And it's funny how these days many people bring up sidechains and lightning as potential solutions to bitcoin's scalability issue
Their reasoning goes as following: whether we rise the block size or not, there will still be scalability problem. Even if the block size limit was 1 GB, the spam attack would cost nearly zero, and there would probably still exist a fee market (much less intensive than now of course). People would really take advantage of the 144 gigabytes of space every day. You can verify that on BSV blockchain, where they have saved whole movies. Needless to say that with 1 GB of block space, one mining pool can take advantage of the time it takes to verify the block and kill off the competition.
This is an indication that we need more than just one layer.
I think that the issue of anti-spam measures kind of solves itself when we have a deflationary and scarce coin like bitcoin.
The minimum on-chain unit is 1 satoshi, which with current prices is at 0.0003604 USD. Ste standard segwit transaction is 140 vBytes according to mempool.space. So even at fees of 1 satoshi per vByte one would have to pay at least 5 dollar cents per btc transaction. Of course, a block size increase doesn't have to completely decimate the fee market, which is a useful feature. A block-size increase could be dynamic and adjustable. There were such solutions presented even circa 2015 with the bitcoin XT and bitcoin classic clients. Under these measures, a spam attack could in par to how expensive it is now.