pull down to refresh

THAT'S why developers/CORE/others are so skeptical of filters.
That's fine and all, but the bad look comes from core removing the option for a node runner to even set that for themselves. They just don't care about what many node runners want. And if those are switching to get that option back, they can't pull a surprised pikachu face. (Though I doubt they are.)
It also feels like a storm in a tea cup to me. Claims of core being captured seem way overblown and too sensationalist for me to take them seriously without hard proof. Are they acting elitist? Maybe. Let them. If that bothers you, run knots. Just beware of not letting your ego alone dictate your choice.
Meanwhile, little has changed. If a node runner wants to block stuff from their mempool, they can. If a spammer wants to include stuff, they can. Either option got a bit easier. Fee estimation is still hard. Big woop.
Fee estimation is still hard. Big woop.
Accurate fee estimation is incredibly important across mempools. "Fragmenting the mempool" as it's called is incredibly bad and should be avoided.
Ya sure you can always look at mempool.space or whatever to do your fee estimation for a transaction... but that's exactly the point. Mempool.space shows everything it holds back nothing and if your own mempool isn't showing you the transactions you're competing with... this damages the network.
In general but especially when it comes to things like Lightning. Being able to open Lightning channels and know what the fees are in a timely manner is extremely healthy for Bitcoin.

As far as removing the slider in future versions of core... I agree with you. However, from a design perspective it may be better that users just run their node at default settings and then actually go and use it. Using the node is the key thing...
A knots node rarely used does way less good for the network, than a "stock settings" node that's heavily used to broadcast and relay transactions. With accurate fee estimation.
Just playing the devil's advocate
reply