inspired.

I expected to like the book and find the material not too difficult to follow (Four years of a Computer Science had to have been good for something! -ed.), but the presentation is so lucid that the material should be accessible to a high school student with no exposure to formal systems and so on. In fact, the author's interest in the subject was sparked when he was in high school and read Godel's Proof by Nagel and Newman.

Definitely one of November's Books of the Month!

Saturday, November 06, 2004

High School curricula

Continuing our math theme, we just conducted our second CS225 mid-term, and I've been grading exams for the last few days. I've noticed something interesting; students tend to provide reasonably good answers to questions which require them to write code, but are often unable to write proofs/justification for running times and correctness of algorithms. When I find a particularly good answer to a mathematical question, I often turn to the front and check the student's name out of curiosity; it appears that the best solutions are often written by non-Americans (judging solely by names). Before I get flamed for being racist, I should reiterate that I mean American, not Caucasian. Students of different ethnicities who went through school in America don't seem to do better than average; at least, I haven't noticed any evidence of it.

The most compelling reason for the discrepancy that I can find (since I absolutely refuse to accept any claim that Americans are 'inherently' less mathematically able than others) is the difference in the way Math is taught at the school level. (Disclaimer: I only have India as a basis for comparison, so perhaps this isn't reasonable either.) An Indian student who enters engineering college has already studied (differential and integral) calculus for two years, probability and statistics for at least as long, some form of co-ordinate geometry (with its emphasis on proofs) and algebra for over five years, besides some combinatorics, matrix algebra, and complex analysis. Some high school syllabi even include a fair bit of group theory! Graders in India also tend to be demanding (bordering on ruthless!) when evaluating exams, so students quickly learn when an answer is sufficiently precise to deserve full credit.

So is that a better system? I'm not sure; many Indian students never see a complex number after leaving high school and years of demanding Math often leave them hating the subject. I know several Indians who feel uncomfortable answering an exam question unless they're reasonably sure how to do it; a lack of partial credit (much less common in India than in America) discourages people from risk-taking. And I've noticed that American students are often more creative with solutions to both exam problems and random problems discussed in section, which makes teaching fun. It looks like they make up most of the relative deficiency in college, so doing less math in school doesn't seem like much of a loss. Perhaps the syllabi will change to embrace the best of both systems.

Right now, though, I often wish that there were more emphasis on rigorously correct solutions. Some of the best students seem to think that proof by example is completely reasonable! And it always makes me feel like I've done a terrible job in section when I have to grade papers which make warm, fuzzy statements like:
Trees are a Good Thing (tm) in general when compared to lists (except when they (the trees) are unbalanced, but that doesn't happen for Red-Black Trees (because they are always balanced), so it doesn't make a difference to this answer) and so searching in a tree is usually better than in a list (because we don't have to look at all the tree nodes), so we would rather use a tree.

(Admit it, that's an exaggeration! -ed.) Granted, but I yearn for a plain O(log n)!

Friday, November 05, 2004

Math was never this interesting

2 is an interesting number : Among other things it is the smallest prime. Ramanujam's number ,1729, is interesting, it is the smallest number that can be written as the sum of two cubes in two different ways. Perhaps the largest interesting number we know of is Graham's number.

Is it possible to define exactly what the set of 'interesting' numbers is? Not in any non-trivial way, if you're working with only the natural numbers. Suppose I is a set of of interesting numbers, such that I' is non-empty. Let d be the least element of I'. Since d is the least non-interesting number, it is therefore interesting. This contradicts the assumption that I' is non-empty. Therefore we have,

Theorem: There does not exist a set of dull natural numbers.

Corollary : Every natural number is interesting.

This argument can be extended to any countable set of numbers. How about the real numbers?

Can there exist a dull set of real numbers? Maybe, as long as they are uncountable and do not contain any natural numbers (or integers or rationals) , for then the least such natural number (or integer or rational) would be interesting.

It might seem that since a countable set can always be embedded into such a uncountable dull set of real numbers, its minimum element must always be interesting, but we have to be careful here. We cannot consider arbitrary orderings since they are not interesting.

Thus, if there exists only a countable number of interesting orderings, then there must exist a set of dull real numbers, because each interesting number can also be represented as the least element of an interesting ordering (interesting because it has an interesting element as its least element). Assuming the continuum hypothesis, the converse follows.

(Disclaimer: This article was written after drinking one too many tequila shots and should therefore be taken with a generous dose of salt, rather like the tequila.)

Wednesday, November 03, 2004

Four More Years!

It looks like President Bush will return to the White House for another 4 years. It was evident for several hours that Ohio would decide the election. With 98% of Ohio precincts reporting results, President Bush leads by 51% to 49%, a margin of 136,000 votes. Provisional and absentee ballots are still uncounted, but given that there won't be more than 250,000 of these (a very loose upper bound) and that some of them will be disallowed, I think it's unlikely that Senator Kerry will win enough of these to offset the Bush margin.

Still, the Kerry campaign is not giving up. Mary Beth Cahill, one of the campaign managers said, "The vote count in Ohio has not been completed. There are more than 250,000 remaining votes to be counted. We believe when they are, John Kerry will win Ohio." A few minutes ago, Vice Presidential candidate Senator John Edwards made this announcement to the Democratic supporters outside campaign headquarters: "We've waited four years for this victory, so we can wait one more night."

Since many Democrats were unhappy with Al Gore for not fighting "hard enough or smart enough" (quoting George Stephanopoulos) after the Florida debacle four years ago, Senator Kerry is understandably unwilling to concede the election until he has explored every avenue that could lead to victory. I'm not exactly thrilled by the prospect of a protracted battle, though; such wrangling could have the effect of polarizing the country further. Surely the statisticians employed by the Kerry campaign have reported exactly how unlikely it is that the Senator can pull off a win.

The Republicans appear set to dominate Congress as well. They held 51 seats in the Senate earlier; they have 52 so far, and are leading in all 3 seats for which final results have not been declared. Democrat Tom Daschle, the Senate Minority Leader, appears to be one of the casualties. The Republicans will also probably have a comfortable majority in the House. We'll have to wait to see how all this plays out.

Tuesday, November 02, 2004

Record Voter Turnouts?

Looking through early news stories, one thing is clear: regardless of who wins, the turnout at this election will be among the highest in recent memory. In spite of heavy rain in many parts of the country, this election is expected to witness at least 118 million voters, and may set the all-time record. Two newsbites that struck me:

From the New York Times,
In North Philadelphia, Valerie Morman, a legal secretary, walked to her polling place at St. Malachy School. "The last time I voted," she said, "was about 20 years ago."

From the Washington Post,
In north Milwaukee, 19-year-old Maurice Dodson waited in a long line to cast the first presidential vote of his life. "No way I'm leaving," he said after an hour with no ballot in sight. "I'm very excited.

Sunday, October 31, 2004

In just a few more hours...

... just over 48, to be precise, the result of the 2004 U.S. Presidential elections will be known. Assuming, of course, that the election doesn't have to be decided by the Supreme Court again; a state of affairs that I'm sure nobody would like. I'm not going to comment further now, though; I'll be up all night on Tuesday as the results come in and will post any thoughts then.

Oh, and a Happy Halloween to all my readers!

Monday, October 25, 2004

Seen in a Slashdotter's signature

"With Microsoft, you get Windows. With Linux, you get the full house" - original source unknown.

Enough said. :-))

Sunday, October 24, 2004

More Bad News in Iraq

The New York Times is carrying two stories from Iraq. The first covers the loss of 380 tons of powerful conventional explosives from Al Qaqaa, a former Iraqi military facility now (at least nominally) under American control. From the article:
The bomb that brought down Pan Am Flight 103 over Lockerbie, Scotland, in 1988 used less than a pound of the same type of material, and larger amounts were apparently used in the bombing of a housing complex in November 2003 in Riyadh, Saudi Arabia, and the blasts in a Moscow apartment complex in September 1999 that killed nearly 300 people.

