[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[FYI] (Fwd) FC: Australian government publishes censorware effectiveness report




------- Forwarded message follows -------
Date sent:      	Mon, 25 Mar 2002 22:53:43 -0500
To:             	politech@politechbot.com
From:           	Declan McCullagh <declan@well.com>
Subject:        	FC: Australian government publishes censorware effectiveness report
Send reply to:  	declan@well.com


---

Date: Tue, 26 Mar 2002 13:08:08 +1100
From: Nathan Cochrane <ncochrane@theage.fairfax.com.au>
Organization: The Age newspaper
To: Dave Farber <dave@farber.net>, Declan McCullagh <declan@well.com>
Subject: Australian Government releases content filtering survey in
time for Philadelphia trial

Hi Dave, Declan

Call it kismet, providence or good timing, but the Australian
Government has released a 90 page report into the effectiveness of
censorware as the CIPA goes to trial in Philadelphia.

http://www.interesting-people.org/archives/interesting-people/200203/m
sg00184.html http://www.politechbot.com/p-03297.html

The report was commissioned by the Australian Broadcasting Authority
(www.aba.gov.au), which is responsible for censoring the Internet in
this country, and conducted by the Commonwealth Scientific and
Industrial Research Organisation (www.csiro.gov.au).

http://www.aba.gov.au/internet/research/filtereffectiveness.pdf

"Effectiveness of Internet Filtering Software Products" gives a basic
background to the problems of censoring Internet content. It then
addresses the different approaches vendors use, such as
inclusion/exclusion, content, source and image filtering. It finishes
by reviewing 14 products and services including Cyber Patrol 5.0, AOL
Parental Control 6.0 and Net Nanny 4.0.

The report doesn't take a moral or ethical stance, but outlines what
can be achieved with the technology available.

Some highlights from the report:

"While it is technically feasible to block access to all undesirable
Internet content, no Internet blocking or filtering scheme will ever
be 100% effective, or resist a determined and informed attacker, but
many of them will be perfectly adequate in normal use."

"A completely safe Internet may well be a very restricted Internet,
especially when new types of content and new distribution technologies
emerge."

"Many filtering products are based on lists of Web sites that are
supplied by their vendor. These lists are expensive to produce, as
they have to be compiled by having people examine and classify
Internet content, and as a result these lists are often closely held
proprietary information. The secret nature of these lists can make it
difficult to know just what content is being blocked and for what
purpose.

"These lists also reflect the values of the organisations and people
who compile them, and may not reflect the values of Australian society
as a whole. Some Internet activists (Peacefire) complain that
commercial filtering products reflect US-based conservative and
religious values, and as such may not reflect the more liberal values
held by Australian society. Cultures differ considerably in their
concepts of acceptable content and filtering products really have to
customise their lists to meet local cultural norms."

"Content filtering is a difficult problem. Even text-based filtering
requires some ability to determine context (and meaning) for words
they discover. Early products were infamous for simplistic filtering,
with the blocking of "breast" cancer content being the most quoted
example. Filtering products have improved since those early days but
the task is still very difficult and moderately high error rates can
be expected. Filtering out non-textual information, such as
photographs or video, is much more difficult and problematic."

"All filtering technologies are fallible, and the more effective they
are, the more they risk intruding on general Internet usage. Products
have to strike a balance between filtering out undesirable content,
and allowing access to (possibly unknown) useful content. The white
list products are the most effective because they are the most
restrictive and constrain users to a very small part of the Internet."

"Much attention is paid to filtering Web pages but undesirable content
can be found in many places on the Internet, including newsgroups and
file servers. Some of the more tightly filtered Internet services,
such as some of those designed for the educational market, resolve
this problem by completely blocking access to all Internet services
other than the Web and e-mail. This approach is certainly safe, but
would be unacceptable for the general Australian community and so
these other sources may have to be filtered as well."

"An emerging problem with filtering Web traffic through the use of
server-side filters is the rapidly increasing use of the Web's
protocol (HTTP) and port (80) for other purposes, such as e-commerce
and Web Services. Filtering all HTTP traffic could result in degraded
performance for major applications, rather than just slowing down
interactive Web browsing."



**********************************************************************
*********** This email and any files transmitted with it may be
legally privileged and confidential.  If you are not the intended
recipient of this email, you must not disclose or use the information
contained in it.  If you have received this email in error, please
notify us by return email and permanently delete the document.
**********************************************************************
***********





----------------------------------------------------------------------
--- POLITECH -- Declan McCullagh's politics and technology mailing
list You may redistribute this message freely if you include this
notice. Declan McCullagh's photographs are at
http://www.mccullagh.org/ To subscribe to Politech:
http://www.politechbot.com/info/subscribe.html This message is
archived at http://www.politechbot.com/
----------------------------------------------------------------------
--- Politech dinner in SF on 4/16:
http://www.politechbot.com/events/cfp2002/
----------------------------------------------------------------------
---

------- End of forwarded message -------


-- 
To unsubscribe, e-mail: debate-unsubscribe@lists.fitug.de
For additional commands, e-mail: debate-help@lists.fitug.de