A popular remote desktop connection issue at present shows up in my head making me wonder whether or not what I am thinking is correct at all. I am a Windows internals novice so please be prepared with my ignorance. For security reason I suppose later versions of Windows (post-XP) have been equipped with security dll functions included to turn off all activities available on the computer that is to be monitored. Therefore the currently logged on user of that computer will be logged off automatically when the connection is done. The included dlls that act as a firewall for remote connection I think are "stuck" on purpose with the OS kernel. However since Microsoft Windows is notorious for security leaks (likely to be by chance, on purpose or even real attacks from competitors ), there are still software that can intercept this cohesiveness and one of its module interface was implemented to perform as if the mentioned security dlls malfunctioned or were non-existing. The connection is thus not only easier and faster but the automatic log-off function is successfully eliminated. Is this real ? My take on this is that the current user must log off and that computer must be on for the remote assistance to work.