More and more often, the major world powers do not fight many wars themselves, but often peacekeeping or humanitarian battles on behalf of another agency. Considering this role, the occasional civilian death due to malfunction or operator error is no longer acceptable. Military technology, and the control and use of it, should be subject to the same kind of examination that other technology upon which human lives depend is. A good example is the space program -- given the difficulties that can arise if anything goes wrong, anything that goes up is analysed to death and has at least two backup systems.
In reading about the recent incidents in Libya, at least one seems to be caused by malfunction. This is said to be extremely rare -- it's only happened once in the conflict so far. That, I believe, is still far too much. Consider the failure rate of your average web server, upon which no human lives depend, and which is guaranteed to be operational 99.999% of the time. Compare that to the track record of any recent NATO action.
It's well-known that both machines and humans can fail. Software can crash, gears can jam, people can push the wrong button. It's the job of a good engineer to design around that problem -- failsafes, backup systems, bombs that simply don't explode if there's an error during the launch process. In an ordinary war, perhaps you want your bomb to go off whether or not it hits the target perfectly, but in days where the people you are trying to help are dying, a new methodology is needed.
The current failure rate is not acceptable, and is preventable. Let the world know that you will not accept substandard military technology used in peacekeeping missions, that needless civilian deaths should not be tolerated whether they're at home or overseas.