Wrappin' it up

Follow me over here. No more posts on this one.

Clear and unambiguous reporting.

Well, that sure cleared things up! Here's the first story on the Toronto Star about Stephane Dion's resignation (this is the entire text of the article):

A very Liberal prediction

Neither Bob Rae nor Michael Ignatieff will be the next leader of the Liberal Party.

Why? Because as soon as the campaign begins, they will start reminding people why they voted for Dion -- Dion! -- instead.

That doesn't look so bad.

This morning's 1000-point drop on the Toronto Stock Exchange was so bad, it broke Yahoo's stock ticker (note, the open was around 10,800 points):

Science: Gross and fascinating. (Grosscinating?)

In only the fifth recorded battle between pythons and alligators, we discover the following:

1. "Pythons versus alligators" is a question of legitimate academic interest; and

2. Yes, it is possible for a snake to swallow something so big, that its stomach explodes.

None of the above: An endorsement

You may have heard that we are facing an election. Let's review the major parties:

The Conservatives. To their credit they haven't governed too badly, mostly keeping the country going in a straight line from where the last administration left off. But two things are deal breakers for me. First, where's the vision? In trying to convince everyone that there's no hidden agenda, they seem to have jettisoned any coherent agenda at all. One has the sense of the Prime Minister in his office, gleefully moving tiny action figures around a huge map of Canada as he thinks up clever political tactics to deal with the issue of the day. Second, where's the credible environmental plan? This is probably the most important international issue of the decade, and Canada risks being on the outside looking in. So no vote for you, Mr. Harper.

The Liberals. Let's put aside the whole "Dion's a nerd and a weak leader" thing, and put our focus exactly where Dion wants it: on the environment. I have plenty to say about the plan itself, but I will restrict myself to this: Canada signed the Kyoto accord in 1997, and ratified it in 2002; both of these events took place under Liberal leadership. To 2005, at which time Canada remained under Liberal leadership, Canadian CO2 emissions had increased over 1990 levels by 26.6%, a larger percentage increase than even the non-signatory United States, and second-worst (to Spain) of the 36 countries for which Kyoto required CO2 reductions. For Mr. Dion, who was the goddamn environment minister from 2004-2006, to be using climate change as a campaign plank is hypocritical in the extreme, and automatically makes his environmental plan not credible. No vote for you.

The NDP. Sorry, but every time Jack Layton opens his mouth I feel like I'm being sold a used car. No vote for you.

The Greens. I was actually thinking about voting Green this time around. Then this happened. See under "Liberals". No vote for you.

The Bloc. Not running a candidate in my outside-of-Quebec riding, so I couldn't vote for them even in the unlikely event that I wanted to. No vote for you.

By the cold, calculating process of elimination, "A Random Process" is proud to endorse: None of the above.

He should have picked Tina Fey

In the past 14 U.S. presidential elections, the ticket with the shorter name has won 9 times, while the ticket with the longer name has won only four times (once the tickets were equally long). So I'm calling it for Obama/Biden (10 characters) over McCain/Palin (11 characters).

Those nine elections for the shorter ticket were:

1952: Eisenhower/Nixon (15) beat Stevenson/Sparkman (17)
1956: Eisenhower/Nixon (15) beat Stevenson/Kefauver (17)
1968: Nixon/Agnew (10) beat Humphrey/Muskie (14)
1972: Nixon/Agnew (10) beat McGovern/Shriver (14)
1980: Reagan/Bush (10) beat Carter/Mondale (13)
1984: Reagan/Bush (10) beat Mondale/Ferraro (14)
1988: Bush/Quayle (10) beat Dukakis/Bentsen (14)
2000: Bush/Cheney (10) beat Gore/Lieberman (13)
2004: Bush/Cheney (10) beat Kerry/Edwards (12)

The four for the longer ticket were:

1960: Kennedy/Johnson (14) beat Nixon/Lodge (10)
1976: Carter/Mondale (13) beat Ford/Dole (8)
1992: Clinton/Gore (11) beat Bush/Quayle (10)
1996: Clinton/Gore (11) beat Dole/Kemp (8)

The remaining ticket was:

1964: Johnson/Humphrey (15) beat Goldwater/Miller (15)

Hijinks ensue

On the morning of Monday, September 1, the Republican National Convention kicks off in St. Paul, Minnesota.

On the afternoon of Monday, September 1, Hurricane Gustav will make landfall on the US Gulf Coast.

