Typesetting my thesis

doctor snapping his latex gloveI’m at the point where I’m beginning to think about actually writing my Ph.D. thesis, and that means deciding what to write it in. My last experience in this realm was a disaster: with three hours to finish and print my Masters thesis, my sections automagically renumbered themselves, something that took me an hour to reverse. If possible I want to avoid a situation where my word processor “fixes” things for me.

My needs are simple: I want a pretty document, with a nice layout, and good bibliographic support. Here’s a quick rundown of my options:

Microsoft Word
World standard, good bibliographic support through Endnote, pretty good global style control, sucks for editing large documents
Adobe InDesign
No bibliographic support, handles large documents, but most features (such as endnotes/footnotes) must be done by hand
LaTeX
No interface, good bibliographic support through Bibtex, handles large documents, good global style control

Like many a Ph.D. student before me, it appears that if I can do without the interface, LaTeX is the obvious solution. No weird side effects, total style control and great support for all kinds of sundries like equations, bibliographic data, and so on. The only thing I worried about were the fonts.

Contrary to previous accounts, installing LaTeX on OS X was a cinch: using the ii installer, packages for teTex were installed on my machine in seconds. And something I was not aware of is the fact that OS X provides a system called XeTeX that has full support for OS X and TrueType fonts. Seriously brother, can a man ask for more?

I took the MIT Thesis LaTeX files, added some Mac fonts like Hoefler Text and Optima (using the fontspec package, which is a breeze), and blamo! I’ve got a great looking thesis. After struggling with my last thesis, this is a weight off my chest.

Hedonic treadmill

the hedonic hamster wheelI’m just about to return a book to the library, something I read a while back and have been meaning to post about for centuries. In their article “Hedonic Relativism and planning the good society**,” Philip Brickman and Donald Campbell give a name to the ongoing state of happiness that we all experience. Despite the fact that external forces are constantly changing our life goals, happiness for most people is a relatively constant state. Regardless of how good things get, we’ll always be about the same level of happy; this they call the hedonic treadmill.

Psychology researchers have observed this phenomenon in a myriad of different situations: lottery winners, tenure achievers, recently handicapped, etc. In all of these situations, despite a massive shift in standard of living or achievement of major life goals, after a short period of time the life-satisfaction levels return to normal.

If this is what we can expect from our own psychology, how does hedonic relevatism affect the way we choose to live our lives? Brickman and Campbell look at this question from a societal level, and suggest that there is an optimal setup for making every member of our culture as happy as possible. You have to give them credit, it was the 70’s and socialism was still a form of utopia. But as far as I can tell, the only way to keep yourself on an increasing scale of happiness is to achieve some small goals on a daily basis, not putting too much emphasis on achieving one over another.

So why am I writing this damned Ph.D.?!

** Brickman, Philip, & Campbell, Donald. (1977). �Hedonic relativism and planning the good society.� In M.H. Appley (Ed.), Social comparison processes: Theoretical and empirical perspectives. New York: Wiley/Halsted.

Bicycle Defense Fund

An organization has arisen in NYC to help support those individuals arrested during Critical Mass events under the name Freewheels Bicycle Defense Fund. This past weekend they hosted an inaugural fund driving event in Brooklyn that apparently made upwards of $4k for the legal battling bikers.

According to my friend Jamie, most, if not all of the arrested cyclists have refused offers of Adjournment in Contemplation of Dismissal (ACD) and decided to take their cases to court. Given the number of cases already pending, along with the expected number to arise from the first big ride of the year in April, the Critical Mass defendants will be needing quite a bit of support for what could be a very defining court battle for bikers rights in the city.

Laptop Withdrawal Affective Disorder

The symptoms started about two weeks ago when I first noticed that my 12″ Powerbook’s hard drive was on the fritz. My applications slowed down, movies stopped playing halfway through, and I was finding it hard to get out of bed in the morning. Then today when I went into my local Apple store for a diagnosis, things really started to set in. I’m grumpy, irritable, and constantly frustrated. I’m afraid I have classic symptoms of Laptop Withdrawal Affective Disorder (LWAD).

