Google released the first beta of its anticipated Web Accelerator software last week, but it has quickly run into problems.
Following its impressive Desktop Search tool last month (which I use and can wholeheartedly recommend), it appears things have not gone so well for its latest free plug-in. The problem resides in the way Google Web Accelerator works, which is to cache popular and recently visited web pages in its own servers. In shorthand terms, pages are requested for a user before they have actually asked for them.
Of course, this sounds like a good idea: smart software is anticipating your every move, but in reality it has created numerous problems. For a start, there is the danger of the Web Accelerator accidentally picking out unrelated sites with inappropriate content, meaning that content could end up on your machine. Potentially even worse, is that the Google servers can cache your information and then send it to someone else.
Incidents of this have already been reported, with users discovering they are automatically logged in forums and even accounts that are not theirs. The security risks here are so clear I really don’t need to spell them out…
In fact, so uneasy have some companies become about the risks of Web Accelerator that they are offering methods to disable it. Most (including perhaps the most high profile by searchenginejournal.com, which can be found here) operate by blocking the Google servers’ IP addresses, thus forbidding them automatic access to their sites.
So where does Google go from here? Clearly – beta or not – more time is required in development. Given that the company is allowing no more downloads of Web Accerator for the time being, it certainly realises this. A case of good idea, shakily implemented.