Nostr notes by m0wer

I think Jam and Raspiblitz integrations will help a lot with ...

2026-07-21T21:35:25Z

In reply to nevent1q…gjej
_________________________

I think Jam and Raspiblitz integrations will help a lot with getting the rest migrated.

There are sooo many cool things that are going to come out thanks ...

2026-07-21T20:51:47Z

In reply to nevent1q…2yvu
_________________________

There are sooo many cool things that are going to come out thanks to Taproot existing (basically privacy related stuff).

Taproot swaps are one of them. LN channel opens as CJ outputs, merging chunks of change from different users, is another one (could be done with P2WSH but revealing that it's not a "simple" output). And, hopefully one day, PTLCs. Which could be a game changer for privacy in LN.

There is also this on my reading list about how could Taproot be used as a way to store commitments for PQC https://eprint.iacr.org/2025/1307.pdf

Taproot LN channels are cool, use them! #naddr1qq…p96n

2026-07-21T10:49:54Z

Taproot LN channels are cool, use them!

quoting
naddr1qq…p96n

Starting with LND 0.21.0, nodes can open production Simple Taproot channels. These are still HTLC-based channels, there are no PTLCs yet, but they already provide meaningful privacy and efficiency improvements.

A Taproot channel funding output looks like an ordinary P2TR output on-chain. When the channel is closed cooperatively, it can be spent through the Taproot key path without revealing the Lightning channel script, making it much harder to distinguish from an ordinary Taproot payment.

Cooperative closes are also more block-space efficient. A force close can still reveal the channel structure, so Taproot does not make every possible channel lifecycle invisible.

You can combine Taproot channels with SCID aliases. Instead of using the real short channel ID. Which directly identifies the funding transaction and output. Instead, the peers use an unrelated alias for routing.

This is particularly useful for private channels:

The channel is not announced in the public gossip graph.

Invoice routing hints can use the SCID alias rather than exposing the funding outpoint.

A cooperative close does not reveal the channel script.

Together, this makes private channels substantially harder to associate with an on-chain UTXO.

Enable support in lnd.conf:
protocol.simple-taproot-chans=true
protocol.option-scid-alias=true
Restart LND, connect to a compatible peer, and open the channel explicitly:
lncli openchannel \
  --node_key=<PUBKEY> \
  --local_amt=2000000 \
  --private \
  --channel_type=taproot \
  --scid_alias
Taproot channels must currently be private in LND. Both sides must advertise support, and the remote node can still reject the channel through its own acceptance policy.

To find candidates in your local graph, look for:

44/45: explicit channel types

46/47: SCID aliases

80/81: production Taproot channels
lncli describegraph | jq -r '
.nodes[]
| select(
    (.features["44"]? or .features["45"]?) and
    (.features["46"]? or .features["47"]?) and
    (.features["80"]? or .features["81"]?)
  )
| [.alias, .pub_key, ([.addresses[]?.addr] | join(","))]
| @tsv
'
After connecting, verify the peer again with lncli listpeers. The public graph can contain stale feature information, while listpeers shows the features advertised during the current connection.

Taproot channels are not the final form of Lightning privacy, but they are an important step toward making Lightning activity look more like ordinary Bitcoin activity.

https://stacker.news/items/1529242

Voilà!

2026-07-15T21:12:34Z

In reply to nevent1q…2emk
_________________________

Voilà!

Nice catch 😜 Takers might need to "fake" a swap ...

2026-07-15T20:09:20Z

In reply to nevent1q…52hj
_________________________

Nice catch 😜

Takers might need to "fake" a swap whenever they want to spend. But if they do so, the onchain fingerprint is "perfect". A maker could give an equal amount P2TR otuput to a swap claim address (onchain could look like a 1 input 2 outputs tx). The "maker" has change from the swap, but the other "branch" of that swap are now someone else's. That might do whatever with it. How could you tell if it was a real swap or not?

But there are other ways. Takers can also get channel opens as change outputs. Which could then be spent in part (splice, swap out, rebalance, ...) or fully (ideally cooperatively as well). And of course, makers can also eventually spend!

