[LINK] error rates for censorware

[David Mitchell]

All statistics are open to manipulation.  A point Peter rightly makes.

Another useful comparison would be the number of sites identified as 
pornographic by the software compared to the number of sites the OFLC 
in Australia would consider to be rated X or RC under the terms of 
the Australian Internet classification legislation.  This would 
indicate how effective the software was at protecting children from 
inappropriate content, and preventing adults from accessing that 
content, and therefore how effective it was at meeting the terms of 
the Broadcasting Services Act 1992 Schedule 5 - Online Services.

Has anyone considered the effectiveness of these products in an 
Australian context?  Why is the ABA happy to accept that US software 
will do a good job of protecting Australian children?

Has the ABA conducted a study examining the effectiveness of 
censorware in the process of determining the adequacy of the schedule 
attached to the IIA codes of practice (which lists approved filtering 
products)?  If so, what were the results?  If not, why not?

Is NetAlert examining the effectiveness of these products or is it 
aware of studies examining the effectiveness of these products?  If 
not, why is it happy to direct parents to them (although to give 
credit where it is due, they acknowledge the inadequacies of such 
products - http://www.netalert.net.au/overview.htm).  Is NetAlert 
going to examine the effectiveness of such filters as part of their 
role of "researching new access management technologies"?  Or will 
their focus be on restricting adult access to R rated content hosted 
overseas?

blah, blah, blah ....

dm

>The error rates from Peacefire's study were:
>
>CyberPatrol 1.7%
>SafeSurf 4.2%
>Bess 0.7%
>AOL 0.1%
>SafeSurfer 1%
>
>These are the proportions of sites incorrectly identified as
>"pornographic" out of the 1000 tested.
>
>-------------------------------------
>Peter Bowditch peterb@ratbags.com
>Mad - Quintessence of the Loon http://www.ratbags.com/loon
>Bad - The Millenium Project    http://www.ratbags.com/rsoles
>Sad - Full Canvas Jacket       http://www.ratbags.com/ranters

-- 

David Mitchell
Research Fellow
Network Insight (formerly MTPG)
e-mail: <david.mitchell@rmit.edu.au>
Home Page: <http://www.ni.rmit.edu.au/>
Postal Address: Locked Bag 2400, St. Leonards, NSW, Australia, 1590.