>You could look at the problem in two ways.
>
>You could represent it in "gods" time - the universe is scaled to fit in a little orb that you can hold in your hand - and distance, and therefore time, are inconsequential.
>
>In this model a super nova would be witnessed as it happened rather than when it was observerd (several thousand years depending on distance).
>
>Or, you could model it relative to the observer's time - ie - the further away someting is, the more the lag between the observers moment and the event's moment.


Well, the requirements are to make a model that accounts for all observers' measurements of time.

So modeling relative time would give you at least one observer's measurements, and any other observer's measurements that can be deduced from that. But it won't include all the observer's measurements in the universe, because they could be outside the light cone (out side the observable universe) of the first observer.

That seems like a pretty tough problem to work around.

But if you chose to model absolute time, and found some way to get relative time out of the model, that might be an easier solution the problem.