A valid point - even if we succeed in reducing the footprint of a single inscription transaction, that might just make it cheap enough to incentivize others to start inscribing.

Perhaps I should not bother trying to optimize a dead-end technology, and simply wait for the fools to burn through their money. There's no way to introduce a new fee structure without a hard fork unfortunately, so it seems like patience and faith in solid economic incentives is the only practical solution.