Internet Archive DMCA Circumvention Lawsuit
The Internet Archive is a wonderful organization which keeps historical records of websites ('The Wayback Machine"). In Healthcare Advocates, Inc. v. Harding, Early, Follmer & Frailey, they are involved in a lawsuit alleging a violation of the DMCA. See William Patry for details and discussion.
The basic DMCA aspect of the case seems to be the Internet Archive uses (current website) "robots.txt" exclusion files to block access to archived material (they talk about "removing documents", but it's really "block access", not deletion). The defendants in the case were able to get to historical versions of the website in question anyway. The court complaint is unclear about exactly what happened. After thinking about it, the following is my speculation as to the technical aspect of the sequence of events (some material below taken for the lawsuit).
1) Fact - In order to decide whether to display the history version
of a website, the Internet Archive queries the current
website for the contents of the "robots.txt" file (to see if that
file prohibits the display or not).
2) Fact - The check is supposed to be done once per day.
3) Fact - There was a bug in this check, which led to the lawsuit.
4) SPECULATION - The bug in the check was that if an attempt to
retrieve the "robots.txt" file failed, that failure would be treated
as if no "robots.txt" file existed, and that means no block on display
(i.e. everything could be displayed for that attempt).
5) SPECULATION - At the time, this bug could be triggered by
repeatedly attempting to retrieve pages from the historical site
(which would, at the time, trigger repeated retrieval attempts of
"robots.txt", some of which might have failed).
Hmmm ... I hate to say it, but if the above is a correct reconstruction, it does begin to at least arguably look like an access control circumvention under the DMCA. The Internet Archive relies on external files to "control access" to archived website content. The defendants here found that sometimes it appears to the Internet Archive as if the external file wasn't present, via an implementation flaw.
I think it comes down to whether buggy "technological measures" still count under the law, and if taking advantage of a malfunction counts as circumvention by the user. It seems to be a much tougher case than it first appeared.
By Seth Finkelstein | posted in dmca , legal | on July 13, 2005 01:38 PM (Infothought permalink)
