People proposed things like that, but it simply didn't happen. Maybe even the initial testnet4 difficulty should not be set to "1", causing a massive premine for the early creators, who tested it with ASICs. Guess what: that's how it should work. And how it was supposed to work, from the very beginning. Because now, after 20 minutes, any CPU miner can submit a valid block, and if it will be propagated fast enough, then it will be accepted. Only propagation speed matters in that case. And later, when some ASIC will join the network, then it should just automatically smash all of these CPU blocks. It is a bug, that ASICs set these blocks with CPU difficulty in stone. They shouldn't. To do that, only one change is needed: to "allow" the difficulty 0x1d00ffff, instead of "require" it. And of course, the code for ASIC miners should ignore all blocks with CPU difficulty from the tip, and mine a new block with ASIC difficulty, which will reorg everything, what CPU miners did. Which also means, that people are not interested that much in making custom block templates, or confirming non-standard transactions, but rather in just getting coins. In this case, rules could be set to the same thing, as they are in mainnet, but just some scripts with Proof of Work may be used, to distribute coins to people with CPUs. Then, it is much less disruptive, if people are making double-spends, and sending different versions of the same transaction, than if they produce different blocks, and disrupt everyone. Also, it is already possible to use Proof of Work scripts on mainnet, so if someone is interested in getting single satoshis, by providing some Proof of Work, then it can already be done there, in a trustless way.
People proposed things like that, but it simply didn't happen. Maybe even the initial testnet4 difficulty should not be set to "1", causing a massive premine for the early creators, who tested it with ASICs. Guess what: that's how it should work. And how it was supposed to work, from the very beginning. Because now, after 20 minutes, any CPU miner can submit a valid block, and if it will be propagated fast enough, then it will be accepted. Only propagation speed matters in that case. And later, when some ASIC will join the network, then it should just automatically smash all of these CPU blocks. It is a bug, that ASICs set these blocks with CPU difficulty in stone. They shouldn't. To do that, only one change is needed: to "allow" the difficulty 0x1d00ffff, instead of "require" it. And of course, the code for ASIC miners should ignore all blocks with CPU difficulty from the tip, and mine a new block with ASIC difficulty, which will reorg everything, what CPU miners did. Which also means, that people are not interested that much in making custom block templates, or confirming non-standard transactions, but rather in just getting coins. In this case, rules could be set to the same thing, as they are in mainnet, but just some scripts with Proof of Work may be used, to distribute coins to people with CPUs. Then, it is much less disruptive, if people are making double-spends, and sending different versions of the same transaction, than if they produce different blocks, and disrupt everyone. Also, it is already possible to use Proof of Work scripts on mainnet, so if someone is interested in getting single satoshis, by providing some Proof of Work, then it can already be done there, in a trustless way.

People proposed things like that, but it simply didn't happen. Maybe even the initial testnet4 difficulty should not be set to "1", causing a massive premine for the early creators, who tested it with ASICs. Guess what: that's how it should work. And how it was supposed to work, from the very beginning. Because now, after 20 minutes, any CPU miner can submit a valid block, and if it will be propagated fast enough, then it will be accepted. Only propagation speed matters in that case. And later, when some ASIC will join the network, then it should just automatically smash all of these CPU blocks. It is a bug, that ASICs set these blocks with CPU difficulty in stone. They shouldn't. To do that, only one change is needed: to "allow" the difficulty 0x1d00ffff, instead of "require" it. And of course, the code for ASIC miners should ignore all blocks with CPU difficulty from the tip, and mine a new block with ASIC difficulty, which will reorg everything, what CPU miners did. Which also means, that people are not interested that much in making custom block templates, or confirming non-standard transactions, but rather in just getting coins. In this case, rules could be set to the same thing, as they are in mainnet, but just some scripts with Proof of Work may be used, to distribute coins to people with CPUs. Then, it is much less disruptive, if people are making double-spends, and sending different versions of the same transaction, than if they produce different blocks, and disrupt everyone. Also, it is already possible to use Proof of Work scripts on mainnet, so if someone is interested in getting single satoshis, by providing some Proof of Work, then it can already be done there, in a trustless way.

People proposed things like that, but it simply didn't happen. Maybe even the initial testnet4 difficulty should not be set to "1", causing a massive premine for the early creators, who tested it with ASICs. Guess what: that's how it should work. And how it was supposed to work, from the very beginning. Because now, after 20 minutes, any CPU miner can submit a valid block, and if it will be propagated fast enough, then it will be accepted. Only propagation speed matters in that case. And later, when some ASIC will join the network, then it should just automatically smash all of these CPU blocks. It is a bug, that ASICs set these blocks with CPU difficulty in stone. They shouldn't. To do that, only one change is needed: to "allow" the difficulty 0x1d00ffff, instead of "require" it. And of course, the code for ASIC miners should ignore all blocks with CPU difficulty from the tip, and mine a new block with ASIC difficulty, which will reorg everything, what CPU miners did. Which also means, that people are not interested that much in making custom block templates, or confirming non-standard transactions, but rather in just getting coins. In this case, rules could be set to the same thing, as they are in mainnet, but just some scripts with Proof of Work may be used, to distribute coins to people with CPUs. Then, it is much less disruptive, if people are making double-spends, and sending different versions of the same transaction, than if they produce different blocks, and disrupt everyone. Also, it is already possible to use Proof of Work scripts on mainnet, so if someone is interested in getting single satoshis, by providing some Proof of Work, then it can already be done there, in a trustless way.