Hi Antoine.
I’d like to say thanks for your answer but you are mischaracterizing my points, getting emotional, ascribing intentions and in the end are not even able to engage with the arguments presented.
I am doing no such thing.
I’m not interested in further entertaining barstool demagoguery, if you came here only to repeat recycled slogans without substance you could have kept it on Twitter.
This is precisely the condescending, arrogant attitude that is leading to the backlash you dislike so much. If you would treat our arguments with the respect we feel they rightly deserve, and actually address them rather than dismissing them, then you would be much more popular and liked. So far I have tried to keep things civil, but it seems you simply don’t respond when I do, so I am forced to be louder and more annoying. Again, the ball is in your court here; I’m willing to have an in-depth good-faith conversation, but you don’t seem willing to reciprocate.
Nobody’s saying that tightening standardness rules wouldn’t impose JPEGers some marginal (and temporary) cost in the form of having to re-build their tools on top of private APIs or alternative p2p relays to the Bitcoin Core one.
Great! I’m glad you agree that filters work 
The point has always been that with such demand for these transactions 1) the costs are ludicrously small
Any increase at all in differential costs between legitimate transactions and spam transactions is a win for filters! The costs are not “ludicrously small”; as a matter of fact, Rob Hamilton tested an attempt to get past the dust filter, and ended up spending 72 hours and 9x the normal fee to get it confirmed. Obviously since Mara implemented their hostile Slipstream service, the cost to go around mempool filters has come down significantly, but the average cost seems to be still around 3x what a normal tx would cost. Again, this is a huge win for filters!
The costs will increase even more once Libre Relay’s DoS attacks on bitcoin are countered by enough defensive nodes.
and 2) the nudge to use direct submission to miners is an additional unnecessary mining centralization pressure
Giving miners whatever fees they want for confirming abusive transactions is a much faster way to cause mining centralization than “nudging users toward direct submission”. Currently only a minority of hashrate (Mara and F2Pool) have “direct submission” services, because these pools are bad actors. The rest of the hashrate has not implemented these abusive services, because they clearly care about bitcoin being usable as money (which they should if they want to avoid undermining their investment!)
So what that means is that Mara and F2Pool are increasing their risk of having their blocks orphaned by not following standardness rules. Of course, they are also opening themselves up to retaliatory action, such as boycotts, by angry bitcoiners. All in all, it doesn’t seem worth the risk for a mining pool to attack bitcoin in this way. Anyway they will always end up passing on this additional risk in the form of increased fees to their customers, which means filters will always be effective, as long as a majority of the network runs them.
Bitcoin Core as a project has always put significant work into addressing mining centralization pressures.
Yes, that has always been a stated goal, but it doesn’t seem to be very effective so far. At the moment, there are only a handful of entities creating block templates for the vast majority of hashrate. Maybe instead of kicking the can down the road by caving to scammers’ demands, core should focus on tech to help re-decentralize mining.
So you can build up strawmans and yell at them all you want, i am confident the project will continue to treat this as a primary concern in the interest of all users of the Bitcoin network.
What “strawmen” are you referring to? You’re just projecting. You are completely ignoring all of the points I made, which (as far as I know) are dealing with your actual positions, and not strawmen. Conversely, your points have all attacked strawmen from the anti-spam side, and you have still not dealt with the steelmen. But please point out where I have made errors and I will apologize and retract.
Nobody’s proposing to change standardness rules to accommodate Citrea or any other side system
I guess I misread your stated rationale then, which explicitly mentions Citrea’s new rollup bridge as a motivation for wanting to remove opreturn limits. Yes, of course there may be “other” rollups that try to do a similar thing, but such rollups can also be treated as hostile (because they are).
Their design works fine for them today, with or without your (or my) blessing
It actually doesn’t. If enough people filter their transactions, their entire system falls apart. Currently their txs are considered standard, but there’s no reason why we can’t filter them. The reason they are not relying on hostile private relay services such as Mara or F2Pool is that they need assurances that these transactions will be confirmed in a timely manner and just using one or two small miners is not a good enough guarantee. They need the cooperation of the public relay network to even launch. They need us. We don’t need them.
The point is to rectify the perverse incentive to use forever-unspendable outputs unnecessarily created by Core’s standardness rules.
Yes, and you still haven’t addressed my point about this being a fake motivation, given brc20s having added 8GB of utxoset bloat because of core’s failure to merge Luke’s anti-inscription PR. The longer you dance around trying to avoid answering this point, the more dishonest you look.
I’m aware that you love this strawman so i’m sorry to break it up to you: Citrea’s usage itself has never been the primary concern
I never claimed this was the primary concern. Citrea was the catalyst, though, and if we respond to such attacks with the hostility they deserve, it’s very unlikely that other similar ventures will follow. Again, they need our cooperation in order for their system to function. We don’t need them - any of them. Bitcoin works just fine without EVM-casino scamcoin rollups. (Actually it works a lot better!)
It was only insofar as it is a manifestation of the perverse incentives which need to be rectified and as it’s always some potentially created forever-unspendable outputs that we are better off not taking
It’s amazing how you keep dancing around the brc20 issue. You don’t see the contradiction in calling Citrea’s usage of fake pubkeys a “perverse incentive”, but somehow letting anyone stuff as much data as they want into the witness, causing a tripling of the utxoset size in 2 years, is not a perverse incentive?? Please let me know how you are squaring that circle.
Another argument that was later brought up is that since large miners already mine these transactions, Bitcoin Core should relay and include them in block templates by default.
They don’t, actually. Only a minority of hashrate mines nonstandard txs. And again, the mining pools that do (F2Pool and Mara) are hostile actors and we should not simply lie down and let them get away with ruining bitcoin’s entire reason for existence (being money) by stuffing it with permanent, toxic junk.
And if a majority of hashrate were mining abusive txs against the will of the noderunners, then we have much bigger problems on our hands and we should find out right away. After all, Nakamoto Consensus falls apart if miners are not at least 50% honest.