Monday, June 13, 2005

Wifi Hotspot Statistics

Good evening.

I was at it again today. Besides cooking, I was collecting data for my study. I have expanded the scope of the study slightly. Instead of focusing simply on the length of connections, it would be interesting to see if the Internet workload is different depending on your connection type. Specifically, is there a specific set of network resources (email, certain websites, etc) that are more often used when temporarily connected to a free access point? Or, do people treat these free wireless hotspots as a second office where they perform the same routine network activities they would from their home or desk?

First and foremost, people need to be aware that when they are connected to these access points, there is very little security. As people sit down and start up Outlook to check their email, a password may be transmitted in plain text making it available to anyone sniffing the network. This security gap can easily be corrected by using a webmail client that offers HTTPS connections or using IMAP with certificates.

If you were to look at the statistics I collected today (displayed to you, the reader, in the following pie chart), you would not notice anything out of the ordinary. However, I was surprised by one thing in particular. It was part of my initial hypothesis that users of these free access points would primarily do web/email. I figured there would be a little AIM chatting also. But, I never imagined that someone parked in a cafe would fire up a p2p app (Gnutella) and start downloading. I have not looked at the specific packets yet, so it is possible that these packets are a result of a system tray application that automatically executes on connection to any network and do not represent a user request. Nevertheless, I am surprised.



I am going back tomorrow for round three of data collection. I wonder if I'll see anything new.

No comments: