A few weeks back we ran a story on the poor – make that abysmal - performance of many desktop software firewalls. Sure enough, one of the vendors criticised in the story has made its feelings plain.

The story recounted tests carried out by David Matousec’s matousec.com, which found that the majority of the products in this category are next to useless – at least at stopping a type of firewall test called a “leak test”.

One of the vendors, Agnitum (which otherwise scored moderately well) was accused of optimising the performance of its Outpost Firewall 4.0 to do well at such tests, a technically-involved if serious claim of underhand jiggery-pokery. It’s not clear to me that this is the case, but the vendor has replied without entirely clearing up the niggly details of this matter in a blog on the subject.

What we don’t get, for the most part, is complaints from ordinary users as to why their expensively-acquired software firewall is probably as useful as a pile of mashed potatoes. This would be the case for almost all the products on offer, the worst offender being the Windows firewall. According to these tests, that is utterly useless as to be worth turning on only to rid oneself of the Windows alert noting that it has 'foolishly' been turned off.