Where will Gustav hit? Well, according to this forecast, the center of the forecast track is bearing down on New Orleans. How's that for convention optics?

How not to teach

A case study in how a professor should not behave in the classroom. Strictly this was a debate competition, but the principle holds. (Caution, lots of swearing.) [via Gawker]

Brave new world

Brain in a jar controls robot remotely. (A few neurons from a rat fetus ... hooked up via bluetooth ... but still.)

I'm not sure if I would be more creeped out, or less, if the rat brain were actually on board the robot.

On Blue 22

My comment on this article:

Arguments against Blue 22 tend to be specious, and those made in this article [are] no exception.

- “only 17 percent of travellers departing from Pearson originate in the downtown area” — I’m not sure where you got those numbers, so I can’t look it up myself, but I’ll bet that downtown is by far the largest single origin/destination for Pearson passengers. It’s also home to an increasing number of people, myself included.

- “SNC-Lavalin intended to charge $20 for a one-way ticket, which would price many out of the market” — First, an air ticket is a premium luxury good, usually purchased by the (relatively) affluent; it makes no sense to give flyers a government subsidy on their transit trip to the airport, and I’m happy to make them pay. Second, if you hang around the airport a lot, you will observe that families (who would not benefit from Blue 22) make up a small minority of flyers. Third, I’m sure the executives of SNC-Lavalin are smart enough to lower the price if they can’t fill trains at $20, or offer family passes if need be.

- “There are several options that should be studied instead of Blue 22″ — Have you ever used Chicago’s blue line? Or London’s Picadilly line? In each case it takes about an hour to get downtown (not including wait times), which is wearying after a long flight, and frustrating after a short one; this is not a compelling option for business travelers, who will simply take taxi. Further, light rail is already going to the airport, via the Eglinton TransitCity line.

- “The advantages of [a rail shuttle] solution include providing a station at Woodbine Racetrack where VIA trains from London, Kitchener and Guelph could gain direct access to the airport” — If your objective is to provide connections with VIA and GO, why not do that at Union? It’s already the largest transportation hub in the region, and it makes no sense to duplicate that function at the airport — especially since the arguments in favor of London, Kitchener, and Guelph ignore points east and south of Toronto, such as Oshawa, Kingston, Hamilton, and the Niagara region.

- Finally, and most importantly, there’s no reason why your suggested routes — which, as you point out, complement an express route rather than replace it — can’t also be built. You talk of “putting the cart before the horse”, whereas it is more like sticking your thumb in the eye of the cart vendor, because you don’t have a horse yet. Given the progress of transit construction in this city, the alternatives are decades away at best, while Blue 22 is ready to go right now and has the backing of government. Rail to the airport is a major missing piece of transportation infrastructure, and it’s time to take the solution that is available. One can turn your argument on its head: if regional, non-express services are still needed, they can be built later.

The weather on Mars

At the site of the Phoenix lander, anyway: Sunny, with a high of -26 C.

[CSA via Environment Canada Weather Office]

Memo to arXiv cs.IT users

Hey guys. I realize that you're all brilliant and have so much to share, but do you think you could possibly condense your ideas a bit? I read interesting abstracts, and download the paper to find -- every damn time -- a 40-page horse pill that I don't have time to read, and that feels like a waste to print if I'm just going to skim it.

The next station is St. George. Repent your sins.

There are a few transit routes I can take to work -- all are about equally fast -- but one of them has me connect from the Bloor-Danforth line to the Yonge-University line at St. George station. In the midst of an unusual number of transit backups and delays (though happily none in my direction), I got off at the B-D platform to make the switch ... only to find that it was impossible to make it up the stairs to my next train. There were so many people waiting on the Y-U platform that the crowd was backing up down the stairs and onto the B-D platform.

It took me a moment to realize that, to get to any exit, I would also have to go through the Y-U platform. So in an emergency, everyone on the B-D platform would have been stuck. I guess we could have made our way down the tunnel, but that doesn't sound like a key to a long and prosperous life -- especially if the problem (like a fire) were in the tunnel in the first place.

Which reminds me of another St. George experience. A few months ago, I was making the same connection -- this time I got up to the Y-U platform all right, but there was a stalled train on the southbound track and a building crowd, as well as an acrid smell of smoke. I actually saw a firefighter make his way down the platform and head into the tunnel ahead of the stalled train, and noticed the smoke getting worse ... but, happily, just then my northbound train arrived and I left the problem behind (I don't know what happened, but it wasn't on the news so it must have worked out).