In all honesty, I’m telling the truth. Well, maybe the part about not getting up in the morning is more about me being lazy, but I am generally off since I realized I’d be without my laptop for a week. It’s not the service it provides, per se, it’s how it’s integrated itself into my life. I’ve had relationships with a few PC laptops in the past, but my Powerbook seems to be the first laptop I’ve ever truly loved. Without it I feel lost, aimless, and like I spend most of my time thinking about how much easier everything would be if I didn’t have to use this damn PC desktop.

It’s definitely a testament to the quality and usability of Apple’s laptops. Sure they’re not the fastest, lightest, or most eye-catching machines around, but they generally work better than anything else. Maybe Apple would settle out of court if I sued them over my current condition. I think my expensive AppleCare support should include some sort of counseling for my condition. Something—anything—to make this pain go away.

OS X keyboard locking issue

I’ve been dealing with this problem with my Powerbook 12″ lately which has been the source of neverending frustration: while using my laptop occassionally the keyboard will stop working. For a while I thought that the only way to cure my ill computer was to reboot it, and with Google providing no help, I was just about to stick my silver friend in the microwave to teach it a lesson.

Over the past month or so, I’ve discovered quite a few features of this peculiar bug. The first thing I recognized was that not all of the keys were frozen: most of the option and F-keys still provided their full functionality, such as App switching (Commant-Tab), brightness control (F1-F2), etc. Still nothing would repair this ailment, and I’d eventually end up rebooting.

But last night I made a major breakthrough: while locked, I tried to Force-quit (Command-Option-ESC), and suddenly my keyboard was returned to normal. I quickly tried seeing if it was related to any of my running applications: not iTerm, not BBEdit, none of them seemed to affect it. But, while switching applications once I tried to not use Command-Tab, and low and behold, the problem went away entirely.

Whenever I use Command-tab to switch applications now, there’s a good chance that my keyboard will lock. Has anyone ever encountered this problem before? Either I’m a really peculiar user or it’s something specific about the applications I use (I just reinstalled last week).

Very strange.

Update: I realized that when the machine gets into this state, any Cmd-Tab operation will render the keyboard unusable (until you run force-quit), but a reboot returns things to the normal operation. Relaunching the Finder has no effect, so it’s as if the kernel gets into a state where application switching causes the keyboard to hang.

Props to Cincinnati/Northern Kentucky

Cincinnati/Northern Kentucky International AirportI’m supposed to be eating Tex-mex in Austin right now, reliving old SXSW’s with that group of people I only see this time of year. Instead I’m in a Courtyard Mariott somewhere in Northern Kentucky. Apparently not the part where the bourbon flows freely from spigots one every street corner.

But I’m not mad, I’m glad. The room is comped, my flight isn’t too early in the morning, and they have free high speed internet access. I guess they do this sort of thing often in Cincinnati because after landing they asked people to raise their hands if they had a connection to make. Even the staff raised their hands. It was a straight up connection party on that plane.

The hour and a half spent on the tarmac in Boston cost Deltinental severely, because I’d say a good third of the plane missed the last flight of the night. When we all staggered to the ticket counter, there was a giant sign that said “hotel accomodations,” and they had already prepared vouchers for room, board, and pre-printed our tickes for the morning. Now that’s what I call service.

So if you’re going to fly through a hub during the snowy season, make sure to fly through Cinci, because they won’t give you this “we’re not liable for flights delayed due to weather” bullshit. They give you the red carpet. Or something. In fact, in 2004 it was rated Best Gateway Airport in the US of A.

Stopping comment spam with keystrokes

In response to the recent post by Simon Goodway, I threw together a simple implementation of the keystroke-approach to comment spam blocking:

MT-Keystrokes

It’s quick to install (one small file and one quick change to templates with a comment form). So far I’ve seen a 100% decline in spam, so much so that I’m thinking about turning off Moderate and going old-skoole with my comments.

If the spammers figure out the field I’m using, it’s trivial to change, and could even be automated. There are a number of ways to make it more difficult for spam purveyors, but my thought is that this 5% added difficulty will never be worth their while.

Weblog ping services

Given the sheer number of weblogs that exist, and more importantly the number of those that could be dead, knowing which weblogs have been updated and when is a critical piece of information. The solution to this problem is a notification service, whereby weblogs alert a system when they’ve been changed. Pioneered by weblogs.com, ping services are the linchpin of every major weblog aggregator.

