Saturday, October 19, 2013

Everybody Knows You’re a Dog

[ed. Evercookie = malware]

Remember 1993? The World Wide Web had already been invented and nobody knew about it. The NCSA Mosaic browser had just appeared in a limited alpha release, but the text-based Gopher service was the closest thing most people had to an interactive user interface to dive into information on the Internet. The commercial use of the Net was still extremely limited in America, as the National Science Foundation’s NSFNet backbone ostensibly prohibited anything but academic use. In other countries, the Internet existed barely, if at all.

Into that void, New Yorker cartoonist Peter Steiner dropped what is now the most popular and well-known panel ever produced for the magazine: "On the Internet, nobody knows you're a dog."

In 1993, the Internet was still a great mystery to most people, if they’d even heard of it. America Online (AOL) was the way most people got “online,” and email was the primary use of the Internet — and for people who weren’t connected to an academic institution or using one of the early Internet service providers (ISPs), this happened via gateways from AOL and other walled gardens.

Steiner’s single panel manages to convey the Internet’s newness and the personal purpose to which most people then put it, as well as its fundamental anonymity. The tools to track someone down or even demand a “real” name were nonexistent. One could be a dog (cats came later), and no one would be able to tell the difference. It was a time that now seems remarkably innocent. (...)

When the “dog” cartoon first appeared, pseudonymity on the Internet was a given, and anonymity wasn’t difficult. With the onset of commercial uses of the network and the appearance of widely available graphical Web browsers later in 1993, general anonymity became even easier because one could pay an ISP for access, but the ISP didn’t give two figs what you did online — nor, in those days, could it possibly have had the routers or storage to track behavior if it had wanted to.

One could post comments all over with little or no connection to one’s identity or location. The Web is inherently stateless — there’s no idea of a continuous session built in — which made it hard in its early days to allow for the association of a browser with an identity. The later addition of Web cookies allowed servers to push a unique code and other information to browsers, which browsers would send back with each request. This strings connections like pearls on a necklace and allows continuity, which allows an account and a login.

Such tracking doesn’t necessarily destroy anonymity, but it does reduce it. When a Web site or outside authority has only an IP address and other browser information by which to identify a user uniquely, the task becomes harder, as many networks share a single numeric address on the public-facing Internet and use private addresses internally. And the public addresses can change over time.

Marketers, of course, want to erase anonymity and even pseudonymity, because the less knowable an individual is, the less value that person has to advertisers. The more accurately they track you, the more lucrative it is to sell ads that cater to you or shop your data to other parties that combine online details with real-world purchasing behavior and credit records.

Over the years, Adobe Flash-based cookies and other breadcrumbs that allow tracking over many sites (or that bypass users’ ability to defeat such tracking) have been baked into plug-ins and browsers, often as unintentional byproducts. The “evercookie” proof-of-concept revealed that it was nearly impossible to kill all the stateful tracking elements for a site that wanted to keep you in its sights.

by Glenn Fleishman, Boing Boing | Read more:
Image: Wikipedia