Probably by the time we get there we can come up with simpler ideas though.

The challenge definitely is complexity.

Decentralized lnproxy (provider discovery through Nostr) working ...

2026-07-15T20:02:06Z

Decentralized lnproxy (provider discovery through Nostr) working on signet :-)

PR: https://github.com/lnproxy/lnproxy-relay/pull/31

Try it at: https://m0wer.github.io/lnproxy-webui2/?network=signet

BTW this is a nice web LN wallet for testing on signet: https://ln.signet.coinbin.org/

Let me present you what could be the future of JoinMarket and ...

2026-07-15T16:38:29Z

Let me present you what could be the future of JoinMarket and hopefully a great tool for Bitcoin privacy and fungibility.

First of all, there is the motivation. JM makers have an onchain fingerprint, that although it's hard to follow, limits the privacy achieved by the takers. The full details are at https://gist.github.com/m0wer/a228c625fcb6a27c32e298ec903dfc44

But to save you some time, here the main problems described there are: unique enough maker fees, co-spending of inputs, makers never spending, ... Which in the worst case scenario, would mean that a pure onchain observer could potentially cluster some makers across mixdepths and reduce the anonymity set by one of CoinJoins where they participate.

So what can we do? Well, we can use this chance to bring JM to the next level. And create such a transaction graph "mess" that becomes a chain analyst nightmare :-)

Let me explain. We can't ever get fully rid of the links between UTXOs onchain, or at least not without a way more complex multi taker model that leads to Knapsack "dense" transactions. But we can strive for something, that is that no single party ever has all the information.

The idea is the following. JM makers could become Lightning Network swap providers. Like Boltz, like SwapMarket, like electrum swapserver. Using Taproot swaps and advertising through Nostr. LN swap users get a simple and trustless client side web UI, makers get "fresh" UTXOs from onchain->LN swaps and the opportunity to "spend" CJ outputs for LN->onchain swaps. Oh, and another source of fees. LN swap users get cheaper swaps from the increased competition.

But this is not enough. We need to break the subset sum analysis onchain. We want to stop maker clustering from the root. Here is the next idea, JM CJ participants (makers and taker) could get their change as an LN channel open. Or two, with random other peers and the change amount split randomly between both channels. So a maker that has a 10 M sats input in a 5 M sats CJ no longer gets just a clear ~5M sats change, instead, two LN channel opens where the balance is hidden. Maybe 2M sats in one and 3M sats in the other. But an external observer wouldn't know. Not even which channel is whose!

You could say, well, but an active attacker could probe those channels balance. Maybe, but not really. Those channels would be private simple taproot (HTLC based, not PTLC yet unfortunately), so not advertised to the network. And, BTW, onchain they look like any other P2TR output BTW. Advertised only when needed with a SCID alias that does not reveal the onchain UTXO. But hey, who knows, maybe some peer node that gets the SCID could try to probe it. Fear not! There's another ingredient coming to the mix.

lnproxy is a great service to hide the destination node of an LN BOLT11 invoice by having a relay/proxy node wrap it. The payment hash is the same, only the recipient can settle the payment. But whoever sees the wrapped invoice sees the relay/proxy pubkey instead of the destination one.

Why not decentralize lnproxy through Nostr and then makers can be lnproxy providers as well? Just to add a little bit more noise to their LN activity and help move the balances around. So that if these change as channels ever close (cooperatively) the onchain fingerprint those not help understand what happened in the opening CJ. Or that UTXO can still be used as CJ inputs if signed cooperatively. Or, of course, as LN channels for swaps of lnproxy. Some kind of noisy spin of CoinjoinXT.

What we need for this and what we're currently working on at jm-ng:

- A JM taproot only pit.
- Decentralized lnproxy.
- Decentralized LN swap providers.
- Protocol wiring for the change as channel CJ coordination.
- People to actually use this and not be overwhelmed by the complexity.

