Does anyone have a link to a bitcoin transaction which has actually been filtered?
Discussion
such a tool could be an interesting project to work on
(Hint, there isn’t one, it’s a rhetorical question)
If the filter worked and made it annoying enough for them not to bother going to a miner out of band, then there wouldn’t be a transaction on the blockchain in the first place (which is the point).
Hard to prove how many transactions the filters have kept keep out by making it harder to use though isn’t it?
You can just post a txhex, if it pays a market fee rate, I’m sure it’ll get included in a block. It takes 2 clicks to get it into slipstream the friction is near 0 at this point
Would a failed spammer want to out themselves tho?
They can just submit the tx to slipstream, the failure rate isn’t there. If you pay the fee, filters don’t work
The fees are the filter
Yeah they filter out people that can’t afford to pay them even if they want to use the network for monetary purposes
Whoever values their transaction most gets confirmed. Simple as that. If "monetary purposes" are not valued as highly as something else that someone else values more then you are probably doing it wrong.
Those dirty poor people just need to get richer, am I right?
Very odd to apply a moral lens to an anarchic system, if you wanted an ESG monetary system may I suggest Fiat?
Way to miss the point lol
Good to know you agree that non wealthy people that can’t outbid the fee of scammers are “doing it wrong”
You can’t escape morality. Bitcoin and everything else humans create are for the purpose of increasing human flourishing, in principle at least. You can’t just look at a piece of code and be blind or obtuse about the purpose of that piece of code or anything else we create for that matter
No. You’re missing the point of Bitcoin, and are one unironic half step into making a Roger ver bcash argument of babies are dying because of transaction fees.
I’m missing the point of Bitcoin? 🤣
It couldn’t be you who supports people wanting to put jpegs on the chain and having all nodes relay and store those jpegs in their mempool without an incentive that is missing the point of P2P cash. Can’t make this stuff up
Why do you accept JPGs in the blocks you receive? Your node is complicit in the spam until you fork off.
That’s not how it works or what the debate is about, but nice try lol
The debate is it requires a consensus change to change any of this, which is obviously true.
The debate is I don’t want to be forces to relay or store spam on my mempool.
You will not gaslight me or anyone else into making life easier for you and your spammer friends
The response to that, is that no one is saying you aren't allowed to do whatever you want with your own software, you are making a point no one is disagreeing with.
The next step to this is, does this have any impact on the network? And the answer is no, and if you want to actually change network behavior you need a change in consensus to do that.
No one was trying to remove the users ability to configure their node to filter out txns with an OP_RETURN exceeding a certain size?
No one is trying to remove filters from the default of Core?
No one is campaigning against filters saying they don’t work and trying to expand the amount of data that can be stored in OP_RETURN so spammers can save money and can more easily put their spam on the chain?
Ok bro, go gaslight someone else
No. Its Free Open Source Software, you can run your own fork and do whatever you want.
Here is GMax explaining this exact point on reddit.

Stop thinking like a serf and just do whatever you want in bitcoin.
No one is arguing filters prevent people from paying a high fee and handing the jpegs directly to Marathon or any other miner
Try creating a txn with an OP_RETURN greater than 80 bytes and see how far it goes
Nodes have no incentive to store spam in their mempools or to relay it to other nodes
You can use libre relay and send larger op returns already.
I can jump a wall and break a door. What’s your point?
People leveraging the bitcoin protocol to become sovereign have no incentive to relay garbage around to other nodes and store that in their mempool.
That spammers can get around by tricking nodes into accepting spam or going directly to miners is not a reason for them start passing around spam.
Telling noderunners to just get on with the program and relay spam because it may end up in the chain anyway and gaslight them into thinking filters don’t work because they don’t stop every single spam txn is a hard sell and bad strategy.
“Tricking nodes” - write the consensus change and fork off, otherwise, you consent to using the code this way and are complicit.
Again with the projection and deflection lol
Anti spammers are not calling for consensus changes
I have no obligation or incentive to relay or store spam on my node
The people that need to rage quit and fork off already are those trying to make it so nodes are forces to relay and store spam in their mempools
My comment was a hint as well
Call me crazy, but if it’s been filtered, it would have a link because it’s not included in the chain. Am I wrong?
You could share a txhex - but if it pays a competitive fee, and is consensus compatible, it’ll get included in a block
So technically no, there is no realistic way for me to share a link of a bitcoin transaction that has been filtered.
Because it was filtered.
That isn’t true, you can share a txhex for a transaction not in the mempool or a block.
Per your question, I don’t believe that would be a link specifically. Can you define what you mean by “THE Mempool”
I only know it as “My Mempool”
You can create a bitcoin transaction, have it as a txhex, and see if it’s in any mempool via the p2p layer, so it’s the collective of all mempools I am referring to
The idea of “the collective of all mempools” is not a defined entity and can't be queried or observed in its entirety.
There is no way to know whether a transaction is in any mempool, because you can’t reach or query every node in practice or in principle.
You can walk the network graph and talk to any node in the p2p network graph with ports open and check to see if they have a tx in their mempool.
Something is telling me you’re playing pedantic word games to avoid the point, which is it’s absolutely possible to show someone a tx which has been filtered, because it would be a transaction not in any block, nor none of the reachable networks mempool.
Here - let me even help you out - you can create a 0 fee tx, and that would be a “filtered” transaction. It takes a few minutes to do it on a core node in the console.
I mean; I did say call me crazy from the start and I’m not ignoring the point: the point still stands.
Miners would not include a 0 fee transaction unless there are other incentives. Just like some nodes would opt not to relay a transaction with arbitrary data less there is policy alignment.
Therefore I cannot “link” a transaction that has been filtered because it cannot be verified on chain (which would disprove its filtered nature) and I cannot “link” a transaction that has been opted not to be relayed around the network because it would be unable to query it.
A filtered or non-relayed transaction cannot be linked, verified, or queried because it exists outside the bounds of both blockchain inclusion and mempool propagation.
But maybe I’m just crazy. I’m still learning.
The linking of a TX is a TX Hex, you can load it into your node, sparrow wallet, and other bitcoin software.
Correct miners would not include a 0 fee TX short of exogenous incentives. Your opinion on how the network topology works for nodes is a non sequitor.
Here, this is a txhex I just spun up on my node:
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
TXID: d3002dbfb6317ebc98107a305e3855969c8049da2384e3e4d8b9d3bb8c7ba9be
Use either sparrow wallet to load it yourself, OR your own bitcoin node using decoderawtransaction.
Whoops, forgot to sign it. Here is the full txhex, also used the wrong vout for the utxo:
TXID: eaece0c1768247d3b8718ed0cab59752d149884d859aef9998d5bbf7648258b2
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
Here is another one (can't spend with both, it'd be a double spend):
TXID: 78996a0c15aa1b01574b99b0899829211d4ea3fad7c29cac0572eef7f1a6018d
02000000000101e58689052bfbff3af1cd60c65d82e0d17b0a60c235817ba50e1a71b4111940410400000000fdffffff0100170000000000002251208f200657b2853dfc759fb2aab5d6cda2e0227ce83041aa37d36b3345ace800ca0140fecfa6b2fb5d91c99d52cc122655bd0cdbc4426bf772098a12a7babbbb6691307cb7ab686fa0c999911ba53f629abd36db0770c80ebae5c7e2a58668544de2e900000000
Transaction, linked:

i'm curious.. can you send one tx with 1000 bytes op return and see if either your node filters it or if it gets successfully broadcasted to the network? then share the tx afterwards..?
If it eventually gets found in a block, it got broadcasted to the network. I think you can run logs on your node to see what is filtered.