Suppose for the sake of discussion that (1) is true. I have no idea whether there are worlds in which there are just 100 happy people, but it does simplify the discussion.
1. w includes 100 happy people existing for 10 minutes only.
The value of w, I think, is ten times the value of w’ in (2).
2. w’ includes 100 happy people existing for 1 minute only.
Now let w” be exactly like w, but add the fact that w” is an eternalist world. w” includes 100 happy people. There is no time in w” at which it is false that 100 happy people exist.
3. w” includes 100 happy people and it is true at each time t that 100 happy people exist.
Since it is true at each time in w” that a 100 happy people exist (and despite the fact that it is not true at each time that 100 happy people exist at that time), the value of w” should be much higher than the value of w. The value of a world is a function (in part) of the number of happy people existing in the world over time. It doesn’t much matter where in the world they are (spatially or, it seems to me, temporally).
But that is a very odd consequence. w” includes no more happy people than w; the happy people in w exist for 10 minutes and cease existing; the happy people in w” exist at t and do not exist at any time 10 minutes after t. The only difference between w and w’ is that it is true at every moment in w” that the happy people exist and it is not true at every moment in w that the happy people exist.
Consider the objection that only presently existing happy people contribute to the overall value of a world.
Isn’t this analogous to saying that only happy people existing at a certain region of a world matter to the value of a world? Once people move to a different region, their happiness no longer contributes to the overall value of the world? That’s false, I think we’ll agree.
Consider the objection that happy people do not contribute to the value of a world for all of the time they exist in that world. Happy people contribute to the value of a world they exist in only for the period of time they exist at in that world.
This objection too has a spatial analogy that is false. This is analogous to the view that happy people existing at region R do not contribute to the value of the world at a distinct region R’. That’s false too.
If Lewis worlds are eternalist worlds then there are lots of happy people who exist, but do not make the actual world better. These are happy people who exist, but do not exist in the actual world. But all happy people that exist in the past (present and future) of an eternalist world w exist in w. And it is true at each time t in w that all of those happy people exist in w and are happy in w. It is difficult to see how, at each time t, the existence of those happy people in w does not contribute to the value of the world. But that fact does make many eternalist worlds too valuable.