There have been some cases of censorship in the past, like when the Samurai team accused Ocean itself. Anyway, game theory takes care of this. If this kind of thing starts happening frequently, the censoring pool will end up losing money, and miners will switch to another pool or even choose to mine solo. Next.
You're right that pool centralization doesn't make the deciding difference when it comes to defending against changes to the consensus rules via hard forks. Nodes are the deciding factor there, and specifically nodes that are actually being used by those who run them, not ones that are just sitting on a shelf gathering dust while the owner broadcasts transactions through whatever the default node is in Sparrow wallet, and checks to see where the transaction is in the mempools using mempool.space.
However, pool centralization is a MASSIVE deal when it comes to what transactions actually make it into blocks. Foundry currently decides what transactions make it into 41% of blocks. Antpool and its slave-pools decide what make it into another 50% or so. That means effectively two entities control what transactions are included in roughly 91% of blocks. That's problematic, to say the least.
Miners should move to pools that allow them to create their own block templates, instead of relying on the pool to dictate what is included in the block template. Currently, the only pool that allows this is nostr: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 that I am aware of, but more pools should adopt DATUM and do the same.
It would not be NEARLY such a big deal that Foundry has 41% of the hash rate, if the individual miners on that pool could create their own block templates based on the transactions in their node's mempool.
Discussion
My only real concern would be a decrease in the number of nodes, as that could indeed lead to centralization! As long as the incentives are aligned, I don't see any issue.
First, it's already happening with big pools excluding transactions in order to be OFAC compliant, and yet those pools are only getting bigger.
Not all censorship results in financial loss that would actually make miners leave due to their bottom-line shrinking.
Those accusations against nostr:nprofile1qqsq9k04vahllseell55m74n3047y88pzlr0z5yany32st29fapqmgsppemhxue69uhkummn9ekx7mp0qy2hwumn8ghj7un9d3shjtnyv9kh2uewd9hj7qgawaehxw309ahx7um5wghxy6t5vdhkjmn9wgh8xmmrd9skctc87ckyu were ridiculous because OCEAN had and still has such a small percentage of the hash rate that any transactions they intentionally excluded would have made it into the very next block, and their intention was ALWAYS to let the individual miners make their own block templates, which we can now do, so OCEAN has no say in what makes it into a block.
Meanwhile, all it takes is two entities to collude and they can keep transactions out of 90% of blocks, which has a meaningful impact on how long it could take for those transactions to be included by a smaller pool.
Most mining companies would not solo-mine, because they need the consistent revenue.
The point is, if a solution to the problem exists, and it does. DATUM works. Then more pools should adopt it, or miners should move to pools that it can be used with.
If the solution already exists, then there's no issue at all. This post was just to drive engagement. Next!