The CDS and RMC

Lt.-Gen. Walter Natynczyk, the government's pick as the next CDS, received his bachelor's degree from CMR in St-Jean. That makes him the fourth CDS in a row who did not graduate from the Royal Military College, Kingston.

The last RMC grad to get the top job was Jean Boyle, who resigned amid scandal 12 years ago.

The new guy is also the first graduate from any Canadian military college to get the top job since Boyle: Baril did his degree at Ottawa, Henault went to Manitoba, and Hillier went to MUN.

A stirring "dunt-da-DUNT-da-dunt"

What's that you may ask? The mating song of the sea elephant? A sofa falling down the stairs?

Neither. It's the Hockey Night in Canada theme song, according to the Globe and Mail.

Good versus best

I'm creating a presentation and ended up looking for some particular clip art. After a brief look around the web, I came across Open Clip Art, which features a huge library of free clip art under a permissive license.

The problem? The clip art is in scalable vector graphics (SVG) format, which can't be used with the most popular free office suite in the world, OpenOffice.org.

Does it matter if you're creating a collection of clip art in the "best" format, if nobody can use it (or at least not without a significant hassle)?

501 reasons to hate the streetcar

Andrew Coyne on taking away transit workers' right to strike:

But even if it did achieve the goal of ending service disruptions, all that would ensure was uninterrupted TTC service: slow, infrequent, obstructive (Toronto is the only city in the world where traffic improves in a transit strike, since the streetcars are no longer blocking both lanes), and unpleasant.

That's a mild -- but only mild -- exaggeration. I've never understood why this city insists on operating railed vehicles as though they were buses, thereby combining the disadvantages of both. Nor why transit "advocates" have fought tooth and nail to keep the streetcars -- and to expand the streetcar system -- without fighting equally hard to see that they are operated efficiently.

On sucking chest wounds

On announcing his retirement from the military, Gen. Rick Hillier compared the pain of leaving his profession to that of a "sucking chest wound". [G&M via Paul Wells].

An ordinary human might think of that as kind of a weird metaphor. However, I'm thinking the good General was made to watch the same first aid video that I had to watch fifteen years ago, as a fresh recruit at the Canadian Forces' Officer Candidate School. Naturally, one of the injuries being treated in the video was the "sucking chest wound". For the rest of my (short, undistinguished) military career, the phrase "sucking chest wound" would get a laugh out of any of my colleagues.

The video made quite the impression. I remember the treatment to this day: you're supposed to find a plastic bag and tape it over the wound, but don't seal the bag tightly over the wound or else fluid builds up (the consequences of which were graphically illustrated in the video as the "victim" coughed up a bunch of fake-looking blood).

Also, apparently the technical term for such an injury is pneumothorax.

Just say no.

Apparently 20% of top scientists are now taking performance-enhancing drugs, such as Ritalin, without a medical need for them. [WebMD]

Here's a memo to all those who are popping the pills to write the papers and grants. I'm not going to get all Olympicky on you and claim that your research results are tainted because of what you're doing. Science is science, as far as it goes, and that's the problem.

What you're doing is far more subtle and vicious. You're using a chemical means to push yourself ahead a few places in the pecking order, to a place where you wouldn't be if you weren't taking the drugs. And since we're all competing for the same jobs, the same grants, and the same tenure, your actions have serious professional consequences for me. The raised expectations created by your drug-taking may eventually force me to take the drugs myself; and while you may be okay with the side-effects, I'm not keen on them being forced on me. Taken to the extreme, we could have an academy where everyone has to take Ritalin just to get ahead, but because everyone is taking the drugs, the pecking order is exactly the same as it was before. In case your drug-addled mind can't make the connection, here it is: it's a classic case of the Prisoner's Dilemma. And you're the first one who chose to rat out your comrade.

One might suggest that I am guilty of the same thing, in the form of coffee. But in fact this is evidence supporting my thesis. I freely admit that I am a caffeine addict, in the technical sense: I experience withdrawal symptoms when it is unavailable. So is essentially every high-achieving professional person that I know. We are addicts not because we want to be, but because we have to be: the caffeine helps us sustain an unnatural level of alertness, which is necessary to compete on the same playing field as all the other caffeine addicts. From the example of coffee, it's easy to see how another drug might have a similar game-changing effect, to the extent that high achievers would effectively need to take them just to compete. And unlike coffee, whose side effects are mild and which is known to have some health benefits, the side effects of prescription stimulants tend to be nastier (including, in many cases, a chance of psychosis).