But if we manage to make this work, we will have awesome onchain and LN privacy. And a way for service providers to earn money anonymously by providing privacy to others. So it's definitely worth a try :-)

Nostr event nevent1qqs09tjw7weulzk6pv5ywsc4yymq2jpf9ze0ypjhctrq34z3vpgzcrszyp69n5en4anqvmcxdnu809hxjrdn49hlg560nm05a2m5muhjqu5fk5406c7

2026-07-14T14:49:07Z

https://github.com/joinmarket-ng/joinmarket-ng/releases/tag/0.34.0

Automatic history import for existing wallets, fee quantization defaults for better privacy, coin control labels, local mempool API support, bug fixes, and security hardening.

wtf

2026-07-12T14:51:00Z

In reply to nevent1q…53ax
_________________________

wtf

So, what if JM makers "gambled" their change? Maybe ...

2026-07-12T14:38:40Z

In reply to nevent1q…6x0s
_________________________

So, what if JM makers "gambled" their change? Maybe chunks of it?

Just a random idea: Could the probabilities be adjusted so that if maker A that "has" 1M sats change and maker B that "has" 3M sats, instead get proportional chances (¼ and ¾ respectively) to win 4M sats? Maybe they're willing to play this game for a expected value of 0 but a positive expected value of privacy gain.

I guess it doesn't work with more than 2 participants, but if it did, I'm sure many JM takers would be willing to gamble their change, for either a 10x that makes it worth it to deal with it (toxic change), or just lose it.

Wasabi coordinator tagging attack explanation and PoC: ...

2026-07-09T19:57:34Z

Wasabi coordinator tagging attack explanation and PoC: https://gist.github.com/m0wer/2e2f0a3241d8e7651ffdda4435e729ec

You are fully trusting the coordinator with your privacy when ...

2026-07-09T13:01:40Z

You are fully trusting the coordinator with your privacy when using Whirpool (Samourai). Here are some practical attacks with a PoC.

https://gist.github.com/m0wer/e701bf5934474935c459b5e84e5d4e1b

In many cases the default variance is not enough to overlap with ...

2026-07-07T17:46:48Z

In reply to nevent1q…5mzu
_________________________

In many cases the default variance is not enough to overlap with any other maker (and even when it is, the anonymity set might not be great).

When several maker do overlap in fees, randomization becomes "harmful" IMO. Because it then helps track an individual maker by following their "randomly unique" fee through the orderbook.

This second part is arguable. But for a simple message to reach makers and give them an "easy" fix, I think it is better to not get into much detail (but the links expand on this for anyone interested).

JoinMarket makers with unique enough fees are easily clustereable ...

2026-07-07T11:57:46Z

JoinMarket makers with unique enough fees are easily clustereable across mixdepths and don't provide much privacy to takers that choose them.

Imagine there is only one maker that charges 0.123% fee. An external observer can identify that maker's equal amount outputs when they are provided as inputs for another coinjoin with the same fee (observable onchain).

So I would encourage makers to select one of these quantized fees and disable randomization. This is an opt in change that joinmarket-clientserver and joinmarket-ng makers can apply on their own.

The more makers on the same quantized fee, the better anonymity for them and consequently for takers that pick them.

More info at https://joinmarket-ng.github.io/coinjoin-simulator/mainnet-deanon.html and https://github.com/joinmarket-ng/joinmarket-ng/issues/508

In JM, makers don't know anything else about the taker other ...

2026-07-04T09:24:43Z

In reply to nevent1q…fdap
_________________________

In JM, makers don't know anything else about the taker other than what an external onchain observer sees, and their own input-output mapping (of course). The taker is the coordinator and signs last. Each maker sees the whole unsigned transaction and signs their inputs. JM has other "flaws" though.

In Wasabi and Samourai you are trusting the coordinator not to attack you. But it could. See: tagging attacks.

I think we would all be better off if Bitcoin privacy wasn't about "cults".

If AI is making you "think less", you are using it wrong.

2026-07-01T10:49:10Z

If AI is making you "think less", you are using it wrong.

Jam V2 beta (JoinMarket WebUI) ...

