I'm having a hard time separating the surveillance model and the capture model. In fact, I feel the two are inseparable in modern society. I'm at a loss to find an example of surveillance that doesn't capture the information it surveys. Nor can I find an example of capturing information that doesn't survey the subjects of capture. One example that Agre presents is the experiment of tracking in the offices of Xerox, where employees where a self-locating badge. I feel like this follows both models of capture and surveillance. The information of the employees' whereabouts is instantly relayed to a database with the ultimate goal of making the workplace more effecient. I feel if anyone were 'out of place', this information could be instantly processed, and the situation rectified. Maybe I'd be an extra-paranoid employee, but information can be captured and processed so quickly that I'd be extra careful where I went throughout my workday. Another example is my Dad's laptop from work. He won't let me touch it. He has no idea how his company tracks its use, but he isn't willing to chance it with non-work related material. He responds to the capture of his information in the same way you respond to a security camera - you don't know if anyone is watching, but you act as though somebody is.
I see how the two differ - the model of the Panopticon versus the model of information capture, one being more transient and the other more permanent. The two, however different in operation, produce similar results in the groups to which they are applied.