To those who are taking these drugs, I say this: You selfish bastards. What you're doing is not only rightly illegal, it is immoral: you force your choice on others by distorting the playing field. You are creating an academy where nothing matters except getting ahead, whatever the cost. In the short run, that's no example to be setting for students, and in the long run, it can only be corrosive to the reputation of the academy, as well as of scientific research in general.

Heh.

"Reading Slashdot these days is like visiting the IT guy at work. He's infuriatingly smug and cares passionately about stuff you don't care about, and views your lack of interest as further confirmation of his intellectual superiority." [Time: Most Overrated Blogs]

On stupid questions in exams

I am, at this very moment, invigilating my final exam, and putting up with the litany of questions from the students, all of which either: (a) are basically asking me to tell them the answer; or (b) are easily answered by reading the question slowly and carefully, sounding out each word in your head, one at a time.

Inevitably it's the same few people who ask all the questions, and I can't shake the feeling that they're trying to get some advantage. Strong statements on the exam, such as "Ask no questions, if something is unclear make an assumption" do not dissuade these questioners. And you have to go up to them when they put up their hand, in case they need to go to the bathroom or something.

I would love to implement the following system. Every time you raise your hand, I put a red mark on your paper, which results in a one-mark penalty. The penalty applies whether I choose to answer the question or not. I have the right to waive the penalty if you actually found a significant error in the exam.

It would never fly, though.

Dear Nintendo

Playing your otherwise excellent game "Zelda: Twilight Princess," I've reached a stage where the hero is sometimes required to wear heavy boots to do certain tasks. Here is one example:

• Hero stands on a floor switch. The switch moves a little, but does not activate.
• Hero's companion suggests that he is too light to activate the switch.
  
• Hero puts on heavy boots. Now the switch is activated.

But here's the thing. I'm carrying the boots. How is it that putting them on my feet makes me "heavy" enough to do something, as opposed to schlepping them around in my pack?

It kind of reminds me of this.

On Influence

Consider Richard Reid. By any measure he is a failure as a human being. Yet every time I'm forced to remove my shoes at an airport, I can't help but notice that he has had far more influence on the ways of the world than I am ever likely to have.

This Week in Peer Review: The Academic Ad-Hominem

I wrote up two reviews this week.

One was for Trans. Comm. (yes, that one): it was co-authored by one of the more prominent professors in the field, and I found it to be a solid and interesting idea, though the writeup could use some polishing. I recommended acceptance, subject to revisions.

The other was for a different IEEE Transactions. The co-author was a professor whose papers I have seen before, and which have yet to impress me. This paper was like the others: a tired and unoriginal idea, with no particularly deep or interesting contribution. I recommended rejection.

But what has been bugging me this week is the idea that I may have let the names of the authors influence my reviews. I'm not saying that reversing the names would have reversed the decisions as a matter of course. But would I have been as easy on the rough writing in the first paper, or as harsh about the idea in the second paper? I'm not sure.

Dear university administrators,

You remember that big snowstorm yesterday? I sure do. Immediately after waking up, I checked your "weather status page", where I read: "The university is monitoring the situation. No decision to close the university has been taken." I also read the part where you said "A decision to cancel classes will be taken by 5:30 AM on the day in question." It was about 7 AM when I read this.

So like a good trooper, I trudged out into the snow and endured a slow ride on public transit to do the only thing I absolutely had to do at the university on Friday, which was to teach my 10:30 class. The class was about half full, which is frankly more than I was expecting given how much snow was coming down.

So I teach, and go back to my office at 11:30, where I open my email to find a panicky message reading: "WEATHER EMERGENCY -- University closed". To this I have two responses. First, what happened to 5:30 AM? And second, most people who were going to be at the university (such as myself) when you took this decision, in mid-storm, were already here. By not closing the university in advance of the storm you put everyone at risk unnecessarily, and your emergency closure improved the safety of basically nobody. Who is making these decisions?

We get letters

Signs of regime change over at Trans. Comm.?

Dear Professor,

Your review on paper number [redacted] titled [redacted] for the IEEE Transactions on Communications is now overdue for one week. Please try to complete your review as soon as possible.

I don't think I've ever received an overdue notification so quickly for any journal. (Yes, I'm a busy man and my reviews are often late, so I'm part of the problem too, but I rely on the editors reminding me of the deadlines, which they often don't.) I think this is a good omen. Even better would be a reminder a week before the deadline.

Clothes make the professor

Shortly after starting my tenure-track job, I had an epiphany while walking around campus in my jeans and T-shirt: I look just like all of them.

