What (doesn't) make a good task in the computer room?

Barcelona, as seen on satellite image provided by Google Earth

In a previous post, I provided links to three articles on what makes a good language classroom task.

There ought to be a lot of overlap between that and what makes a good task if you are taking your learners to the computer room, and I think there is.

Here's an example of a task that I think is poor, which comes from the Winter 2006 issue of a magazine I like a lot, iT's for Teachers (which incidentally has a lot of good things online).

The task (the fourth for a lesson plan that began by looking at five aerial photographs of historic sites, including the Great Wall of China):

Get your students to use Google Earth to search for places around the world, including one or more or the places they have seen in the photographs. Can they find an aerial picture of their school or home?

What's wrong with that as a task...?
My doubts are as follows. I provide only questions -- if you want to suggest answers, that's what the "comments" feature is for...

• What is the aim of the task?
• What language are they going to learn or practise in doing the task?
  
• What are they going to do with what they produce?
  
• What's the return on investment?

I'd ask pretty much the same questions of most computer room language tasks, and one of my answers to the third would almost invariably be "Well, I guess they could blog it...!"

Barça's Nou Camp stadium, pictured via Google Earth. Wow...! But what language do they learn from it.

Technical note
Google Earth requires (free, easy) installation: note that you might not be able to do that on your school's PCs.

And -- again, importantly -- does the amount of language they are going to learn from the task really compensate for the time it's going to take to install?