Filtering the Internet in Schools

Soon after the school year started [[Oliver]] reported to me that he’d tried to access this website from one of the computers at [[Prince Street School]] and that he’d been blocked. I later heard from a friend with a spouse who teaches at a local high school that they’d run into a similar issue. 

And so over the past few weeks I’ve been learning a lot about Internet filtering in Prince Edward Island schools.

Internet filtering is a touchy subject with me, as in my heart of hearts I’m a strict information libertarian. Filtering the Internet is an effort that’s both ultimately futile, and takes our collective eyes off the real issue, which is outfitting our children to be their own information filters.

At the same time I know that’s a rather Utopian vision of information society, that it’s an almost impossible political sell to the culturally and spiritually squeamish, and that without some sort of filtering there would likely be no Internet in schools at all.  (I’ve also seen enough of the Internet’s seedier side to know that there are some things you can’t un-see, and I’m not sure what the answer to that is).

Which is all to say that I went into my “learning about Internet filters in schools” exercise with a mixture of open-minded curiousity and light moral indignation.

Here’s what I learned.

Back in the spring of 2007, the Province of PEI issued a tender for “the supply of a web filtering solution” (tender documents as a PDF). The “filtering solution” was to cover not only Prince Edward Island schools, but also computers within the public service and the health system; the successful bidder was a Guelph company called Netsweeper, not surprising as it was one of only two products that the tender pre-qualified (the other was Blue Coat).

Netsweeper’s content filtering methodology works, by their own description, as follows:

At the heart of the Netsweeper application resides an artificial intelligence engine that will categorise URLs in real-time based on actual website content.

When a user makes a request to the Internet, this request is intercepted and the URL is identified. The URL name is then sent to the Netsweeper Policy Server to determine what action is to be taken, meaning to allow or deny access at the site. If a user visits a website that is not in the local database, it is requested from the central database or the Content Naming Server (CNS). The CNS provides the information about the requested website to the Policy Server, which caches the information so as to be ready should the request ever be made again. This ensures that the Policy Server only has relevant URLs in its cache.

If the website requested by the end user is not in the CNS, the URL is sent to the categorisation engine that will download, scan and then categorise the page on demand. This category information will then be sent back through the chain to the Policy Server. This entire process happens in milliseconds, resulting in virtually no degradation in performance for the end user.

The categorisation system is also integrated with the Master CNS. As unknown URLs are found and categorised, the Master database is updated and will synchronise with all Policy Servers on a global level.    Today, this central database consists of over 1 billion unique URL addresses.

On a local level, the Department of Education developed a list of Netsweeper categories to which access would be denied (shown on this document in red), the end result being that when a teacher or student in a PEI school uses their browser to access a URL, the URL they entered is first handled by Netsweeper which determines the category it falls in, determines whether that category is on the “denied” list, and, depending on what it finds, either allows or denies access.

You can get a sense of how the categorization process works using the Test a Site tool on Netsweeper’s website.  This site, for example, is categorized under:

• Journals and Blogs
• Arts & Culture
• General News
• Sales
• Sports
• Travel
• Portals

With the exception of “sales,” “sports” and “portals” that’s a relatively accurate categorization (sports?!). 

None of the categories appear on the list of denied categories for PEI schools, so, in theory, this site should be available in schools.

And, sure enough, this morning I was able to confirm that, in fact, it is available: the technology coordinator at Prince Street School allowed me to test this in the school’s computer lab, and the site was accessible without issue. This means that either something has changed since Oliver originally tried earlier in the year, there was a non-related technical issue when he tried, or simply that he entered the address incorrectly. 

Irony aside, it’s good to know more about Internet filtering is set up inside schools; while I remain convinced that it’s wrong-headed, at least it’s not at least slightly more transparent.