Since that day I have always made an effort to dress professionally for my job -- just collared shirts at first, but this year I've moved up to suit jackets (sans tie). I think it's working; I seem to get more respect from students, and even some compliments from colleagues.

Not everyone reacts positively. I'm not aware of anyone like this in my department, but for some reason, many professors view it as their God-given right to wear a burlap sack to work, and for them to dress nicely is to pervert academic freedom. I'm actually serious.

So it's nice to know that there are a few others of us out there.

This Week in Peer Review

I had two papers accepted for ICC 2008. Generally the review process was efficient and well-organized, in line with my previous experience of ICC. Here are my general observations:

• Paper submission deadline was Sep 28, 2007; anticipated decision date was Jan 10, 2008; decisions were received on time.
  

For such a large conference, a 3.5 month turnaround is quite reasonable.

• Paper 1: general subject area was wireless network security (submitted to the Wireless Networking symposium)
  

This paper received four reviews, which is impressive; I have never seen that before on a conference paper. Reviews were generally positive. Two of the reviews were brief, but the other two had substantial and helpful comments.

• Paper 2: general subject area was cooperative diversity (submitted to the Communication Theory symposium)

This paper received two reviews, which is average for an ICC paper. Both reviews, though generally positive, were short (like, 2-3 lines each) and not particularly helpful. One review contained a criticism of our paper that was either incoherent or logically inconsistent, I can't decide which.

Disclaimer: These views are my own and should in no way be construed as the views of my coauthors.

IEEE Transactions on Amateur Hour

I would like to bitch about IEEE Transactions on Communications, the 'flagship' journal of the Communications Society. Scare quotes are intentional.

It is hard to imagine how my experience with this journal could have been worse. In 2005 (!) I made two submissions -- a letter and a regular paper -- for publication. For the letter -- a short submission for which a quick turnaround is expected -- I waited 13 months for the first decision, which turned out to be a rejection. So thanks very much, that's over a year that I'm not going to get back to improve the paper. I'm not even mad about the rejection so much as the ridiculous amount of time taken.

The regular paper, though accepted, ended up going through a 2.5 year review process, which I'm sure would have been far longer in the absence of repeated and insistent requests for status updates from me. At one point the associate editor handling the paper had failed to respond to any of several repeated requests for an update over a period of about a month. Then, after the paper was accepted, I had to harass the publications editor to give me an estimate on the publication date. Finally, it is supposed to appear in the Jan 08 issue; yet, it is January 20th and the January issue has yet to appear. I checked -- this is pretty much the only IEEE journal that has not already printed its January issue.

The contrast is striking when compared to my experience with IEEE Trans. on Wireless Communications -- I submitted a paper last February, and it has already been accepted and is on the path to publication, probably later in the year. Pretty much everyone on the publication path was professional and attentive to their responsibilities.

My advice to any telecommunications researcher is to AVOID Trans. Comm. if at all possible. I did not encounter anyone at this journal who seems to realize that careers are on the line whenever they engage in this kind of thing.

UPDATE: Jan 21: the issue has now been posted.

I'm going to China

Both papers submitted for ICC 2008 were accepted, so I'm going to Beijing in May to present them. Hen hao!

Conversations with Steve Munro

As a frequent transit user, I often check out transit advocate Steve Munro's blog. I'm not entirely sure why, but I find myself frequently disagreeing with his viewpoint on the Toronto transit system.

A recent exchange on Blue 22, the proposed high speed link from Union Station to Pearson airport:

Me: The same arguments used in favour of high speed rail to the airport are also used to justify the continued operation of the island airport. Surely you would agree that the expense of Blue 22 would be justified by the death of the island airport, even if it makes no sense from a transit perspective (though I would dispute that as well).

Steve: I won’t speak for others, but my objection to Blue 22 is that this is a line that should be part of the local transit system, not a privately developed, premium fare service that serves a minority of potential demand in this corridor.

What does this mean? Do you mean that you like the polluting eyesore that is the island airport, or do you mean that you would rather wait for your preferred transit solution (which has been neither proposed nor planned by any level of government), while hoping that the island airport will go away on its own (which, in spite of Mayor Miller's 2003 mandate, is still alive and well)?

A million-dollar idea, in Monopoly money

You know what would be awesome? The ability to (easily) delete individual items from saved form information. I hate it when I misspell, e.g., a search term for Google and am confronted with my mistake every time I try to enter something similar.