Alex A. Hutnik

Yes, that title is making fun of some of the noobs on DC++ (and the internet in general — think about your typical MySpace user, then you’ll get the idea). If you still don’t understand, you’re probably better off.

For my Econometrics (ECON 326) class, with the help of my good friend Colin, I’m running a custom DirectConnect client that records searches. I’ll be publishing my report in about a month and a half. Until then, I’ll be collecting data on searches. I’m using the number of searches as a metric of DC++ activity on campus. The more searches, the more active the network is. TTH searches occur when other people were downloading files and were disconnected. Now their computer (if properly configured) is searching for other copies elsewhere. Since the IP address comes along with the searches, I can attribute searches to buildings on campus. And, since I know some buildings are freshman or Sophomore only, or Greek or otherwise, I can attribute certain searching activity to a particular segment of the student body. Here are some general things I’ll be able to determine:

• Are Case students more likely to be using the network during prime “going out” time (Friday nights, weekends, breaks, etc) or not. Or more exactly, are freshmen using the network more on Friday nights than upperclassmen?
• Which fraternity searches for the most porn? Which freshmen dorm residence hall?
• What TV shows have popular followings? Are people searching for “Heroes” more than “Star Trek?”
• Is there any correlation with the people living on South Side having less to do because they are geographically isolated so are more likely to use DC++? In other words, are South Siders using the network more because they are away from almost everything else?

You get the idea. Stay tuned for updates and for my final report, which should be published here some time before Winter Break.

kthxbye.

Popularity: 6% [?]

Premise

Case incorporates the JA-SIG Single Sign-On service as a way to offer convenient, cross-service authentication on services like the Case wiki, blog, and Blackboard. The basic idea is that a user connects to a service (let’s say blackboard.case.edu), which in turn forwards them to the login.case.edu SSO service. The SSO service issues the user a ticket for that service. On consecutive connections to blackboard.case.edu, the server checks the ticket submitted by the user against the SSO service to see if the user is still logged in. If yes, the session continues. If not, they are requested to log in.

Tracking Visitors

The SSO service provides a ‘user look-up function’ that allows web developers to take user-submitted tickets and see what Network IDs are registered to them. This is how Blackboard knows that you are abc123 when you log in. This also allows web developers the opportunity to track website visitors, even if their site doesn’t require authentication for anything. I set up a demo page here. You can click on the link at the bottom to see the other Network IDs I’ve recorded.

So What?

This feature allows visitors of the site to be tracked and targeted for advertising. Anyone with a computer on the Case network can use the SSO service to track visitors to their website. It would not be a stretch for people to use this information, combined with LDAP to match Network IDs to real names and the IP Subnet lookup page, to determine who accessed their site, and from where. Based on this information, you could determine plenty about your users and how they’re using your site. A simple JavaScript snippet could call the X and Y coordinates of the user’s mouse and asynchronously send them back to the server. Later, those coordinates could be played back to see the user’s mouse movements and gestures. Do some people highlight the text on your page with their cursor? Maybe that’s because your shoddy CSS makes the text look indistinguishable from the background.

By being smart about user tracking, you can glean a lot of information about your site’s visitors.

Popularity: 14% [?]