<<< Date Index >>>     <<< Thread Index >>>

[IP] Google search and seizure, etc. vs. technologists





Begin forwarded message:

From: Phil Karn <karn@xxxxxxxx>
Date: December 3, 2005 7:10:30 PM EST
To: dave@xxxxxxxxxx
Cc: ip@xxxxxxxxxxxxxx
Subject: Re: [IP] Google search and seizure, etc. vs. technologists


From: Lauren Weinstein <lauren@xxxxxxxxxx>

1) Any practical attempt to "swamp" Google's database in such a
   manner is unlikely to succeed, given the sheer volume of legit
   queries that they receive.  I suspect they'd be smart enough to
   detect abuse patterns fairly easily.  That kind of analysis is
   their bread and butter.

The idea is not to "swamp" Google. It's simply to create a little plausible deniability -- i.e., reasonable doubt -- that a given search was entered by the user and not by the automatic daemon.

2) Attempts to purposely "abuse" Google in such a manner (faked
   requests) may well violate their Terms of Service, and if they
   don't now you can be sure that they will in some future version
   of the ToS.  The likely result will at a minimum be bans and ISP
   actions, and at the max lawsuits.  Pull out your wallet.

Again, "swamping" or "abusing" Google is not the intent, nor is it very likely given Google's strong emphasis on performance and scalability. The idea is simply to create doubt that a given query was generated by a human, not by the robot. The "quality" of the synthetic queries is much more important than their quantity.

Still, the extra traffic just might have the effect of encouraging Google to adopt a stronger privacy policy. Not that I'd place much stock in that, of course (see below.)

3) Routing queries through anon proxies will provide some protection
   for the technological elite who understand such things.  They will
   not protect the average user, who most likely doesn't understand
   the risks and issues, and will never use such proxies, even
   assuming that they were trivial to use.

I wish I had a nickel for everything I've been told "the average user" would never understand, need or be able to use. Back in the 1970s, the "average user" would never understand, need or be able to use a personal computer. In the 1980s, the "average user" would never need a local area network in his home. In the early 1990s, the "average user" would never understand or need the Internet. And so on.

It is no more necessary that the "average user" understand how an anonymizing Google proxy works to use it effectively than to understand the fields in TCP/IP packet headers. The whole idea of civilization and commerce is that many people can benefit from specialized knowledge and skills that they themselves lack. The open source movement and the Internet itself have certainly demonstrated this.

Personally, I prefer the anonymizing proxy over the random query generator. The proxy is likely to be more effective, and it generates no extra load. I mention the generator mainly to be complete. My point is that there *are* technical defenses against potential privacy abuses, and we can implement them ourselves instead of naively demanding that Google respect our privacy against their own commercial interests.

And even if Google were completely honest, they would still be subject to Patriot Act abuses that we would never know about.

The sad fact is that "national security" has become the root password to the Constitution. The only effective defense against a "rooted" system is not to put any sensitive information in it in the first place.

--Phil




-------------------------------------
You are subscribed as roessler@xxxxxxxxxxxxxxxxxx
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/