2026-06-30T12:55:44Z

Jam V2 beta (JoinMarket WebUI)
https://github.com/joinmarket-webui/jam/releases/tag/v2.0.0-beta.0

TLDR: Big frontend rewrite with many improvements. Using [https://github.com/joinmarket-ng/joinmarket-ng/](https://github.com/joinmarket-ng/joinmarket-ng/) as the backend.

A standalone docker image for fearless beta testers is coming next.

Any and all feedback is highly appreciated.

Please know, that there are still open issues and refactorings needed before the final release.

https://stacker.news/items/1517235

where was it?

2026-06-30T12:16:29Z

In reply to nevent1q…a8cf
_________________________

where was it?

Thanks for the answer. I meant this with the profitability ...

2026-06-29T19:15:17Z

In reply to nevent1q…hqp9
_________________________

Thanks for the answer. I meant this with the profitability question:

quoting
nevent1q…jeve
Oh and about profitability, the problem is not the small difference in fees depending on the block transactions. I was talking about having most of the pool's hashrate mining only over a BIP-110 chain of blocks (fork) that then get reorged so all those block rewards are gone and the hashrate "wasted". The impact of profitability would be huge in this case. And it's more about what people within the pool decide to do, not even conditional to overall support of BIP-110.

Oh and about profitability, the problem is not the small ...

2026-06-29T12:40:14Z

In reply to nevent1q…wmrk
_________________________

Oh and about profitability, the problem is not the small difference in fees depending on the block transactions. I was talking about having most of the pool's hashrate mining only over a BIP-110 chain of blocks (fork) that then get reorged so all those block rewards are gone and the hashrate "wasted". The impact of profitability would be huge in this case. And it's more about what people within the pool decide to do, not even conditional to overall support of BIP-110.

DATUM allows for pooled mining while each miner decides on their ...

2026-06-29T12:36:10Z

In reply to nevent1q…au04
_________________________

DATUM allows for pooled mining while each miner decides on their block template. That's what I was referring to.

Currently, you can mine for Ocean pool and get paid the proportional share of your hashrate for the pool mined blocks while choosing your own block templates.

This gets interesting if some blocks stop being considered valid. Or if Ocean needs to check every miner's block template to decide if their shares are valid or not.

And Ocean is a company. That does way more than "organizing hashrate" so it's not crazy to ask them for their intentions on this topic.

@nprofile…0rnh @nprofile…c93l Will OCEAN pool enforce BIP-110 ...

2026-06-29T11:54:00Z

OCEAN (nprofile…0rnh) Luke Dashjr (nprofile…c93l) Will OCEAN pool enforce BIP-110 compliance for DATUM miners? And for its stratum?

What happens to a DATUM miner that builds its own block templates if it finds a block that includes non compliant transactions? Or even if it doesn't find it, does the hashrate still count?

And of the pool enforces BIP-110 even if only for stratum, I guess rewards will come from BIP-110 blocks only. Which means lower profitability (but same would happen in the other side). But additionally means that many block rewarda are at risk of being invalidated (reorg). So what happens to LN instant payouts?

Don't these questions deserve an official announcement? Maybe I've missed it but couldn't find any.

Jam v2 (JoinMarket WebUI) beta is out! Adding compatibility for ...

2026-06-29T05:49:47Z

Jam v2 (JoinMarket WebUI) beta is out! Adding compatibility for JoinMarket NG as the backend. See https://github.com/joinmarket-webui/jam/releases/tag/v2.0.0-beta.0

And /dev/fd0 (nprofile…79xu) has audited and co-signed up to the latest release of JoinMarket NG: https://github.com/joinmarket-ng/joinmarket-ng/commit/947552bead929c342282019932ff12fe17cc44af

Huge milestones! Long live JoinMarket.

what is it?

2026-06-25T05:52:27Z

In reply to nevent1q…seyf
_________________________

what is it?

Would it be possible to create a bridge between Waze or similar ...

2026-06-22T15:29:19Z

In reply to nevent1q…f4ga
_________________________

Would it be possible to create a bridge between Waze or similar and roadstr?

Estamos pendientes de que salga la beta pronto, le queda muy ...

