Software security after the cyber war
As of 2017, we did not solve most of the security problems that existed in 2002.
Most classes of vulnerabilities known back then are still current. Most of the same classes of exploits still work in the same way.
The cost to break into any operating system is around $100K. Often, much less. Mobile including. Same applies to every browser and every major piece of software. Nothing's really changed over the past 15 years.
The move to the cloud did not improve any security. It simply replaced large number of moderate holes with fewer disastrous ones. On average, the risk remained the same if not worse. And cloud providers keep pushing the risk to the end customers, by demanding more and more of personal information -- like phone numbers or real names -- in order to 'protect' the informaiton already stored in a very centralized way.
Most of large cloud systems are likely penetrated by the APT who sits there in the shadow. The only reason they don't cause a disruption is that it is not profitable to them, yet.
But we are heading into a cyber war and into a disastrous cyber security catastrophe, probably soon -- around 2020, give or take couple of years. In any case, the cost of staying connected to the Internet by then would probably exceed the benefits from doing so.
Current security industry (as represented by Black Hat for example) is mostly a theater and is disconnected from reality. Most of these people will lose their jobs after the catastrope.
If you are seeking to re-build a new security after the war, hire me. That's why I put this page but made it discoverable only to those who know what they are seeking.