The International Atomic Energy Agency publicly warned about the danger of these explosives before the war, and after the invasion it specifically told United States officials about the need to keep the explosives secured, European diplomats said in interviews last week. Administration officials say they cannot explain why the explosives were not safeguarded, beyond the fact that the occupation force was overwhelmed by the amount of munitions they found throughout the country.

After the invasion, when widespread looting began in Iraq, the international weapons experts grew concerned that the Qaqaa stockpile could fall into unfriendly hands. In May, an internal I.A.E.A. memorandum warned that terrorists might be helping "themselves to the greatest explosives bonanza in history."

The explosives could also be used to trigger a nuclear weapon, which was why international nuclear inspectors had kept a watch on the material, and even sealed and locked some of it.

"After the collapse of the regime, our liberation, everything was under the coalition forces, under their control," Dr. Omar (Rasheed Omar, the Iraqi minister of Science and Technology) said. "So probably they can answer this question, what happened to the materials."

Officials in Washington said they had no answers to that question. One senior official noted that the Qaqaa complex where the explosives were stored was listed as a "medium priority" site on the Central Intelligence Agency's list of more than 500 sites that needed to be searched and secured during the invasion. "Should we have gone there? Definitely," said one senior administration official.

Granted, it might not have been possible to secure every pound of arms and munitions in Iraq, but 380 tons of HDX and RDX? Assuming that the looters used 10-ton trucks to carry the explosives away, they would have needed a convoy of 40 trucks! And no-one noticed? (Perhaps they took it away in smaller chunks at a time - ed.) That's even worse; it implies that they waltzed in and out of the facility on a regular basis! Incompetency doesn't begin to describe this. One can only shrug in wonder at his capacity for self-delusion when President Bush portrays himself as the only candidate capable of winning the 'War on Terror'.

In the second story, Edward Wong reports that fifty freshly trained Iraqi soldiers were ambushed and killed by insurgents dressed as police officers in eastern Iraq. He writes
The executions of the Iraqi soldiers on Saturday evening - and what may also have been three civilian drivers in their convoy - raised disturbing questions about the training process and the recruits: Why were the guardsmen allowed to travel unarmed and without protection, given the frequent attacks on the Iraqi security forces? Why did men trained as soldiers not put up a fight, especially when there were so many of them? How did the insurgents get police uniforms and information on the travel plans of the soldiers?

Iraqi and American officials said they had no immediate answers.

One can't invent this kind of story... and there's no need to comment on it. In other news from Iraq, Edward J. Seitz, a 16-year employee of the State Department was killed this morning in a mortar or rocket attack on Camp Victory, near Baghdad International Airport. Camp Victory is the U.S. military's operations center in Iraq. Mr. Seitz seems to be the first American diplomat to be killed in the war.

After a day filled with bad news, a little thing can sometimes make a huge positive difference. This statement in a Washington Post article on voter rights and harassment moved me nearly to tears.
"I'm excited to cast my first vote," said Heidi Carrillo, 24, a new registrant who was born in the United States to illegal immigrants. "They can ask for ID. They can make me last in the line. I don't care. I'm voting!"

Isn't that what democracy is all about?

Sunday, October 17, 2004

Biological Nomenclature

Slashdot ran a story on Taxonomy and amusing nomenclature today. Mark Isaak collected some names worth a second look. The page is fairly detailed, but here are some of the more amusing names.

Ba Humbugi (endodontoid snail) from Mba island, Fiji.
Eubetia Bigaulae (tortricid moth, pronounced You betcha, by golly)
Pieza Kake, Pieza Pi, Pieza Rhea (mythicomyiid fly)
Strategus Longichomperus (Honduran scarab with elongated mandibles)
Ytu Brutus (water beetle)
Eristalis gatesi (a flower fly, named after Bill Gates. Someone decided he deserved a bug named after him!)
Montypythonoideriversleighensis (An extinct python whose remains were found at Riversleigh in Queensland, Australia, it was named in a tribute to comedy)
Fiordichthys slartibartfasti (a fish named after Slartibartfast - the award-winning Fjord designer in the HitchHiker's Guide to the Galaxy)
Commelina (day-flower, named after the three members of the Dutch Commelin family, two of whom prospered. The flower has three petals, one of which is small, pale, and shriveled.)
Mephitis mephitis (common striped skunk, variously translated "noxious vapours noxious vapours", "stench stench" or my favourite, "smelliest of the smelly")

Scientists with a sense of humour... my favourite form of life!

Photographs of Champaign-Urbana

Champaign-Urbana is a beautiful town, and Fall is one of its best seasons. I've been meaning to capture some of my favourite views, but never seemed to have the time. I dug my camera out this weekend and finally took a few photographs. I didn't intend to photoblog, but since Faraz and a couple of other people asked for photographs, here they are:

The University of Illinois at Urbana-Champaign campus.
Photographs of our apartment.
Fall in Champaign-Urbana.
Siebel Center for Computer Science. This is very much a work in progress; my snaps of Siebel are very sketchy at the moment.

I uploaded them to Ofoto, but suspect that eventually I'll host them myself at UIUC. Another task for my to-do list. Sigh!


Ditch (Deepak Ramachandran for the non-cognoscenti) has launched a blog of his own. As you can tell from its title, it's meant to be a weblog about AI research.
... (M)ostly this blog is about the discipline of AI as a distinct subfield of Computer Science - The effort to build machines that think, know, learn and are aware. Along the way, if we could define what these terms mean exactly, well that would certainly help. I hope to take it in the same direction as Lance Fortnow's Computational Complexity Blog.

Ditch was an occasional contributor to Pseudo-random Thoughts; with luck we'll hear more from him now. I hope he'll continue to grace us with guest posts here, though.

Wednesday, October 06, 2004

The Cost of War

The National Priorities Project documents what federal tax and spending policies mean to America. Among other things, they describe the dollar cost of the war in Iraq. The war has other, more important costs such as the apalling loss of life (mostly Iraqi, and to a lesser extent, American soldiers), the harm to America's credibility and reputation, and so on. They describe these as well, but they are well documented elsewhere (see Iraq Body Count, for example) and often speak for themselves. I'm going to focus on the money, then.

The Cost of War maintains a counter showing the expenditure on the war in Iraq. As of this afternoon, it had reached $138,693,576,310 and was climbing rapidly (yes, that's 138 billion dollars!). This amount could have been used for any of the following:
a) Completely insuring the health of 83 million children for a year
b) Hiring 2.4 million public school teachers for a year
c) Building 1.2 million additional housing units
d) Fully funding global anti-hunger efforts for 5 years, or global anti-AIDS efforts for 13 years.
e) Ensuring that every child in the world would receive basic immunizations for the next 46 years.

What really makes the site stand out, though, is this wonderful statement by President Eisenhower. He was a Republican, and (having been Supreme Commander of Allied Forces in Europe in World War II, Chief of Staff of the U.S. Army, and Supreme Commander of NATO) knew war first-hand; no-one could accuse him of being a 'bleeding-heart liberal'. He said, on April 16, 1953:
Every gun that is made, every warship launched, every rocket fired, signifies in the final sense a theft from those who hunger and are not fed, those who are cold and are not clothed.

Food for thought.

UPDATE: This might help you appreciate how rapidly the costs are increasing; in the time it took me to write this, over 2 million dollars have been spent!

Saturday, October 02, 2004

Selecting a Research Area

The Computer Science department at the University of Illinois, Urbana-Champaign requires all new Ph.D. students to attend a seminar in their first semester. As part of the seminar, various faculty members from different research groups describe the work they are currently engaged in. This is a fantastic idea, as it gives students the opportunity to learn about all the research being conducted in the department and helps us choose which area we would like to work in.

