Three counterpoints.

Liquid has one minute blocks, so it is, at minimum, a 10x scaling solution. In private industry we have a saying: build your solution to scale 10x, when you need to scale beyond this refactor it for another 10x. Scaling 100x before you need to is premature optimization and your efforts are better spent solving other problems.

But I think Liquid is far more than a 10x scaling solution because it has regular block times. Bitcoin relies on a random lottery to produce new blocks. This leads to frequent long stretches between blocks. When we have to wait 45m or an hour for the next block and fees are already high, they go absolutely bananas. Liquid sidesteps this whole problem by removing the time variance of block production. So Liquid is a categorical improvement in worst case fee spikes. This lowers fees overall and makes it a fantastic refuge from fee storms on the main chain.

Finally, Liquid is a terrible place to do inscriptions. Inscriboors are attracted to the absolute scarcity and permissionlessness of the bitcoin blockchain. Jpeg degens want their shitty pixelated frogs or whatever to live forever on the timechain. Liquid is a captured network that has no real guarantee of longevity. Once it has outlived it's usefulness Blockstream will shut it down. This makes it a bad place to write your jpegs. Token casino degens thrive on the permissionlessness. They know they are doing dumb shit with their money and they don't care. No one can stop them on bitcoin. Liquid, OTOH, is run by the multisig federation. If Liquid comes under BRC-20 spam attack they can and should simply censor these transactions. This makes it a bad place to mint your worthless pump and dump tokens. So all of that fee pressure is removed from Liquid.

In summary: Liquid is a fantastic scaling solution for economic bitcoin transactions. It is not a permanent solution, but in the long run we're all dead anyway.