Filtering

There are three approaches to filtering:

Exclusion of particular web sites or URLs by blocking access – blacklisting. It is impossible to block all objectionable material without also blocking useful content. There are three methods of blocking:
- Keyword blocking where a site or URL is disallowed if various given words are found on it. This is not foolproof; blocking the word ‘breast’ can block access to sites or URLs concerned with ‘breast cancer’.
- Packet cancelling blocks entire web sites. Many large sites such as www.geocities.com act as hosts to literally hundreds of web pages. Within such a site there will be a mixture of acceptable and unacceptable materials. For example, WebCom hosts over five thousand sites including www.national-health-care.com and www.naked-teen.com.
- URL blocking blocks specific web pages. This avoids the blanket blocking of good and bad, but it is virtually impossible to achieve given the number of new pages being added to the Web daily.
Inclusion of particular web sites or URLs on a permitted access list – whitelisting. Typically a database contains a list of allowed web sites. The list can be edited by a teacher, and it may be possible to set different permissions for different groups of pupils. Such permitted lists are popular in education, and a number of specialist ISPs provide such a service. The list may consist of permitted URLs or permitted sites. The same problem applies in reverse when listing sites – a whole site is either included or excluded.
Labelling – categorisation of sites or URLs by certain criteria, and then placing a digital label on that site or URL. A number of labelling systems have been developed. PICS (Platform for Internet Content Selection) is not a filtering system, but is a means by which the originator of a web page, or a third party, can attach a digital label to the header code of the page. Labelling is time consuming and relies on the label being trustworthy. Though browsers such as Microsoft’s Internet Explorer are able to use PICS too few sites are currently labelled to be able to use the system effectively.

The Recreational Software Advisory Council (RSCAi), SafeSurf and Content Advisor have developed categories for sites and URLs which require a human editor to do the categorising. Such lists can be used by filtering software either to blacklist or whitelist sites. Content Advisor has twenty nine different categories.

Paul Resnick, chairman of the PICS working group of the World Wide Web Consortium wrote: "Labelling requires human time and energy; many sites of limited interest will probably go unlabeled. Because of safety concerns, some people will block access to materials that are unlabeled or whose labels are untrusted. For such people, the Internet will function more like broadcasting, providing access only to sites with sufficient mass-market appeal to merit the cost of labelling.

"While lamentable, this problem is an inherent one that is not caused by labelling. In any medium, people tend to avoid the unknown when there are risks involved, and it is far easier to get information about material that is of wide interest than about items that appeal to a small audience." (Resnick, 1997)

Whatever method of filtering is used in a school, it is important that it is flexible enough for school staff to allow or disallow sites additional to those in the database maintained by the vendor. It is also desirable for teachers to be able to turn off the filtering when necessary to facilitate their own Internet usage.

Return to Contents