There's one thing I dislike, though; so far, some of the presenters (and almost all the students who I've spoken to) appear to believe that one of the primary reasons to select a research area is the potential impact that your work can have. Don't get me wrong; I completely understand that this could be a motivating factor for many people. Should it be the most important, though? What about research because you enjoy it? Is creating (or discovering) knowledge for its own sake not the most powerful motivation for a researcher? Sure, it must be incredibly gratifying to see your work in daily use. But if that's your reason for research, you might have done better to join a software company and work on products which are used by thousands of people. You could have even joined Microsoft!

Note: I'm not denigrating people who choose such careers; where would the world be if we all spent every waking minute wondering about the precise relationship between NP and BQP? Sure, there are people who would love to work at Microsoft; I'm not one of them. (What about Google? - ed. Ok, you got me!) What irks me is that there isn't anyone presenting their research area and telling us that we should choose it because it's beautiful; because we might love the field, its techniques and open problems. Now that would motivate me. I get a little tired of "Join us because we're relevant!" and "Join us because we're hot today/this week/year!". That's not what I hoped for; it's not what one would expect from people as deeply in love with their field of study as many of the faculty here are.

To be fair, though, even if that's what motivates the faculty, it might be difficult to convey it to students. I wouldn't be able to explain why I love algorithms and not databases to someone who had never taken a theory course. So perhaps this isn't a good way to attract new students to your field. Still, it would be nice to hear it once in a while. And the theory group is presenting their research next week; there's hope yet!

One final comment; what good research does not have an impact? For years, number theory was considered the "purest of pure" mathematics with no practical applications. Today, computational number theory is the basis for cryptography; its impact cannot be denied.

Thursday, September 23, 2004

The Sum of all Knowledge

That's the goal of Wikipedia, the free online encyclopaedia: Creating a world in which every single person is given free access to the sum of human knowledge. Wikipedia is free both as in beer (meaning that you don't pay to have access to encyclopaedia contents) and as in speech (meaning that you can take the content and do anything to/with it, except deprive other people of that right). Material is released under the GNU Free Documentation License.

Wikipedia is an encyclopaedia written collaboratively by its users; anyone can contribute. Pick a subject that you know something about, write an article, and it gets saved to the encyclopaedia database immediately. Millions of people will instantly have access to what you wrote. Considering the number of articles Wikipedia contains, though, it's probable that someone else has been there before you. In that case, you can freely edit their work and improve the article. That is, essentially, how Wikipedia works. They harness the power of thousands of volunteers who share knowledge of a wide variety of subjects. For more information, you might want to read the Wikipedia article on Wikipedia.

Why am I posting this now? After all, Wikipedia has been around for a while. This week is special; Wikipedia reached 1 million articles! In comparison, the Encyclopædia Britannica contains about 120,000 articles.

The Wikimedia Foundation, the parent organisation of Wikipedia, is involved in a Fundraising campaign to support Wikipedia and its sister projects. Spread the news; help Wikipedia today!

Sunday, September 19, 2004

Take back the web!

The preview release of Firefox 1.0 has been available for the last few days. The team set themselves the target of 1 million downloads in 10 days, more than ever before. Incredibly, they reached that target in just over 4 days! The new goal is to reach 2 million in the original 10 days.

Determining the actual number of firefox users as opposed to downloads is a little more difficult. For one thing, this count only reflects downloads off the firefox site, not mirrors or bittorrent. Often, a single download is used for multiple installs, as in business offices or educational institutions. On the other hand, some users download the browser several times for different computers at home or work. Still, this is a remarkable achievement by any standards. In all probability, it's a new record for beta-tested software.

(I thought you were an Opera advocate! - ed.) I am, but that doesn't mean I can't appreciate other good software or even software marketing. Many (dare I say most?) extensions for Firefox are copied from those integrated into Opera by default and the Opera versions are frequently better. Still, Firefox has better press and I'm happy to support it as long as it gets people to switch from IE.

Note: This is not Firefox 1.0, just a preview release of that version. Firefox is still officially in beta testing, and will remain that way until version 1.0 is finally launched.

P ?= NP

This post will only appeal to a limited audience; if you aren't in CS you probably have no idea what it means.

If you could choose to prove either that P = NP or that P ≠ NP, which would you pick? I'm not asking which you believe; you can decide mathematical truth here. Given that you have the ability to prove either, which one would you rather prove?

Remember to leave your name along with comments.

Back to Blogging!

Now that I've settled down in Urbana-Champaign and have bought a computer, blogging will resume on a regular basis. Thanks to all the people (particularly Faraz) who kept asking for updates.

I'll try at some time in the future to describe UIUC, Champaign-Urbana and life here, but for today I just wanted to post a link to a speech made at the 2004 (American) National Convention of the Society of Professional Journalists.

There are times I seriously consider a career in journalism.

Wednesday, August 04, 2004

I'm off to see the US...

..., the wonderful US of A. (Note: To be sung with a heavy dose of sarcasm.) Seriously, though, I'm looking forward to it, even if it isn't going to be Oz. I'm told that I'll have some free time until classes begin, so perhaps I'll blog from Urbana-Champaign. I'm leaving tomorrow and will spend a couple of days travelling. Expect more posts in a week at the latest!

Tuesday, July 27, 2004

Last Post!

It's 3 a.m. (and yes, it's Wednesday Morning!) and I'm extremely sleepy, but this is probably my last chance to post on this blog. I'm packing up my computer tomorrow and sending it to Coimbatore for my aunt to use for the next two years while my parents are in Brunei. Blogging has been a fun experience: From searching desparately for subjects to write about on a slow news day to being accused of incompetence for maligning Rumsfeld, I've enjoyed it thoroughly. I'd particularly like to thank my readers and commenters for all the feedback and comments, and the folks at Blogger for designing a wonderful publishing service (What about your parents and dog? What is this, an Oscar acceptance? - ed.)

Anyway, there might be more posts, but probably not for a while. I'm spending the weekend visiting relatives, and then will be home for only a few days before leaving for Urbana-Champaign. With luck I'll be able to blog from UIUC; let's see how it goes.

Signing off for now,

A Sample CS Curriculum

This post is best read in conjuction with, or immediately after the previous post, which describes the general outline for the curriculum. Here I present the courses which a hypothetical student majoring in Computer Science could take to satisfy the requirements. His (unrelated) interests are mainly Theoretical Computer Science and Physics.

First Semester
Mathematics I (3)
Physics I (3) (Selected topics from Mechanics and Optics)
Chemistry (3)
Biology (3)
Computing I (3)
Engineering Workshop (3)
Methods I (2) (Writing)

Second Semester
Mathematics II (3) (Algebra)
Physics II (3) (Selected topics from electromagnetism)
Applied Mathematics I (3) (Probability and Statistics)
Measurement Techniques (3)
Computing II (3) (CS version: science, not programming)
Engineering Graphics (3)
Methods II (2) (Design)

Third Semester
Mathematics III (3) (Differential Equations and Complex Analysis)
Circuit Theory and Introduction to Electronics (3)
Engineering/Science Elective I (3) (Modern Physics)
Humanities Elective I (3) (Introduction to Economics)
PC 1 (3) (Discrete Mathematics)
PC 2 (3) (Digital Electronics)

Fourth Semester
Applied Math II (3)
Engineering/Science Elective II (3) (Mechanics of Solids)
Humanities Elective II (3) (Principles of Management)
PC 3 (3) (Data Structures and Algorithms)
PC 4 (3) (Microprocessors)
PC 5 (3) (Graphs and Networks)

