1. Once a person exists, they can form preferences, including a preference not to die. These preferences have real weight. These preferences can also adjust, although not eliminate, the pain of death. If I were to die with a cryonics team standing over me ready to give me a chance at waking up again, I would be more emotionally comfortable than if I were to die on the expectation of ceasing to exist. Someone who does not exist yet does not have such preferences.

2. People do not all die at the same time. Although an impermanent death is, like a permanent one, also a loss to the living (of time together), it's not the same magnitude of loss. Beyond a certain point, it doesn't matter very much to most people to be able to create new people (not that they wouldn't resent being disallowed).

3. It's not clear that anyone's birth will really be directly prevented by cryonics. (I mean, except in the sense that all events have some causal impact that influences, among other things, who jumps whose bones when, and therefore who has which children.) A society that would revive cryonics patients probably isn't one that has a population problem such that the cryonics patients make a difference.

I don't know why I said "more seriously" when it's by far the less defensible argument.

I think Eliezer is saying cryonics is a positive good

I don't think so; when people say "shouldn't you argue that people give the money to SIAI", he says "why does this come out of our saving the world budget, and not your curry budget?"

I think this is a very weak point and the far future will probably be able to make whatever kinds of minds they like, but we could have scanning/WBE long before we know enough about minds to diversify them.