Enemytalk, Power Laws, & My Favorite App

Last week I finally delivered my talk at Columbia University. I must admit I was a little nervous beforehand. After doing so many interviews about Enemybook, I've become practiced at speaking with reporters, but this was my first time giving a whole Enemytalk, and I didn't know exactly who my audience was going to be. Thankfully the folks at the EdLab unit of Teacher's College were gracious hosts, and provided a fun and informal atmosphere. They also asked some good questions.

Two questions in particular stood out. The first was technical. Someone wanted to know whether the expected out-degree of the nodes in the enemy graph followed a power law. This is a very interesting question, but since the question itself may have lost half the people reading this blog, I'll devote a whole post to it later, where I'll explain it in more detail. Right now I want to address the other question, which wasn't technical at all. The question was quite simply, "what's your favorite app?"

Now I admit I was hard pressed to think of an app that I even liked, much less my favorite. Since I was on the spot, I mentioned one of the first apps that came to mind, Introductions by Wayne Mak. The idea behind Introductions is that you can use it to ask for your friends to introduce you to people you might want to meet (e.g. "someone with film experience," or "someone who can make a facebook app"). While I like the idea behind Introductions, at least in principle, in retrospect I think it was a bad suggestion. Why? Because I never actually use it (and since I last used it, it's changed quite a bit into a sort of "hot-or-not"-esque game, so while perhaps it's useful for a different purpose, it's not exactly the app I had in mind).

After much contemplation then, I'm announcing that my real favorite app is The Compliment Machine, by Aaron Iba and David Greenspan. This incredibly useful app contains a highly sophisticated robot that generates "laser-accurate" compliments. It comes recommended by no less than Sir Winston Churchill, who was so stirred by the app that he rose from the grave to review it. I can honestly say that The Compliment Machine has given me as much enjoyment as any other app out there, without spamming a single one of my friends. It's really a shame that last I checked, it only has 6 daily active users.

One major suggestion for The Compliment Machine: I want the little robot to appear on my profile, so that my friends can use it to complement me! I'm guessing that Aaron and David haven't added this feature because they're too busy founding their own company, Appjet, which provides a tool for quickly developing web apps (for Facebook and more). The Complement Machine was written in Appjet, and they've open-sourced the code, so hopefully it's only a matter of time before somebody comes along to add this very important feature.

Enemybook to Bring World Peace

Hot on the heals of Enemybook's appearance in Timeout London, I've been asked by a fairly well-known NGO to make a more serious version of the program. The goal of this NGO is to prevent genocide, and their hope is that a more serious Enemybook will act as a sort of war criminal "watch list." The idea is to help educate the public about some of the alleged perpetrators of atrocities being committed internationally (specifically in Darfur), and get people to put pressure on the UN to stop these atrocities and bring the appropriate people to justice.

Admittedly, I have my doubts about the efficacy of a Facebook app in furthering this (or any) humanitarian cause. But if there's one app that can stop genocide and bring world peace, Enemybook is it. Time permitting, I'm going to see what I can do for this organization. More details to follow in the coming weeks.

Facebook Cracks Down on "Forced Virility"

The people who work at Facebook read this blog religiously. At least I assume they do, because that's the only explanation for the news up on their developer site. Responding to my earlier post (as well the academic research backing it up), Facebook now explicitly bans what I termed "forced virility:"

"... as part of our ongoing efforts to improve Platform (sic) through policy and technology changes, applications are prohibited from dead-ending users at an invite-friends page, and must never again prompt for invites after the user has declined.

More generally, unnecessarily gating access to application features behind inviting friends is not a best practice...it is misleading to entice an investment of effort or promise a result and then -- without warning -- hold expected content hostage behind an invitation ransom; this is now expressly prohibited...."

This is a long overdue move on Facebook's part. Right now there's no way for Facebook to automatically enforce the new rules. They'll have to rely on users to report apps that are in violation. Most users probably won't do this, because they won't know that the apps are doing anything wrong; the precedent for forced invites has already been set.

It'll be interesting to see when (or if) things change.

Improvements on the Way

I've been meaning to make some improvements to Enemybook for awhile now, but grad school and other side projects have prevented me from doing so. Recently I've been preoccupied with a paper deadline, but that finished up today, so hopefully I'll have some time in the next couple of weeks to get cracking on Enemybook. Improvements I want to make are:

1) The ability to enemy "pages" for products/celebrities/politicians. Right now you can sort of do this, but the pictures don't display properly on your profile.

2) The ability to enemy groups.

3) The ability to define your own enemies by specifying a name and uploading a pic. This way people can get more creative about who/what they enemy, without having to go through the rigmarole of creating dummy facebook profiles.

Of course there are plenty of other improvements that could be made, but for now this is more than enough to keep me busy. Any suggestions for other things you'd like to see?

Enemybook Gives Computer Science a Bad Name

A few days ago The Chronicle of Higher Education wrote a piece on Enemybook. I know what you're thinking- it's about time, right?

What's interesting is the reader response to their article. The comments section sparked a debate about whether computer science deserves to be called a science. Normally I don't like to get embroiled in such debates, but in this case, I'm tempted. One of the commentators compared me to a drug user, while another commentator compared me to someone who develops drugs (specifically anti-viagra).

Now let me clear up some misconceptions. Though I am on drugs most of the time, when I'm not, I am indeed hard at work designing drugs to cause impotence (I'm almost done- I call them SSRIs. They're sure to be a huge hit.).

Second, the question "Does computer science deserve to be called a science?" is the wrong one to ask. The real question is "Does the research done by computer scientists have value?" The existence of Enemybook is largely irrelevant to that question, since Enemybook is not at all indicative of the research done by computer scientists. I'm an academic, and I made Enemybook, but I did so as a side project, not an academic pursuit (not yet, anyway). From a sociological perspective, social and anti-social networks may very well be worthy of academic inquiry. But it's not the kind of thing most computer scientists study. My own research to date has been on algorithms, a rather different area (for the curious, you can find my papers here). Criticizing all of computer science because you don't like Enemybook makes about as much sense as criticizing all of chemistry because you don't like a particular chemist's cooking.

Other posts in the comments section talked about freedom of speech and privacy concerns. Though I don't think these issues are much different for Enemybook than they are for Facebook itself, they are complex issues and I don't want to comment on them glibly. I'm glad people are thinking about them though. After all, the real intent of Enemybook is to first make you laugh, then make you think. Let's hope the Ig Nobel prize committee is listening.

The Ivory Tower Backs Me Up, Again

I've been invited to talk about my experience with Enemybook at the Columbia University Teacher's College, in their Edlab seminar. I expect my talk won't just be about Enemybook, but about the entire space of social apps.

The seminar is currently scheduled for Feb 27, 2008. More details to follow.

UPDATE: The seminar has been rescheduled for Wednesday, March 12, 2008 at noon. Details will be posted here: http://edlab.tc.columbia.edu/index.php?q=node/1710

The Ivory Tower Backs Me Up

I am pleased to report that my anecdotal observations from the previous post have been confirmed by empirical research in academia. The ACM Transactions of the Web published the following paper by Leskovic, Adamic, and Huberman entitled The Dynamics of Viral Marketing. The authors analyze a recommendation-referral system run by a large retailer. They conclude:

"Marketers should take heed that even if viral marketing works initially, providing excessive incentives for customers to recommend products could backfire by weakening the credibility of the very same links they are trying to take advantage of."

This is part of a rapidly growing body of academic work on social networks. It's only a matter of time before academics start studying anti-social networks as well...