Although this article reads like a paid ad for IBM (journalists, like everyone else, have to eat) I did a summer internship in Almaden in 2000 and the place was

Although this article reads like a paid ad for IBM (journalists, like everyone else, have to eat) I did a summer internship in Almaden in 2000 and the place was awesome, albeit not as quite open-minded as described here perhaps. The important thing many companies don't realize is that having 20% (or "wild duck") projects is only the first tiny step in the right direction. Many tech companies have these and it means nothing by itself. It needs to be coupled with a healthy way of percolating these ideas upwards in a combined peer-manager controlled way. Doing this is hard. When I read comments from their managers I get the feeling that they are just doing it for the looks/fun/PR, i.e. they're not taking it as a serious source of their future revenue. Well, with this approach, it will never become such a source.