Filters Work....kinda not

Thought this was an ironic combination...American Libraries Online January 13, 2006 issue reports that L.A. County has decided to spend $300,000 on Internet filters. Meanwhile, Groton, Connecticut reports that their filtered Internet terminals may have been used to download child pornography (police are investigating it now).

I offer these random thoughts:

1) Filters don't do much better than 80-85% at blocking what you want them to block while not blocking anything else. And that's only the good ones. In other words, filters are not exactly precision tools.

2) There is no such thing as a CIPA-compliant filter (which Groton believes they have). That is unless someone has come up with a universal definition of 'obscene' while I was out of town.

3) The investigation in Groton resulted from library staff reporting the guy doing the downloading. In other words, the filter was useless in this case.

4) I sure hope L.A. County doesn't spend $300,000 on software. Filters filter based on categories and/or on-the-fly keyword analysis. To get them to be CIPA-compliant-like, you have to buy a good one and you have to decide what you want it to do and then spend lots of time configuring and tweaking it so that it matches your policies. And then you have to monitor what its actually doing. The price of the software is probably less than half the cost of of filtering -- if it's done right. And that's just the first year.

It's not that I think filters can't be useful, they can help in certain situations. But library trustees, city council members, library staff and the public need to be a bit more realistic about what they can and can't do and respond accordingly.