As someone who's been involved in setting up a corporate incubator, this is a pretty common dilemma.

The way that we viewed it at my previous company was basically it's two different businesses, two different P&L's and two different sets of metrics and KPIs. You could see them as two different business units that can succeed or fail independently.

If one Business Unit is struggling, you can't just fill the holes by taking funds from another BU to fill the gap. You need to solve the inherent profitability problems in the struggling Business Unit in some other way, typically by selling more, increasing the price/mix or reducing the cost.

Don't know if this is how Mozilla reasons, but this was definitely the way we reasoned in the last two large corporations I worked for.