I suspect one of the major reasons for the implementation Steve has given is that it works with the existing "contacts" mechanic in Aurora - the current system does not have any way to represent "the enemy is approximately here..." and I think implementing anything like that would be a lot more work than Steve wants to do for not all that much in terms of gameplay interest (where "realistic" =/= "good gameplay"). I think the simplistic idea Steve has presented is reasonable especially if a large EM signature is emitted which would be sufficient to detect the ship generally and represent the "they're now very aware that you're coming" effect while preventing targeting.
Agreed with what seems to be the motivation behind the design decision. I don't know what the answer to this scenario is though:
An AI fires missiles at you from maximum range, and you _then_ turn on your sensor jammer, putting your fleet outside of active sensor and MFC range of the fleet that just fired on you. Those hostile missiles now have no active contact to seek onto. The current rules have those missiles self destruct at this point. 100% anti-missile kill rate, if you can keep the range open, until their magazines are dry, then sail in close for the kill.
This leads to always-on being simpler to solve - game design-wise. But if it's always-on, then that's a constant annoying EM emission. So therefore no EM signature.
I don't say I like it, but the missile cheese scenario I describe above needs to be solved to unpick this chain of effects.
A solution could be simply that missiles continue to the location of the last known active contact, and _then_ self destruct. So could steering onto passive contacts (but still requiring an active sensor fix from either the launching MFC, or an onboard sensor, to actually execute an attack).
Both of these would allow time to re-acquire the contact. Personally I already wish we could fire on passive tracks for other, unrelated reasons, but whatever the solution, it's more coding for Steve to do..