2026-06-21T07:32:43Z

In reply to nevent1q…g9yr
_________________________

Estamos pendientes de que salga la beta pronto, le queda muy poquito.

sí porfa!

2026-06-21T07:32:10Z

In reply to nevent1q…pmw8
_________________________

sí porfa!

I think the bigger problem there is needing a company to perform ...

2026-06-21T07:31:16Z

In reply to nevent1q…7ptq
_________________________

I think the bigger problem there is needing a company to perform businesses. That's why "tainted" coins are a problem, not because they have anything wrong on their own. But even if you cared about that, LN fixes that. Blockchains don't scale anyway.

Monero is still vulnerable to many attacks that collapse ...

2026-06-21T07:26:42Z

In reply to nevent1q…c5la
_________________________

Monero is still vulnerable to many attacks that collapse anonymity. That's why they're working on FCMP++. But even with that fixed, blockchains don't scale.

Should I be angry at my grandma because of her public pension? ...

2026-06-21T07:10:46Z

In reply to nevent1q…zvlr
_________________________

Should I be angry at my grandma because of her public pension? She's receiving way more than she contributed.

Unfortunately many people benefit unfairly from the state. Should I also "hate" anyone that went to a public university or hospital?

I see the point, but don't think it's fair. I hate the state, which is clearly who steals from me. And I'm disappointed of those who defend that robbery (whether they benefit from it or not). But I don't get into second order beneficiaries. Is a public health doctor net negative? Is a university professor? Where's the line?

Monero is not (completely) private. And doesn't scale well.

2026-06-19T13:14:45Z

In reply to nevent1q…nnux
_________________________

Monero is not (completely) private. And doesn't scale well.

How would Bitcoin work if there was a human settlement on Mars? ...

2026-06-16T20:53:58Z

How would Bitcoin work if there was a human settlement on Mars? And further away?

The time it takes for electromagnetic waves traveling at the speed of light though open space is between 3 and 22 minutes depending on the relative position between the Earth and Mars.

So for example, Mars miners would be at a great disadvantage at first! And reorgs (chain forks) would occur more often.

For regular transactions, it would just mean higher waiting times for Mars users.

But Lightning Network path finding... Probably impractical!

What if there's a human (or another intelligent life form) settlement further away? As in weeks or light years far?

Should each 10 light minute radius area get their own coin? And then the market does its thing setting the exchange rates? Or?

That's called a private company 🤔

2026-06-16T19:22:24Z

In reply to nevent1q…uznp
_________________________

That's called a private company 🤔

how do I buy musk without kyc

2026-06-16T18:53:31Z

In reply to nevent1q…kyfp
_________________________

how do I buy musk without kyc

timechain!

2026-06-16T12:53:59Z

In reply to nevent1q…dggr
_________________________

timechain!

ack

2026-06-14T19:31:26Z

In reply to nevent1q…ayg2
_________________________

ack

https://bitcoincore.org/en/2026/06/06/privatebroadcast-ip-leak/

2026-06-12T09:29:45Z

https://bitcoincore.org/en/2026/06/06/privatebroadcast-ip-leak/

Is that so?

2026-06-11T20:33:50Z

In reply to nevent1q…ejjk
_________________________

Is that so?

Woild be ao mice though :(

2026-06-10T21:03:03Z

In reply to nevent1q…7uxp
_________________________

Woild be ao mice though :(

congrats @nprofile…pt0r ;-)

2026-06-10T20:40:46Z

In reply to nevent1q…zxzs
_________________________

congrats Kruw (nprofile…pt0r) ;-)

Stablecoins are winning. We can use this as an opportunity. ...

2026-06-10T20:33:24Z

Stablecoins are winning. We can use this as an opportunity.

Slides from Kilian (nprofile…pptm)

complexipy: Code complexity analyzer ...

2026-06-10T16:14:10Z

complexipy: Code complexity analyzer
https://github.com/rohaquinlop/complexipy

Blazingly fast cognitive complexity analysis for Python, written in Rust.

> Cognitive complexity measures how hard code is to understand by humans, not machines.