Fifth Semester
CDC 1 (3) (Theory of Computing)
CDC 2 (3) (Operating Systems)
CDC 3 (3) (Database Management)
CDC 4 (3) (Theory of Programming Languages)
Applied Math III (3)

Sixth Semester
CDC 5 (3) (Computer Networks)
CDC 6 (3) (Computer Organisation)
CDC 7 (3) (Compiler Construction)
CDC 8 (3) (Artificial Intelligence)
Control Systems (3)

Seventh Semester
Elective (3) (Parallel Computing)
Elective (3) (Advanced Algorithms)
Elective (3) (Computational Complexity)
Elective (3) (Theory of Relativity)
Elective (6) (2 Independent Study projects)

Eighth Semester
Project/Practice School

As a practical matter, some of the courses listed need not, strictly speaking, be requirements for Computer Science students. In the place of 'Theory of Programming Languages' and 'Compiler Construction' a single course could be taught in the second semester. Artificial Intelligence could also be removed as a hard requirement. In the two vacancies created, students could choose from one of several two-course sequences that the department may offer, which could include (among others):
a) Advanced Algorithms and Computational Complexity
b) Artificial Intelligence and Machine Learning
c) Image Processing and Multimedia Computing
d) Data Warehousing and Data Mining

The Proposed Curriculum

Here it is... the much-delayed proposal for a curriculum (see previous posts). I've procrastinated far too long, but packing up an entire house should be some excuse. Before I begin, the abbreviations I use are:
PC - Preparatory Course, an elementary course in a student's field that is required for the more advanced CDCs. An example could be Discrete Mathematics for a Computer Science student.
CDC - Compulsory Discipline Course, a required course for students majoring in a discipline. A CDC typically covers one field in reasonable depth, though not as much as a specialised elective. CDCs roughly correspond to 'higher-level' courses at several universities. Microelectronic Circuits might be compulsory for Electronics students, and Statistical Mechanics for Physics students.

Further explanations will be required, but it will probably help to see the curriculum first. I have listed courses in a semester-wise pattern, with between 18 and 20 credit hours per semester. One course is listed per line, with the number of credit hours in parentheses.

First Semester
Mathematics I (3)
Physics I (3)
Chemistry (3)
Biology (3)
Computing I (3)
Engineering Workshop (3)
Methods I (2)

Second Semester
Mathematics II (3)
Physics II (3)
Applied Mathematics I (3)
Measurement Techniques (3)
Computing II (3)
Engineering Graphics (3)
Methods II (2)

Third Semester
Mathematics III (3)
Circuit Theory and Introduction to Electronics (3)
Engineering/Science Elective I (3)
Humanities Elective I (3)
PC 1 (3)
PC 2 (3)

Fourth Semester
Applied Math II(3)
Engineering/Science Elective II (3)
Humanities Elective II (3)
PC 3 (3)
PC 4 (3)
PC 5 (3)

Fifth Semester
Four CDCs (12-14)
Applied Math III (3)

Sixth Semester
Four CDCs (12-14)
A systems / lab course (3)

Seventh Semester
Six electives (18-24)

Eighth Semester
Project/Practice School

A long series of explanatory notes is obviously called for, so here goes.

Numbering: Why are Preparatory courses numbered 1, 2, ... while Mathematics Courses are I, II, ...? As I said in earlier posts, one of the goals of this curriculum was to increase the amount of choice available to students. A Roman numeral after a course indicates that it is part of a series where the student has a certain amount (sometimes more, sometimes less) of freedom to select individual members in the series.

Mathematics: Mathematics I, Calculus is required of all students. In addition two more pure mathematics courses (6 credits) are required to cover Algebra (2/3 credits), Differential Equations (2/3 credits) and Complex Analysis (1 credit). A student can select as his first course either Math IIa (3 credits of Algebra) or Math IIb (2 credits of Algebra and one of Complex Analysis). If he has chosen Math IIa, he will have to take Math IIIa (1 credit of Complex and 2 of Differential Equations); else he will take Math IIIb (3 credits of Differential Equations). A student may choose either path depending on his interests or preferences, or on what he believes will prepare him better for future courses. Departments may recommend a stream for students of their major, but it can only be a recommendation, not a requirement. As an example, the Computer Science department may recommend Math IIa as better preparation for Discrete Mathematics and Theoretical Computer Science. On the other hand, a chemical engineer might prefer to place more emphasis on differential equations than advanced linear algebra.

Physics: Two physics courses are required, but students can choose from one of two streams. The first provides a general introduction to mechanics, electromagnetic theory and optics at a level slightly above high school. The second covers selected topics in greater depth: introductions to fields like relativity and quantum mechanics may be provided.

Computing: The first computer science course is common to all students. It provides an introduction to computers and computing, operating systems, programming, etc. The second course has two avatars: the first on programming, with topics such as file I/O, and the second covering abstraction, basic algorithms and data structures. Students hoping to major in Computer Science or related areas would be required to take the latter course; all others may choose.

Applied Mathematics: A three course sequence covering Probability, basic Statistics, Numerical Methods and Operations Research (including Optimisation techniques).

Methods courses: I cannot describe these better than Timothy Burke does in his proposal for the 21st century college. Note that Measurement Techniques (in the second semester) is also essentially a methods course, but it is named in the curriculum because it is required of all science and engineering students. The other two may be selected from a subset of those Tim Burke lists, with the proviso that either Reading or Writing must be chosen.

Engineering/Science Electives: This is one of the more complicated parts of the curriculum. Every student must, in his second year, choose two courses outside his own discipline from basic sciences or engineering fundamentals. Every department must offer one or more courses that meet this description, forming a pool from which students select. Typically, preparatory courses for students of one discipline could fill this elective slot for other students. Thus, Microprocessor Design and Interfacing is a required preparatory course for Electronics students (and hence cannot fill this elective position), but could be an elective for a Chemistry student. The idea is to provide a student with an elective slot to pursue an 'outside' interest, and/or expose him to a technical area he might not have otherwise studied.

Humanities Electives: To round out the sophomore year, a student must take two courses from a pool of humanities electives. These may include History, Economics, Management, Music, etc. The same rule applies here as previously; a student may not select a course that would normally be required for his major.

Every student is also required to do several preparatory and advanced courses from his own discipline; this is usually a minimum of thirteen named courses (in addition to some that may be common to all students) and a few electives. The seventh semester is reserved entirely for electives; students may use them for advanced study in their own discipline and/or to explore interesting areas in other fields.

I'll post a sample curriculum for a Computer Science student soon, but I'm done with the basic outline. I'd love any feedback: criticism, suggestions or compliments. (Note: compliments are especially welcome!). I may not be able to respond immediately, but I will reply to all comments.

Before I forget, the credit (or blame! - ed.) for this proposal isn't entirely mine. I really have to thank Karthik Narayanan, Sohini Roy, Divya Devarajan, Prof. Sundar Balasubramaniam and Prof. G. Sundar from BITS, Pilani for all their time, ideas, and help in various forms.

Saturday, July 24, 2004

Why do we need a new curriculum?

In my last post on the subject of science/engineering undergraduate curricula, I described some possible goals of any curriculum and a few thoughts that I had. The comments I received were interesting; I'll summarise some of them and explain the rationale behind my ideas today before posting the proposed curriculum tomorrow.

Ranjith said the the chief problem was not with the curriculum followed or even the quality of professors, but rather with the attitude of students. They should break away from a mark-centered school system and learn to explore subjects independently, but often don't realise this. Dileepan agrees that students are marks-oriented and proposes the solution of evaluating performance based on labs, case-studies or assignments instead of emphasising tests that encourage rote learning.

