I think that the case for longtermism gets stronger if you consider truly irreversible catastrophic risks, for example human extinction. Lets say that there is a chance of 10% for the extinction of humankind. Suppose you suggest some policy that reduces this risk by 2%, but introduces a new extinction risk with a probability of 1%. Then it would be wise to enact this policy.
This kind of reasoning would be probably wrong if you have a chance of 2% for a very good outcome such as unlimited cheap energy, but an additional extinction risk of 1%.
Moreover, you cannot argue that everything will be OK several thousand years in the future if humankind is eradicated instead of “just” reduced to a much smaller population size.
Your forum and your blog post contain many interesting thoughts and I think that the role of high variations in longtermism is indeed underexplored. Nevertheless, I think that even if everything that you have written is correct, it would still be sensible to limit global warming and care for extinction risks.