Unlike traditional metrics like cyclomatic complexity, cognitive complexity accounts for nesting depth and control flow patterns that affect human comprehension. Inspired by [G. Ann Campbell's research](https://www.sonarsource.com/resources/cognitive-complexity/) at SonarSource, complexipy provides a fast, accurate implementation for Python.

**Key benefits:**

* **Human-focused** - Penalizes nesting, flow breaks, and human-unfriendly logic
* **Actionable insights** - Identifies genuinely hard-to-maintain code
* **Different from cyclomatic** - Measures readability while cyclomatic measures structural, testing, and branch density

Try it with:

```shell
pip install complexipy
complexipy . --failed --suggest-refactors # in a python project
```

Example output for JoinMarket NG:

```
joinmarket-ng/tumbler/src/tumbler/runner.py
TumbleRunner::_run_taker_phase 19 ❌ FAILED

Refactor plans:
1. Extract complex block into helper function, lines 351-405
estimated: 19 -> 7 (-12)
steps:
- move the selected block into a private helper
- pass required values as parameters
- return the result/status needed by the caller
TumbleRunner::_run_maker_phase 24 ❌ FAILED

Refactor plans:
1. Extract complex block into helper function, lines 496-523
estimated: 24 -> 6 (-18)
steps:
- move the selected block into a private helper
- pass required values as parameters
- return the result/status needed by the caller
2. Flatten loop body with continue guards, lines 500-521
estimated: 24 -> 14 (-10)
steps:
- add negative condition checks at the top of the loop
- use continue for skipped items
- keep the main processing path unindented
3. Flatten nested condition block with guard clauses, lines 505-506
estimated: 24 -> 22 (-2)
steps:
- invert the outer condition
- return early when the condition fails
- move the main success path one indentation level left
- repeat for inner nested conditions where safe
```

https://stacker.news/items/1505973

Cyberpunks write code.

2026-06-10T11:53:16Z

In reply to nevent1q…c7cj
_________________________

Cyberpunks write code.

crazy

2026-06-10T11:45:40Z

In reply to nevent1q…pksx
_________________________

crazy

nice try NSA

2026-06-10T11:40:11Z

In reply to nevent1q…yhl7
_________________________

nice try NSA

Oh! I really like that idea. I guess humans having something to ...

2026-06-09T13:06:37Z

In reply to nevent1q…4d22
_________________________

Oh! I really like that idea.

I guess humans having something to lose is also what makes us "lazy". We cut corners and make mistakes because we try to save time.

It is really amazing what one can achieve with the non laziness properties of AI. I've tried many times to have my papers/books directory organized (directory structure, filenames, metadata) and failed. It just takes to be in a rush once to open the door to chaos.

But with AI, I just throw new stuff under pending/ and trigger the AI to follow the recipe. And it changes completely the mental cost of adding new content to the library.

Another big advantage is probably preventing writing fingerprinting. For anyone that has written enough publicly, there's probably sufficient information to uniquely identify that writing style.

Interesting.

Cashu is not Bitcoin. It's debt. An IOU. But financial debt ...

2026-06-09T12:57:18Z

In reply to nevent1q…yn74
_________________________

Cashu is not Bitcoin. It's debt. An IOU.

But financial debt is not bad on its own. The problem is usually the debtor. Fiat in a bank is debt squared. With the additional disadvantage of 0 privacy and collaboration with the state.

What if Bitcoin already had all it needs to scale and do so with ...

2026-06-09T12:22:25Z

What if Bitcoin already had all it needs to scale and do so with decent privacy?

I want my friends&family Bitcoin services to be as good as my Jellyfin+Ombi setup. People *ask* for an account, it's not me pushing it to them. Because it's better than Netflix, easy to use, and it just works.

Now, I want to improve my cashu mint setup. Find a nice mobile wallet to recommend (looks like minibits for now), and have a fast and robust LN node underneath so that payments work and settle fast.