Rahul concurs with Ranji and Dileepan, but feels that there is a problem other than attitude. Students think that they need knowledge in their own areas of specialisation and believe that this is affected negatively by learning things that they don't need. They would love to know how the whole world works but don't want their core compromised. Though BITS permits you to do electives from any area, students choose them from subjects that they think they need (either to increase their specialised knowledge or programming courses to help them get software jobs). While doing a three-and-a-half year course, they give their discipline their first priority and compromise on stuff they would otherwise like to do.

While trying to design a new curriculum, I began from premises similar to Rahul's. I hoped that I could simultaneously increase the number of requisite discipline-specific courses so that people could choose electives as electives instead of using them to address a lack of knowledge that should have been covered and to increase the number of elective slots available to an average student. An ambitious goal, but I think I succeeded partially.

To respond to Ranji and Dileepan, if students were required to choose - early in their college lives - a few electives that were specially designed to explore new areas, with an emphasis on individual thought and few (if any) written tests and exams, would it help change their attitudes? The first-year courses at Timothy Burke's 21st Century College would seem perfect for this purpose.

I'm looking forward to tomorrow's post, where I have to actually describe a curriculum that achieves all the targets I've set for it. A pretty tall order, and one I don't know if I can fill; most of my previous attempts at curriculum design fall short in one way or another. An interesting challenge!

Thursday, July 22, 2004

I hate packing!

Why, you ask? Besides the fact that it keeps me from blogging, it's one of the most boring forms of work imaginable. My parents will be working in Brunei for two years, which means that we're packing up the entire house. And, of course, I have to take stuff to Urbana-Champaign.

Anyway, I'm nearly done, so with luck I'll get to do some blogging again. In particular, I have to finish the series on science/engineering curricula. Thanks to everyone who left comments, and I'm sorry for the long gap between the last two posts!

9-11 Commission report released

The National Commission on Terrorist Attacks Upon the United States (better known as the 9-11 Commission) has just released its final report. Unfortunately, it's a 7.4 MB PDF. You could download the summary instead, but at 5.9 MB (That's a summary! - ed.), it's not much better.

If you don't have the patience to download and read the report, CNN has good coverage, as do many other media organisations. The chief recommendation appears to be the appointment of a national intelligence director (possibly with cabinet rank) to co-ordinate intelligence gathering among the (approximately) 15 agencies involved in it. The Director of Central Intelligence (head of the CIA) would no longer have this responsibility, which probably explains why the CIA opposes it.

Hat tip: Dan Drezner.

Sunday, July 18, 2004

Blogging resumed

I just got back from Madras, where I had a wonderful weekend. I had intended to continue writing about university curricula, but I'm just too tired at the moment. Expect the post tomorrow morning.

Also, thanks to everyone who left comments. Ranji, if this were Slashdot, you would be +5, Insightful. Dileepan, I'm looking forward to yours. And Arun, welcome back! :-)

I think I've mentioned this before, but could readers please not leave anonymous comments? If you don't have a login, that's ok; you can include your name (or initials, even!) in the text of the message. It makes life so much easier on me.

Friday, July 16, 2004

Curriculum Redesign, contd.

If you haven't read my previous post, do that first. I was speculating on the possibility of revamping the course structure of an Indian science / engineering university. It seems reasonable to measure the efficacy of a curriculum by how well it meets the goals of its designers. I believe that the goal of a university's undergraduate curriculum should be to ensure that its graduating students:
a) have a working knowledge of various fields of study, are able to meaningfully apply this knowledge, and can increase their knowledge without assistance should they choose to do so.
b) possess the confidence and ability to begin learning any subject they find interesting.
c) are sufficiently qualified to begin either their professional lives or more rigorous graduate study in their area of interest.

To be successful, any proposed curriculum would have to require a broad set of foundation courses to fulfil the first two requirements and permit sufficient specialisation to fulfil the third. Unfortunately, Indian universities seem to focus only on the last requirement. At the few colleges where a broad range of subjects are taught, students are often 'spoon-fed', leaving them unable to go further on their own and thus leaving the first requirement unfulfilled. Sadly, as I said earlier, any college that decides to focus on providing a strong general education is usually criticised for not teaching students enough about their own discipline.

BITS is a case in point. It attempts to teach a common core of subjects to all students during the first two years, but still fails to provide a good general education. Students acquire random bits of knowledge about several fields, but little or no epistemology. (A couple of Timothy Burke's Methods courses would do well here.) All the discipline-specific subjects are compressed into the third year, which is plainly insufficient to do justice to them all. Many students have criticised BITS for not spending enough time on these important courses, but increasing the number of discipline-specific courses (hereafter CDCs) seems to require a decrease in the common 'core' of the first two years. Given that the core already seems to do its job badly, decreasing the number of courses further is hardly likely to help. At first glance, then, there seems to be no solution to this problem.

Perhaps, though, we're approaching it in the wrong way. Why assume that this is a zero-sum game? Can we strengthen one aspect of the curriculum without weakening the others? Can we simultaneously satisfy all the requirements of an effective curriculum? One (naive) workaround is to substantially increase the total number of courses. Students are not machines, so that seems unlikely to work, though a slight increase might be acceptable.

I spent a large part of last semester wrestling with these problems and produced several curricula. Before I present the latest iteration, I'd like some comments. First, are my goals for a curriculum meaningful? Can you think of better ones? Even if they seem reasonable, are they practically achievable? Am I being too ambitious? Second, do you have any ideas for a science/engineering curriculum? While this is a hypothetical exercise, the faculty at BITS has expressed some interest, so I'd like to keep it practical. That means that we would preserve sections of the current BITS curriculum wherever possible. Don't let that chain you down, though. If you have some radically new ideas, go ahead and post them. Quoting Dr. L. K. Maheshwari on a possible curriculum change, "Sometimes you have to destroy a structure before you can begin rebuilding."

I'll post my ideas on Monday; I'm going to Chennai for the weekend. I'll look forward to reading all the suggestions when I get back.

One last note: I've been saying science/engineering throughout these posts for a reason. Any engineering college will have to offer science courses as prerequsites for several subjects, and as electives for interested students. It is technically possible to have science departments that don't offer degrees, but such departments are unlikely to attract good faculty. A 'service' model, where faculty are retained only to teach introductory courses will almost always be unsuccessful. An integrated science/engineering university (like BITS) seems like a far better idea than a 'pure engineering' college.

The 21st Century College

Timothy Burke has written a proposal for what he calls the 21st Century College. It's too long for me to reproduce here, but you can read the whole article on his website. From the prelude:
I have come to the conclusion that so many of the problems of contemporary academia are wrapped up together in the same knot that incremental reforms may not be able to deal with these issues. I think we can do a better job that we presently do, though I also think—and have been scolded for saying so—that at its best, contemporary academia is good enough and performs many useful functions. While I perceive an overall crisis, a failure to live up to expectations, I also think things are not as dire as the strongest critics claim. I think liberal arts colleges can continue credibly as they are with some reforms, though I also think research universities, particularly public ones, are staring some more fundamental problems in the face in the near-term future.

However, I’m also attracted by the idea of trying to imagine a radically different kind of institution that carries forward some of the virtues of higher education but implements them in a radically new form, partly precisely because I think it’s not necessary to storm the ivory tower and burn it to the ground.

There are discussions of the essay on Critical Mass, Cliopatria and Crooked Timber. His ideas and comments apply mostly to liberal arts colleges in the US, but they got me thinking about how one could apply them in an Indian university. More specifically, can one redesign an Indian science/engineering university curriculum in this fashion? Some of you may remember that I spend a significant part of last semester considering a change in the curriculum of BITS, Pilani. The idea was to simultaneously increase a students preparation in his/her (Oooh... politically correct, are we? - ed. Not any more... too tiring!) own subject and improve his general education. Lest I sound presumptuous, let me clarify that. I don't mean that students of Indian universities are generally uneducated, but that students from American universities often seem to have a broader knowledge of subjects outside their own disciplines. Colleges outside India seem to have more of an emphasis on the humanities and subjects of general interest. How many Indian engineering colleges require (or even offer!) courses in political science, economics, or history? The purpose of an undergraduate education shouldn't be only to provide you with the skills to do a particular job - that's what a vocational education is for! Even the so-called 'professional' colleges should turn out students educated in a variety of fields, students who are able to pick up any subject they find interesting and gain a reasonable appreciation of it.

