3 ms·
I'm sorry its a bit hard to have this knee-jerk reaction without at least having some background into the matter. Did Aaron in fact break into any systems? Or d
by dlikhten 15y ago
I'm sorry its a bit hard to have this knee-jerk reaction without at least having some background into the matter. Did Aaron in fact break into any systems? Or did he legitimately access data and then get "flagged" for knowing too much? http://en.wikipedia.org/wiki/The_Man_Who_Knew_Too_Little http://en.wikipedia.org/wiki/The_Man_Who_Knew_Too_Little
- aw3c2 15y agoI had the same reaction like you (I think). This is a current affair and it seems not clear what exactly happened and why. I would not sign any petition before knowing more.
- shareme 15y agono he circumvented the MIT guest system to download those JSTOR DOCs..
- apgwoz 15y ago"circumvented the MIT guest system"? If by circumvented you mean, registered a fake name, clicked accept and got an address via DHCP, then yes. edit removed the "I bet you're guilty.." and added the below: Should Aaron have asked for permission from JSTOR first? They seem to have asked the feds not to prosecute, they might have been willing to help originally. Maybe he should have, but JSTOR is unlikely to say "yes." So Aaron did what most of us would have done. He wrote a script to acquire the information he wanted to--fine. Have you ever mirrored a website? It's likely that you felt guilty doing so, and maybe even went to your local coffeeshop to do it on their network instead. You don't want your IP banned. Aaron seems to have done basically the same thing, but instead of a coffeeshop, which wouldn't allow him access to JSTOR, since the access is via proxy, he went to a neighboring college campus that likely had a JSTOR subscription. Now, the thing that's troublesome to me is that MIT didn't stop him while he was doing it in the first place. Surely they'd have noticed 30,000+ requests from the same IP to a protected resource, no?