I don't think there's anything in the wording of the Lindy effect that means it's only a one way inference. It says the life expectancy of a thing is proportion

I don't think there's anything in the wording of the Lindy effect that means it's only a one way inference. It says the life expectancy of a thing is proportional to its age. That makes sense to me. If a product is going to die, it's likely to die fast. If a product sticks around five years, it's more likely it'll be around for ten more. It's why startups go under all the time, but you don't hear of a lot of 50 year old companies going out of business. I think the error is interpreting these things literally. If we did, then if it were 1990 we'd say Windows wouldn't be see the next century. Probabilities are attached.