The problem, though, is that students here - and their parents - often obsess about a professional education, pushing their kids to become doctors, engineers, etc. An engineering / science college which reduced the professional component of its degrees to teach social sciences would find itself rapidly becoming unpopular. I know I wouldn't have wanted to attend a college that didn't teach me a substantial amount of Computer Science. (And yet you went to BITS! - ed.) These conflicting needs to generalise and specialise makes curriculum design particularly challenging.

How does one resolve this tension? Stay tuned for the next post.

Hat tip: Critical Mass

Thursday, July 15, 2004


To begin with, let me apologise for the lack of posts recently. I was busy with setting interesting questions for What's the Good Word? and had time for little else. I couldn't even sleep without dreaming of words chasing each other around in my head! Thankfully, I'm done with it; the contest was held this morning and it went off well, even if I do say so myself (Well, no-one else is going to say it! - ed.)

You may have noticed that I redesigned the layout of the blog slightly. I moved the blogroll up (something I should have done a long time ago) and added a promotional section to list products/services that I think are worthwhile. For now, the promo contains links to Opera and Firefox; more will be added later.

The real reason for the title Redesign, though, is that Swarthmore Professor Timothy Burke has described a radically new kind of college, reshaping higher education. This is a subject I'm particularly interested in, so look for a post on the subject soon!

Tuesday, July 13, 2004

What's the Good Word?

Yesterday was Ditch's first blog post; let's hope there are many more and that the Cyc project succeeds. I, for one, will welcome our new AI overlords! :-)

C.M.C. is holding it's annual sports/cultural festival, Pegasus, this week and I've been asked to conduct What's the Good Word? For those of my readers who've never seen it, it's a contest where teams try to guess words, similar to dumb charades. Each team has 2 members, one of whom knows the word and gives clues to his partner. Only three single-word clues are allowed. For example you might provide the clues attorney, court, prosecutor for your partner to guess lawyer.

Lakshmi and I were talking about it this evening and discussing our favourite funny, strange or generally pleasant and interesting words. Some of them were: discombobulate, dulcet, effulgent, epiphany, halcyon. Are there words that amuse, intrigue or please you? Leave a list of your favourite words in the comments.

Sunday, July 11, 2004

Philosophers Wanted. Enquire Within.

As an AI researcher, I find my internship at Cycorp Inc. a stimulating experience. But job satisfaction at Cyc is probably highest for the 20-odd philosophers that Cyc employs. That's right, philosophers.

The goal of Cyc is to build a complete Formalization of Human Knowledge, stuff we take for granted every day as "common sense". This includes seemingly trivial facts like "Rain is wet but ice is not." but also arcane wisdom such as "The temporal intersection of two events is disjoint with their spatial intersection." This kind of thing is, of course, right up the Philosophers' alley. Lunch-time discussions at Cyc are peppered with wry observations and vigorous debate over issues the rest of the world would dismiss with a "Whatever". Of course in true philosophical fashion, Cycorp view the rest of the world as a special case.

For example, did you know that the set of all unicorns is not equal to the set of all vampires; but the collection of all unicorns is equal to the collection of all vampires? Because a set is defined intentionally, ie. It has a meaning independent of its elements, whereas a collection is defined extensionally, it is merely the sum of its parts, which in both the cases above is zero.

Go ahead, scratch your head.

(To explore the wonders of Cyc more thoroughly, visit Opencyc - the Free [as in Beer] release.)

The Browser Wars

Recently, there's been talk about the return of the Browser Wars. Internet Explorer has been the subject of a lot of criticism for it's vulnerability, with even the US Government's Computer Emergency Readiness Team (CERT) recommending that users consider switching browsers. (Ok, to be honest, they listed that as the last update, after applying Security Updates, downloading the latest Antivirus Software, etc.) There's also been a lot of media coverage of alternative browsers. One result is that downloads of these browsers have increased sharply. This article in Wired News discusses how Mozilla has gained from IE's woes.
Downloads of Mozilla and Firefox -- an advanced version of Mozilla -- spiked the day CERT's warning was released, and demand has continued to grow. According to Chris Hofmann, engineering director at the Mozilla Foundation, formed last July to promote the development, distribution and adoption of Mozilla Web applications, downloads of the browsers hit an all-time high on Thursday, from the usual 100,000 or so downloads on a normal day to more than 200,000.

Hofmann said the Mozilla team wasn't surprised when CERT issued its warning.

"Mozilla and Firefox downloads have increased steadily since last fall, with the Firefox user base doubling every few months, as more people seem to have reached their threshold level of frustration dealing with problems with IE and Windows, and have found the Mozilla software a good solution to solving those problems," said Hofmann. "CERT's recommendation is just a reflection of the trend we have seen for quite some time."

I'm thrilled that people are beginning to realise what a terrible piece of software Internet Explorer is. It's slow, bloated, doesn't follow standards, and has more holes than Swiss cheese! What I'm depressed about, though, is that people don't seem to have heard about Opera. Everyone talks about Mozilla and Firefox and praises them for their innovative features (particularly tabbed browsing and mouse gestures). These were both Opera innovations! Even normally technically aware people like regulars at Slashdot don't get this. There has recently been a flurry of Slashdot stories on browsers (the two most recent). The vast majority of comments, though, deal only with Firefox. Commenters say that they have long wished for a browser to have certain features and hope that Firefox will incorporate them soon. Opera already implements most of them! Also, Gmail doesn't support Opera!

Opera seems (to me) to be both technically superior and have a far better User Interface, besides being almost infinitely customisable. I'm not alone, PC World voted Opera best browser of 2004. (And that was version 7.23; 7.50, the latest version is even better.) So why don't more people talk about (and use) Opera? There's one obvious reason to prefer Firefox; it's Free Software, whereas Opera is not. (Note to my non-tech readers: Free Software means free as in "free speech", meaning that you can view the source code, modify and redistribute it, not as in "free beer", meaning that you don't have to pay.) Both Opera and Firefox cost nothing, though the free version of Opera has text ads. (The paid version, which costs $39, does not.) The ads appear in a narrow band on the screen, though, and they are always relevant. I actually like them, which I never thought I'd say about ads. That annoys some people who don't want ads and don't want to pay, either. Firefox allows them to have their cake and eat it too. The ads aren't a big deal, though; in version 7.50, they're absolutely unobtrusive.

To many people the Freedom of software is important, and you have to respect that. If you don't want to use proprietary software, by all means use Firefox. It's one of the better options there is. But many people don't care about Free software, and just want the best software they can get. Do yourself a favour; try Opera today. If I haven't convinced you, and you're still running IE, at least check out Firefox. It's nearly as good, and both browsers are miles ahead of Internet Explorer in terms of features and security. In addition, they're smaller, faster (both to start up and to load web pages, especially on a slow connection like a typical Indian dial-up), and standards-compliant. Just get as far away from IE as you can!

Note: Both Opera and Firefox are cross-platform; they run on Windows, Linux, Mac OS, and many others.

Disclaimer: I strongly support the Free Software Foundation, think that the GPL is a Very Good Thing (tm), and use a lot of Free Software myself. However, honesty compels me to admit that Firefox isn't, in my opinion, the best browser out there right now; Opera is. I wrote this post because I felt it was unfair for such a great product to receive so little attention.

