> Good security happens when the incentives are right. Has this ever happened before? I can't name a single major company on earth that takes customer security

> Good security happens when the incentives are right. Has this ever happened before? I can't name a single major company on earth that takes customer security seriously except through farcical characterization of the concept. It's just too much of an impediment to profitable business to be incentivized in any serious sense. Even the concept of "security" without parameterization of a threat model seems to make a joke of the concept. I'm not claiming that free software offers security that's better per threat model, but they at least allow customization of the threat model itself. Without this being up for discussion such a claim to discuss security in general is extremely difficult to take seriously. The Google threat model openly lauded in the article seemingly doesn't take state actors as a serious threat model, for instance. They'll just hand your data over to the state that asks for it because they want the business that happens under that state.