1996-09-11 - Re: shutting down anon.funet.fi – tough case

Header Data

From: Greg Broiles <gbroiles@netbox.com>
To: Law & Policy of Computer Communications <CYBERIA-L@LISTSERV.AOL.COM>
Message Hash: 39ab85e79322acfc86b7ddc5d12297cb29a104aaeaf880715d912dd6120dbd1f
Message ID: <2.2.32.19960911000616.00711a9c@pop.ricochet.net>
Reply To: N/A
UTC Datetime: 1996-09-11 03:01:49 UTC
Raw Date: Wed, 11 Sep 1996 11:01:49 +0800

Raw message

From: Greg Broiles <gbroiles@netbox.com>
Date: Wed, 11 Sep 1996 11:01:49 +0800
To: Law &amp; Policy of Computer Communications              <CYBERIA-L@LISTSERV.AOL.COM>
Subject: Re: shutting down anon.funet.fi -- tough case
Message-ID: <2.2.32.19960911000616.00711a9c@pop.ricochet.net>
MIME-Version: 1.0
Content-Type: text/plain


At 05:22 PM 9/10/96 -0500, gene o'regon  wrote:

>        the forced closing of the anonymous remailer from finland shows how
>the law has to adapt to the internet.  aside from the witch-hunt aspects of
>the case ["they are the biggest porno distributor, etc."], there is real
>substance to the claim that you can't just ignore blatant violoations of
>coyright law.

As I understand the situation, penet.fi wasn't forced to close but will
apparently be forced (again) to compromise the anonymity of one of its users
- and the operator, Johan Helsingus, chose to stop providing the service
because he isn't able to protect the privacy of the users. 

It also might be confusing to think about anon.penet.fi as a single "case",
since it's been the subject of frustration/attention from several directions
- the church of scientology, the government of Singapore, and the recent UK
newspaper article identifying it as a source of child porn. 

Further, I don't think there's been a real claim that penet or its
operator(s) have violated any copyright at all; merely that they have
frustrated the efforts of copyright holders to identify alleged infringers. 
I don't see how any service provider of even modest size can fail to "ignore
blatant violations of copyright law" given the difficulty of figuring out
who's got a right to make which data available.

Even a relatively easy-to-spot infringement (say, an image with "Playmate of
the Month" text visible) requires that some person who's familiar with
Playboy view the image and come to the conclusion that an individual user
probably doesn't have the right to make images available which were
apparently produced by a well-known international publishing operation. The
creation of a prescreening systen would require the development of
protocols/systems which do not now exist, as well as burden the ability of
users to communicate quickly. It's also expensive in terms of human time
required, and likely to fall far short of its mission - e.g., people can
perhaps spot Playboy centerfold images by their vertical-to-horizontal size
ratio (I mean the image, not the model) plus the "Playmate of the month"
text, and they can probably spot popular songs or images .. but matching two
images or arbitrary selections of text is a difficult problem.

Prescreening seems to require access (on the part of the screener) to a
database of already copyrighted works, or a database of hashes (think of a
hash as a digital fingerprint) of copyrighted works. The most likely
scenario I can see would be for the Library of Congress (and equivalent
foreign institutions) to maintain a database of all copyright works (yow!)
in electronic format; and then to compare hashes of the work being screened
to those works in the database. Obviously, such an effort - especially one
which could respond in close-to-realtime - would be *incredibly expensive*. 

(The side effect - that we'd end up with an electronic version of every
copyrighted work in the Library of Congress - is alluring. But if we had
that the obvious step would be to make it Web-accessible, which would so
radically transform the publishing business that copyright infringement
wouldn't even be interesting any more.) 

Keeping a database of hashes online instead of a database of works is less
useful because many of the hashes must be recomputed dynamically - e.g.,
such a system should be able to spot substrings or resizings of copyrighted
works.

And all of this ignores the possibility of fair use. So I guess my point is
that the problem of "acting responsible with respect to letting others post
copyright-protected works without permission" is a lot harder than it sounds
like it ought to be. Good faith on the part of the ISP/remailer isn't even
close to good enough. 
--
Greg Broiles                |"Post-rotational nystagmus was the subject of
gbroiles@netbox.com         |an in-court demonstration by the People
http://www.io.com/~gbroiles |wherein Sgt Page was spun around by Sgt
                            |Studdard." People v. Quinn 580 NYS2d 818,825.






Thread