> The problem with this is you keep factoring the costs of a computer, or electricity or an ASIC. It's expensive or the user. But for uTorrent, that computer is

> The problem with this is you keep factoring the costs of a computer, or electricity or an ASIC. It's expensive or the user. But for uTorrent, that computer is free. How is that a problem in my analysis? It's important. For example say you bought a Macbook Air and came with mining software, used 100% of your CPU and GPU and made your computer unusable. And it generated 1 penny per year in return. And it'd cost you $100 in electricity. Would you consider it normal to say 'But for Apple it's 1 penny, for free! Your analysis is flawed because all those costs, who cares, Apple isn't paying for them, for them it's a free penny'. Of course not. It's ridiculous. Screwing over your user is not without consequence, there is a cost involved. A moral one, for one, which if you can't express financially still matters. That is, locking up 25% or 50% of a CPU, reducing the lifetime of their CPU and costing them $20 in electricity per year is screwing over your users. But secondly there is also an economic cost, just not directly. But indirectly in the long run, doing things like this will lead people to uninstall and avoid your product. In the exaggerated example about Apple above, it's obvious nobody would ever buy that Macbook Air again. A similar thing will happen for uTorrent on a smaller scale, and that's a cost. And if your little scheme generates peanuts in revenue, then that cost of losing users is likely bigger than your revenue if you'd built a less intrusive scheme. As for your example, it's from spring 2013. Spring 2013 the hashrate was about 50k, today it's 350 million. So literally, the same infected computers today would earn $1.17 in the same period, or about $28 per year. And the electricity costs would be thousands of dollars. Even if you didn't have to pay for it, a business model where you screw with your user's hardware performance, generate thousands of dollars in costs for them to make $28 is ridiculous. Even if you scale that up from 14 thousand to 14 million users, not with heavy GPUs but average users using their CPU, it's absolute peanuts. I showed that already. I'm not making this up. This is exactly my point, as I mentioned, a few years ago this would've been a brilliant idea, hell even spring 2013, today, not so much.