I also see this commitment would go a long way in creating and documenting the means to support OOo an LO in very large installations, helping develop the technologies companies would use to deploy them.
I am not sure the Brazilian gov hiring developers is the best course of action. I'd much rather have it fund Apache Soft, Mozilla, or any other organization that knows how to do it.
In my experience, pretty much ever govt software I've interacted with (imposto de renda, nota fiscal paulista, the portal brasil fluke?) is incredibly bad. I fail to see the hand on clever people in them.
Aren't there already counterexamples of government products costing more to end users than privately produced products, even in the world of software. Certainly there is no general principle in economics that that which is done by state actors is less expensive than that which is done by private actors. Often the situation is exactly the other way around, with government services being much more expensive than privately delivered services.
Why doesn't the US gov just create a telco? Or any large supplier, really?
This sounds good in theory, but in practice it tends to put the state machinery in charge of managing things it's largely incapable of.
I'd much rather the government would show support either by donations (and let Apache Foundation / whomever manage their own thing), or just creating stimulus plans with the money saved (say, for example to make sure computers did not costs 3x as much in Brazil as they do in the US).
The public sector has many restrictions (elections every few years changing management, very much subject to political pressures, larger bureaucracy etc) that make it a bad competitor in a number of business.