Firefox runs amok - thanks to Google

A hidden feature of Firefox can result in the downloading of cookies and Web pages that the user did not intend by doing a Google search.

Among the plethora of stories surrounding the release of Firefox 2.0, I saw one that dealt with the prefetching and how to turn it off. This sparked my interest and a little bit of research shows that this is a long standing issue that not many people seem to know about.

A hidden feature of Firefox can result in the downloading of cookies and Web pages that the user did not intend by doing a Google search. When some searches are complete Google will instruct the browser to fetch the top result, the browser dutifully complies and the end result shows the user has downloaded materials that they have not selected or viewed.

The prefetched pages and cookies will then reside in a user's cache but will not be noted in the user's history. This means that the user will have no idea of the material that is prefetched without looking into their cache themselves - a task that is beyond most users.

In the standard work environment, it is widely assumed that all pages visited are logged for later use by system administrators. This may lead to a situation where a user is facing censure for visiting sites that violate a company's computer use policy.

While generally a handy feature to speed up browsing performance, the drawbacks are seen when an innocent search query will return illicit or pornographic results. An example of such a term is "bear videos", the top result being an explicit personal site complete with videos.

For people that still pay for bandwidth on a per megabyte basis, this feature will also cost money. Web masters will now have to deal with increased traffic if they are Google's top result without any guarantee that a user has actually seen their site. Analytical tools need to be able to handle this or there will be thousands of false positives littering the stats.

Headers exist for Web masters to handle these prefetch requests but Mozilla concedes that they are non-standard.

Users can disable this feature by following the instuctions on Google's help page. How a standard desktop user can be expected to navigate the about:config listing successfully is a interesting question.

That this behaviour is active by default and there is no GUI to control it means that the vast majority of users do not even know it exists. Asking people to regularly clean their cache because of the way a hidden feature works is unacceptable.

By Chris Duckett

Some would say that it is a long way from software engineering to journalism, others would correctly argue that it is a mere 10 metres according to the floor plan.During his first five years with CBS Interactive, Chris started his journalistic advent...