Subpoena Sought For Browsed News Articles

The Xoxo Reader writes "A new filing in the Autoadmit Internet defamation lawsuit (previously discussed here on two occasions) reveals how the plaintiffs' lawyers have attempted to discover the identities of the defendants, who posted under pseudonyms on a message board without IP logging. The defendants had posted links and excerpts of several Web pages that mention the plaintiffs, including a Washington Post article, a college scholarship announcement, and a federal court opinion. Now the plaintiffs are asking those Web sites for logs of everybody who accessed those articles in the hours before the allegedly defamatory content was posted. (All the more reason to read the web through Google cache!) The plantiff's motion for expedited discovery includes copies of the lawyers' letters to hosting providers, ISPs, and others. It also includes replies from the recipients, many of whom point out that the lawyers' requests are technically impossible to fulfill. No matter; the plaintiffs are asking the court to issue subpoenas anyway. This thread contains a summary of the letters in the filing."

Comments

Google Cache doesn't help

[dereference]

(All the more reason to read the web through Google cache!) You're horribly misinformed if you think that will help. Any images, scripts, or other "external" content referenced from the cached HTML is still fetched from the original server by your browser. Unless it's a self-contained text-only page (as if many of those exist) your IP address is still in their logs.

Re:Google Cache doesn't help

[strider1551]

(All the more reason to read the web through Google cache!) All the more reason to use Tor. There we go...

Re:Google Cache doesn't help

[martin-boundary]

It's worse than that. Google (cache) is a single point of failure. All it takes is a valid subpoena from a court to the company Google, and Google will divulge all the information it has about an IP or a user account etc. That may include web pages visited, search terms, emails, etc., eg your whole online life.

From a privacy perspective, it's a lot better if the information that the lawyers are looking for is physically scattered over many websites, each owned by a separate legal entity with different hardware capabilities, different backup strategies, different technical expertise, and different political beliefs.

Remember, the company Google has enough information already to reconstruct your email conversations (if you use gmail or you exchange emails with someone who uses gmail), your searches terms on Google and partner search engines (agreements to pass along complete search histories to Google in exchange for help with searching), and the actual pages on third party web sites that you visit (world wide text ad tracking by Google/Doubleclick is enough for browsing history reconstruction).

Proxies?

[Idiot with a gun]

This is a pretty good reason to view the web through a proxy, or even Tor (if you can stand the performance hit you'll take).

Re:Proxies?

[DustyShadow]

Many of the users of autoadmit do use Tor now. Threads about it pop up every now and then.

Re:hmm

[nomadic]

Unless there is some sort of local rule or local law I'm not noticing that's not right at all. For parties attorneys can make requests without going to a court. But, for non-parties, which these miscellaneous ISPs would be, they need a subpoena from the court to get the info. If the holders of the info are parties then yes the attorneys could get it by just asking. Assuming the other side doesn't refuse for whatever reason they might have, then a judge would be called in to settle the fight.

Under Federal Rule of Civil Procedure 45 an attorney can sign and issue subpoenas for third parties, as an officer of the court. I know it's the same for my state, and considering most states base their rules to a large extent on the Federal ones, I'm assuming it's the same for most jurisdictions.

Can't issue subpoenas until discovery begins

[HazelrahXo]

This explanation was posted on Autoadmit:

Plaintiffs can't issue a subpoena without the court's permission since discovery hasn't began. That's the whole nature of the motion for expedited discovery.

http://www.autoadmit.com/thread.php?thread_id=753317&mc=167&forum_id=2#9222459

Text of some of the letters

[HazelrahXo]

Excerpt from letter to HighBeam Research, from page 48 of the Justia link. The plaintiffs actually asked for a week's worth of access logs:

We have reason to believe that at least one defendant used your service to post allegdly tortious content regarding out clients on AutoAdmit.com and/or to email that content to third parties. That defendant posted or sent that information on March 7, 2007 at 5:23 p.m. EST after using your service to access an article titled "Ex-World Bank Official Disappears From Trial," which is found at http://www.highbeam.com/doc/1P2-915261.html.

We hope that, given the egregious conduct that is alleged in this case, HighBeam will disclose certain information that may allow us to determine the identities of the named defendants in this lawsuit. To that end, we request that HighBeam provide us with information that identifies the person(s) using your service to locate the above-identified article, including but not limited to first and last names, present or last known mailing addresses, telephone numbers, e-mail addresses, and all logs containing the source Internet Protocol ("IP") addresses of all access to the location or file "1P2-915261.html", "915261" or other variants thereof between 12:00 a.m. EST on March 1, 2007 and 5:23 p.m. on March 7, 2007.

Response from HighBeam, page 86:

We spent a fair bit of time determining whether it was technically possible to comply with this request. We have determined that our system simply does not allow us to obtain the information you are seeking about the identity of a person accessing a particular story at a certain time.

I cannot begin to explain to you the complex technical reasons for this. However, if you have a technology consultant working with you, I will be happy to put that person in touch with our CTO for a complete explanation.

Not necessarily; &strip=1

[Xelios]

Copy the cache URL, paste to address bar and add &strip=1 to the end. It will ignore anything not in the cache'd record. You won't see any images and the like, but you also won't end up in the website's logs.