[edsheppa]

A second common criticism of ID is that it is untestable.

The author should have said parenthetically (see previous paragraph).

I think Sober has it wrong. The problem with ID as a scientific theory is that it makes no predictions. Take his example that the statement that an intelligent designer created the vertebrate eye entails that vertebrates have eyes, which is an observation. Vertebrates having eyes is not a prediction of that statement, it is baked in.

Alternatively one could view this lack of predictiveness as a lack of explanatory power.

[curiosity]

I think Sober has it wrong. The problem with ID as a scientific theory is that it makes no predictions.

I thought that was his point. What am I missing?