Instead of just crypto learning, another option is to implement proprietary algorithms / logic inside a third-party WinRT component. This way you can be sure that your own algorithm is protected by compilation. If you decide to implement in .NET, there is an opportunity for someone to reverse engineer it.
The idea is to write your client in JS / HTML5, apparently, it will be somewhat simple if you do not have huge patented information. Then you write your WinRT component in C # / C ++, which contains your own "Sausage Production Process". You call this WinRT component to create some βsausageβ with some data entry. This approach means that your secret recipe for Sausage is safe, but at the same time provides you with the simplicity of the platform.
Is this an acceptable solution?
source share