Saturday, July 10, 2004


#1 A few people asked me why I didn't comment on the budget. Two reasons: I'm not an expert on the subject (Like that ever stopped you before! - ed.) and much more importantly, I think there's plenty of good information and analysis available, both online and of the Dead Tree variety.

#2 At his request, Ditch will be appearing as a guest blogger on occasion. I think most of my readers know him, with the notable exception of Arun.

#3 I know there was a 3... now I'm going to be preoccupied until I can figure out what it was. Will update this post when it occurs to me. UPDATE: Finally! Critical Mass has a discussion of the National Endowment for the Arts recent report entitled "Reading at Risk: A Survey of Literary Reading in America". Also check her previous post on the writing of Tupac Shakur being taught in American schools.

Friday, July 09, 2004

Attempt to weaken Patriot Act unsuccessful

A proposal by Independent Rep. Bernie Sanders to water down the controversial Patriot Act failed to pass by a single vote in the House of Representatives today. The proposal would have barred the federal government (specifically, the Justice Department) from demanding records and reading lists from libraries and bookstores without showing probable cause.

From the New York Times article:
The vote, a 210 to 210 deadlock, amounted to a referendum on the antiterrorism law known as the USA Patriot Act and reflected deep divisions in Congress over whether the law undercuts civil liberties. Under House rules, the tie vote meant the measure was defeated.

Federal law enforcement officials say the power to gain access to such records has been used sparingly. Still, the provision granting the government that power has become the most widely attacked element of the law, galvanizing opposition in more than 330 communities that have expressed concern about government abuse. Critics say the law gives the government the ability to pry into people's personal reading habits.

"People are waking up to the fact that the government can walk into their libraries, without probable cause, without any particular information that someone was associated with terrorism, and monitor their reading habits," Representative Bernie Sanders, the Vermont independent who sponsored the measure, said in an interview.
... [T]he Justice Department on Thursday sent a letter saying that at least twice in recent months "a member of a terrorist group closely affiliated with Al Qaeda used Internet services provided by a public library."
Last September, Attorney General John Ashcroft accused critics of the government's library powers of fueling "baseless hysteria," and he grudgingly declassified government data showing that the Justice Department had not yet used the power to seize library records.

But the department has refused to say how often the authority has been used since, saying the information remains classified. The American Civil Liberties Union said last month that documents disclosed in court challenges showed that the Federal Bureau of Investigation had sought to use that section of the law soon after Mr. Ashcroft's declaration.
The Washington Post article provides some more information.

At one point, the proposal seemed sure to pass, with the tally reading 219 in favour to 201 against. But the vote, scheduled to last for 15 minutes, was extended by Republicans to 38 minutes as they desparately tried to persuade defectors from their ranks to support the bill. Their efforts were successful; 9 representatives (8 Republicans and 1 Democrat) changed their votes to ensure the tie.

Of course, there's no point expecting logic from Ashcroft, but this is even worse than usual. I'm sure 'members of terrorist groups closely affiliated with Al Qaeda' used the U.S. postal system or public transport. Is the government going to demand records be kept on all letters/parcels and taxicab customers? Would you approve of the government scrutinising your reading habits or keeping tabs on who you write to? And using the argument that the government had not examined such records prior to last September (though it has since then) is ridiculous. The frequency with which a law may be applied should have nothing to do with whether it is passed.

The admininstration will, I'm sure, hail this as a 'victory in the war on terror'. It's a shameful defeat in the war for freedom.

Meanwhile, I hope the Indian Government keeps their promise to repeal POTA, and soon.

UPDATE: Slashdot (slightly late, as usual) has a discussion going. My favourite part is this quote from Thomas Jefferson: "I am really mortified to be told that, in the United States of America, a fact like this can become a subject of inquiry, and of criminal inquiry too, as an offence against religion; that a question about the sale of a book can be carried before the civil magistrate. Is this then our freedom of religion? and are we to have a censor whose imprimatur shall say what books may be sold, and what we may buy?"

Thursday, July 08, 2004

Up the Down Staircase

A few years ago, when they were leaving the country, some friends gave us a large pile of books. Many of them are extremely uninteresting, so I went through them a couple of times and extracted the ones I thought were worth reading. Somehow, I missed this treasure until today.

Up the Down Staircase [] is the story of Sylvia Barrett, a young teacher who begins her career at an inner-city high school. From an Amazon Review:
Miss Barrett arrives at Calvin Coolidge High, to teach English to a motley band of students. Among them are: Hormone-addled Linda; resentful, angry Joe Ferone; woman-hating Rusty (who repeatedly tells Barrett that he would like her if she weren't "a female"); Edward Williams Esq., who thinks that everything is racially-prejudiced; soppily romantic Alice, and a slew of others. Miss Barrett realizes over time that the kids are screaming out not just for education, but for love and understanding. But will her idealism break through to them?

The dialogue is funny, especially since quite a few of the students don't spell-check. ("Fuk"?) There are also suggestion box excerpts ("You think it's fair when a teacher takes off 5 points on a test just because I misspelled his name wrong?"; "We're behind you 85%!") and book reports ("We study myths to learn what it was like to live in the golden age with all the killings"; "We read it because it's a classicle").

Up the Down Staircase is similar to To Sir, With Love [] in its portrayal of an idealistic teacher who struggles to break through students' distrust, but with significant differences. Up the Down Staircase is more light-hearted and personal, funny and quick-moving. It isn't a novel in the traditional sense; the story is narrated entirely through letters, memos, circulars, excerpts from a suggestion box, minutes of staff meetings, and so on (reminiscent of 84 Charing Cross Road). Well worth the read; I stayed up with it all night.

Wednesday, July 07, 2004

Back to Blogging!

My last few posts weren't displayed for a couple of days; Blogger seemed to be having some trouble with their servers. It looks like the problem has been fixed, so I can post again! I hadn't realised until now how much I enjoy blogging; it was incredibly frustrating to be unable to comment on things I found interesting.

Dileepan just told me he has a blog. Also, I found Shivku's blog via Orkut.

And to make a good day better, The Hindu is carrying a story about BITS being excluded (temporarily) from the AIEEE counselling. This is merely until the Supreme Court decides on the petition filed by several students. The case is due for hearing on July 12th.

Monday, July 05, 2004

The Politics of Software

Via Slashdot, I came across this article in the New York Times:
In a campaign season of polarization, when Republicans and Democrats seem far apart on issues like Iraq, the economy and leadership style, it is perhaps not surprising that the parties find themselves on different sides in the politics of software as well.

The Web sites of Senator John Kerry and the Democratic National Committee run mainly on the technology of the computing counterculture: open-source software that is distributed free, and improved and debugged by far-flung networks of programmers.

In the other corner, the Web sites of President Bush and the Republican National Committee run on software supplied by the corporate embodiment of big business - Microsoft.

Reflective of their respective positions on intellectual property? Or just coincidence? Either way, check the lively discuussion on Slashdot.

Pay Cut for Disrupting Parliament?

Lalu Prasad Yadav has come up with a novel idea to reduce disruptions of parliament by MPs: dock their pay for each day they prevent parliament from functioning. The reason for his concern appears to be that the NDA has hinted that it might boycott the presentation of his first Railway Budget, though the BJP has deferred the final decision until tomorrow morning. From Rediff:
Railway Minister Lalu Prasad has suggested the enactment of a law to stop an MP's salary on the days he/she disrupts proceedings in Parliament.

"A new law should be adopted to see that those disrupting the House proceedings should not be paid salary for the days they disrupt the proceedings," he said on Sunday while participating in a programme on BBC Radio.

He said despite the National Democratic Alliance-led Opposition's threat to boycott the Railway Budget, "I will present it on the 6th [of July] whether anybody listens or not. Through Parliament I will inform and address the Indian public, I don't care, damn care," he was quoted as saying.