Nowadays, every weblog author has a number of options to choose from when they setup their blog. The big three are:

There are a number of other ping services for specific communities (e.g. geographic, topical, etc.), but most people decide on the big three. In addition to these choices, the Pingomatic meta-ping service has emerged as a way to easily manage your pinging adventures. Since most of the smaller systems also troll Weblogs.com and Blo.gs, this seems like a suboptimal solution. Why ping 15 individual services when you can ping just one and get the same effect?

As for who receives your ping, Technorati, Bloglines, and Feedster are all closed systems, and pings sent to them are available only to their service. As the weblog economy grows, there will only be more and more competition for each ping, and I assume these companies will protect their data, and for good reason. Any additional updates they get above and beyond the free, open services provide an advantage over other companies. For me, the choice of an open ping system is obvious, but i fear that new webloggers will choose the only name they recognize and many smaller services will lose out at the expense of better-marketed systems.

In the realm of open ping systems, Blo.gs has taken a leap ahead with respect to the efficiency of ping delivery. Most ping services use HTTP as a delivery mechanism, requiring a user to poll a URL to get an XML list of recently updated sites. This is highly inefficent given the nature of the process, namely using a pull technology to push information. Probably due to demand, Blo.gs has moved to a push system whereby blog aggregators can recieve updates as they roll in.

While the future of Blo.gs is unknown, their source code is available, and it has remained a freely available system from the outset. As the author of a blog aggregation system, their efforts have made my life that much easier over the course of the past few years. I sincerely hope that this system, or another open system like it becomes the industry standard for providing update information. Otherwise many of the smaller weblog systems will suffer.

Fill My Closet

nordstrom's gift cardOver the past few days I’ve received emails from a few friends asking me to sign up for a service called “Fill My Closet,” which appears to be yet another viral marketing campaign.

Their proposition is simple: sign up for a free trial from a host of companies, get 5 of your friends to do the same, and get a free $250 gift certificate to Nordstroms. Paypal was the first company to utilize this sort of strategy when they paid cold cash for every new signup a user could generate. I fell pray to this offer (who could resist free money?), as did millions of other people. Paypal subsequently turned out to be a great service, and I would have signed up anyway.

FillMyCloset on the other hand is essentially a viral marketing frontend for all of the old school junk snail mail—magazine subscriptions, free product trials, cd clubs—seemingly great offers with huge drawbacks. These junk services are brokered by a company called MetaReward.com which allows people to easily sign up for services (one-click shopping for thigns you don’t want) Unlike the postal service, which is a random attempt to find gullible people, FillMyCloset uses social networks to find cliques of gullible people.

If each person that completes the required networking service, and gets 5 people to sign up for a service, the cost to the service provider would be $50. I’m sure that none of the participating companies would pay that much money per new customer, so the service must depend on people not being able to quite achieve their goal. If they set the number of friends too low, everyone will succeed, and if they set the number too high, no one will think it’s possible. Five seems to be the magic number, although I’m not sure why it’s so hard.

The site is a service of Free Super, an incentives-based viral marketing company. Their website provides very little information about the company, except that they can “generate thousands of new customers every month, at a fraction of the legacy cost of acquisition.” At least they’re upfront about their sliminess.

Google’s AdSense spam

Google has embarked on a pyramid-based incentives program directed at bloggers. They’ve been pushing really hard for me to put AdSense on Blogdex, presumedly because it has high PageRank. I was first contacted directly by someone located in New York under the pretense of a “potential partnership between Google and Blogdex.net.” This was a personal email requesting me to contact them directly (their name has been changed to protect the innocent):

Subject: Blogdex.net partnership with Google
From: John Doe <jdoe@google.com>
To: blogdex@media.mit.eduI’d like to discuss a potential partnership opportunity between Blogdex.net and Google. I work here in the New York office.

Please let me know when might be the best time to follow up with a call.

thanks,
-John

John Doe
Google, Inc. | Partner Development
1440 Broadway, New York, NY 10018

I called them back, not sure what this “partnership” entailed. It turned out that they wanted me to put AdSense on Blogdex, nothing more. I politely declined noting that it was a research project on academic bandwidth, and forbidden from containing advertising.
Continue reading “Google’s AdSense spam”