One is that future contingents like 'It will rain tomorrow' lack a truth value. This definition of the open future denies bivalence, that every sentence is either true or false. But there are some views of the open future that are compatible with bivalence, so Torre thinks this isn't the best definition formulation of the open future.
Another is that the "future is indeterministic" in the sense that all the untensed truths now plus the laws of nature do not entail all the future-tensed truths. (Usually one says that the laws or the universe is deterministic.) Torre's main objection with the formulation of the OF is that it doesn't account for the asymmetry between past and present, because the past could be indeterministic in this sense, yet we do not think it is open.
The last that Torre offers is that we can affect the future but not the past, and in that sense the future is open. This is in line with Lewis's formulation in terms of counterfactual dependence.
On the block theory, all events are temporally related by the earlier-/later-than relation (so time is linear), and on the branch theory, this is not the case. Suppose a moment t branches to moments t1 and t2. Then t1 is neither earlier nor later than t2.
Yes. Consider a growing block view on which the future doesn't exist but where the laws of nature are deterministic. In such a world the future is determined even though the future doesn't (yet) exist.
Yes. The so-called Thin Red Line view has it that not all branches are on an ontological par since only one of them is marked out (metaphorically!) by a thin red line representing the actual history and future.
He does so in terms of the asymmetry of counterfactual dependence. E.g. if I had eaten a lot of pancakes for breakfast this morning, I would skip lunch this afternoon, but if I had not eaten a lot of pancakes this morning, I would not skip lunch this afternoon. On the other hand, whether or not I had eaten a lot of pancakes for breakfast this morning, it would still be true that men landed on the moon. So counterfactual suppositions about the present "affect" which future-tensed propositions are true, but which past-tensed ones are true.
Intuitively it needs to be intrinsic, otherwise the sense in which the future is open on Lewis's divergence picture looks merely epistemic; it is one on which we simply don't know which world we're in. In another sense, it's not. For even if we knew all the untensed truths of the present together with the laws, we could not determine which future-tensed truths hold. So one difference between divergence and intrinsic branching is that future contingents are true on divergence, though not determinately, while they may be neither if branching is intrinsic.
On the divergence picture, future contingents are true or false, though not necessarily determinately true or false. Given that the future exists in the block, looking from the outside we can see whether it rains tomorrow, even if it is indeterminate in virtue of the existence of other worlds sharing our laws that are like ours up to the present but diverge thereafter (in the sense of making different propositions true).
That between the factual future tense ('There will be a sea battle tomorrow'), the strong modal future tense ('There must be a sea battle tomorrow') and the weak modal future tense ('There might be a sea battle tomorrow').
If time is linear then the natural semantics for 'It will be such that A' is that it is true at a moment just in case there is a later moment at which A. But this is the same semantics that would be given for 'It must/might be suche that A' if time is linear. So all three turn out semantically, hence logically, equivalent.
If time branches, we have three options for interpreting 'It will be such that A'. The first is that A be true at some later moment. But then 'It will be such that A' and 'It will be such that not-A' may be mutually consistent, which seems wrong. The second is that A be true at some later moment on every future continuation; i.e. it gets interpreted as 'A is inevitable'. But that gives it strong modal force, equivalent to 'It must be such that A (in the future)'. The final option is that A be true at every later moment. This is the strongest interpretation and implies the previous one, so it too has strong modal force. In each plausible case (the last two), we get a collapse of the factual with the modal future tense.
He thinks that we should "investigate the beliefs, intentions, etc. of the speaker to decide whether he is prepared to go all the way to ['It must be the case that A' (in one of the two senses)] or is making a guess, an founded prediction, is only reasonably sure, and so forth, in which case we can take him as asserting ['It might be the case that A']" [p. 288].