It's surprising how he manages to come up with good ideas once in a while. For outrageous originality, there's no-one to touch him in Indian politics.

Sunday, July 04, 2004

The Great Visa Give-away

Lakshmi had her visa interview yesterday; she was told she'd get her (F-1) visa in a day or two. Arpith, Pradheep, Shastry, Suresh, Vishaka, Bugs and I all have ours as well. In fact, I haven't heard of a single F-1 visa rejection this year. Most of us had extremely cursory interviews, and it often seemed that the interviewer didn't really care about the answers. I'm not complaining, though. It's good to see that genuine students are being granted visas without much of a problem.

In other news, the Dhananjoy Chatterjee case (more detailed coverage here) seems to be the subject of intense debate. My mother has one comment to make: Why is everyone asking the hangman for his opinion? He has a vested interest in the execution; he gets paid Rs. 10,000 if Chatterjee is hanged.

I'm not sure whether I agree with all the demands for the execution. While rapists deserve to be severely punished, the death penalty is final. A conversation from the Lord of the Rings comes to mind, Frodo and Gandalf discussing Gollum.
'I can't understand you,' [said Frodo]. Do you mean to say that you, and the elves, have let him live on after all those horrible deeds? Now at any rate he is as bad as an Orc, and just an enemy. He deserves death.'

'Deserves it! I daresay he does. Many that live deserve death. And some that die deserve life. Can you give it to them? Then do not be too eager to deal out death in judgement. For even the very wise cannot see all ends.'

Let he who is without sin among you cast the first stone.

Saturday, July 03, 2004

Control Room

Via an op-ed in the New York Times on Al-Jazeera (and Fox News), I came across a review of "Control Room", a documentary made by Jehane Noujaim on the difference in perception of the Iraq war by Americans and Arabs. Quoting from the movie's brochure:
In the early days of the war in Iraq, Americans could turn on their televisions twenty-four hours a day and take a front row seat with coalition troops careening across the desert. We could follow the action live as precision bombers brought Iraqi cities to their knees and American POWs were rescued and triumphantly returned home as television heroes. We could watch soldiers toppling statues of Saddam Hussein.

But as Americans witnessed U.S. victory at home, a different story unfolded on television sets throughout the Arab world. Qatari-based Al-Jazeera broadcast images of Iraqi civilian casualties and American POWs that were taboo in the American media. Many claim that as a result, America barred Al-Jazeera journalists from reporting on Wall Street and bombed their headquarters in Baghdad.

As the saying goes there are always two sides to every story, but in a media-managed war where does the truth lie? With exclusive behind-the-scenes access to Al-Jazeera, American journalists, and the players at Central Command, CONTROL ROOM takes an unprecedented look at the business of war. Uniquely qualified with a cross-cultural perspective, Egyptian-American filmmaker Jehane
Noujaim (, travels to the headquarters of Al-Jazeera and U.S. Central Command to capture the staging of the war in Iraq and the media’s vital role in writing history.

This quote from Rumsfeld seems to describe the gulf perfectly.
"We know that Al Jazeera has a pattern of playing propaganda over and over and over again," Don Rumsfeld complained during the war. "What they do is, when there's a bomb that goes down, they grab some children and some women and pretend that the bomb hit the women and the children. . . . We are dealing with people that are perfectly willing to lie to the world to attempt to further their case — and to the extent people lie, ultimately they are caught lying and they lose their credibility."

Rich, isn't it, coming from him? What was that proverb about glass houses and stones again?

Anyone know where I can get my hands on a copy of this movie? Might make an interesting counterpoint to Fahrenheit 9/11, another documentary to watch this summer.

Friday, July 02, 2004

President (Kalam, not Dubya) recommends Open Source

From an article in The Hindu:
The President, A.P.J. Abdul Kalam, today advised defence scientists to shun proprietary systems and opt for open source codes to enhance software security in defence networks.

"Open source codes can easily introduce the users to build security algorithms in the system without the dependence of proprietary platforms. We should take maximum care to ensure that our solution is unique to protect our own defence security solutions implemented on open platforms," the President observed at the silver jubilee celebration of the Indian Navy's Weapons and Electronic System Engineering Establishment (WESEE). During his interaction with IT experts in civilian and defence fields, Mr. Kalam has consistently advocated the use of open source software over proprietary software whose building block is rarely disclosed by corporates who own and sell them.

This is why we love the man. I can't imagine that many heads of state recommend Free Software, or even know what it is.

Thursday, July 01, 2004

ISPs permitted to read users email

Wired News reports:
E-mail privacy suffered a serious setback on Tuesday when a court of appeals ruled that an e-mail provider did not break the law in reading his customers' communications without their consent.

The US First Circuit Court of Appeals (covering large parts of New England) ruled that Bradford C. Councilman was not in violation of wiretap laws when he copied and read customers email so that he could monitor their business transactions with rival
Councilman, owner of a website selling rare and out-of-print books, offered book-dealer customers e-mail accounts through his site. But unknown to those customers, Councilman installed code that intercepted and copied any e-mail that came to them from his competitor, Although Councilman did not prevent the mail from reaching recipients, he read thousands of copied messages in order to know what books customers were seeking and gain a commercial advantage over Amazon.

Authorities charged Councilman with violating the Wiretap Act (formally known as the Electronic Communications Privacy Act or ECPA), which governs unauthorized interception of communication. But the court found that because the e-mails were already in the random access memory, or RAM, of the defendant's computer system when he copied them, he did not intercept them while they were in transit over wires and therefore did not violate the ECPA, even though he copied the messages before the intended recipients read them. The court ruled that the messages were in storage rather than transit.

However, the court did acknowledge that the Wiretap act might be inadequate to protect consumers privacy over the Internet. In spite of all the criticism being levelled at them, the decision could have been correct; the court was merely interpreting this law, and not legislating, as American courts too often do. (Roe v. Wade, anyone?) The judge, in his decision (PDF) appears to rely on the distinction between electronic and wire communications. Quoting from the ECPA:
"wire communication" means any aural transfer made in whole or in part through the use of facilities for the
transmission of communications by the aid of wire, cable, or other like connection between the point of origin and the point of reception (including the use of such connection in a switching station)... and such term includes any electronic storage of such communication

"electronic communication" means any transfer of signs, signals, writing, images, sounds, data, or intelligence of any nature transmitted in whole or in part by a wire, radio, electromagnetic, photoelectronic or photooptical system that affects interstate or foreign commerce...

"electronic storage" means--
(A) any temporary, intermediate storage of a wire or electronic communication incidental to the electronic transmission thereof; and
(B) any storage of such communication by an electronic communication service for purposes of backup protection of such communication;

Note the absence of the phrase "electronic storage" from the definition of "electronic communication", even though it is included in the definition of "wire communication." The judge cites precedents which uphold the principle that
"When Congress includes a particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the the disparate inclusion or exclusion."
As RAM most definitely is electronic storage, he appears to have interpreted the law correctly.

It seems that the judge sympathises with the authorities, but believes that a strict interpretation of the law compels him to acquit the defendant. In order that this miscarriage of justice is not repeated, they believe that the ECPA should be revised. To that extent, I agree. If this is what the law says, then (quoting Mr. Bumble) "the law is an ass" and we need a new law. The implications for online privacy are stunning. Any email you send over the net can now be legitimately read by your ISP and/or email service provider, and they can freely use the data however they see fit. I can't even begin to describe the consequences! Compared to this, the automated parsing of email by Gmail is entirely innocuous. The ECPA should be modified immediately. Posters on Slashdot have a brilliant solution: Gillbates writes "I feel like starting an ISP and offering free email accounts to congressmen, judges, FBI agents, etc... The time difference between an embarrassing email leak and legislation outlawing reading another's email is left as an exercise for the reader...."