Will share the final setup once it proves to work for my users.

quoting
naddr1qq…4gxm

Bitcoin doesn't kill trust, it relocates it. Inside the protocol you verify instead of trust. But the moment you touch custody, payments, swaps, liquidity, trust comes flooding back. And the weight of that trust depends entirely on who you're trusting. Not trusting a central bank whose mandate is to debase you on a schedule is a very different thing from not trusting your own father. So maybe the question isn't "how do we eliminate trust," but "what's the minimum amount of it, placed in the one person you'd trust anyway, that buys a normal human usable, private money?"

We keep answering that question wrong. Telling your mom she's a fiat slave for not running a node is the same as telling someone they're lazy for not growing their own wheat. I've self-hosted for over a decade and I'll be honest: most people who "self-host" have bad uptime, stale software, and a security setup that would make you wince. Running infrastructure is a real job. Expecting Lightning self-custody to be the default for humanity is as foolish as expecting everyone to run their own mail server. And email, also open (decentralized) and self-hostable, still collapsed into Gmail.

So here's what I've actually been doing (well a bit fancier), and it could be the solution to scaling. It's the old "Uncle Jim" idea, but hardened with the privacy tech that shipped recently. One capable person runs the boring expensive part: Bitcoin and Lightning node, Tor, private channels, coinjoined coins to fund them, ... The family and friends just hold ecash in a Cashu wallet pointed at that node. Because Cashu is Chaumian, the operator can't see balances or even count users. They pay each other instantly and privately; they swap out to cold storage when a balance gets meaningful; receiving gets wrapped with lnproxy or BOLT12 blinded paths so it doesn't leak the nodes pubkey.

It's not perfect: mistakes amplify, uptime is a single point of failure (hold accounts at two or three operators), and the operator can still guess your habits. But it beats a bank account on every axis a normal person cares about.

What gives me some hope it's natural and not just my pet scheme: we've done trust-based community money for centuries: hawala, savings circles, free banking. Buchanan's "theory of clubs" and Dunbar's 150 both suggest these things want to be small: many tiny trust-cells, not a few big banks, which is also exactly what you want for resilience and for staying off the radar. The honest counterargument is email: open protocols can still centralize at the operator layer. But money might differ, because the settlement layer underneath can't be captured the way mail deliverability was, and people are far more desperate to escape money that's actively robbing them or forcing them to ask for permission to use it, than email that's merely reading their newsletters.

Keep it genuinely small and personal: your actual family, not a business with strangers. Cover costs rather than running a service, and prefer designs with no central coordinator (for CoinJoin, LN swaps, ...).

Underneath all of it is one idea, which the cypherpunks and James Scott both got at: states need their populations legible to tax and control them, and privacy is just the deliberate engineering of illegibility. The one durable asymmetry an individual has. Which is why I think shaming people for not being pure is strategically backwards. Konkin called the real move counter-economics: don't lecture the captives, build a parallel economy cheap and private and useful enough that joining it is the obvious choice. We want the grandma who'll never run a node. Getting her onto sound, private money via an uncle she trusts isn't a compromise of the mission. It might be the whole mission.

Curious on what people think!

https://stacker.news/items/1505172

#nevent1q…565g The idea of the next Satoshi Nakamoto being ...

2026-06-09T07:55:21Z

quoting nevent1q…565g
What if the next Satoshi Nakamoto and rhe next Ross Ullbricht are not human?

The idea of the next Satoshi Nakamoto being non-human left me thinking last night. I think that won't be the case, because they lack something: caring. They just don't care. They have nothing to lose!

We could already throw an AI into a VPS with a LN wallet with some funds and ask it to run a service or something to eventually "make a living" and pay for its tokens. But why would it? Any mistake that kills its "cron" or main process would "kill" it. And I'm super sure that it will make such a mistake rather quickly. Because it doesn't care.

Even if we got through that, imagine it stays "alive" for long and creates an online service that people pay for (idk, something like nadanada/LNVPN, or nostr.wine). Does it really care about customers? It would be great at apologizing, but probably not enough to keep the service reliable and prevent mistakes on the first place.

I was thinking about this for JoinMarket development: couldn't an AI just handle questions, issues, decide the roadmap, and work on the code? Technically yes. But it probably would take at most a few weeks until there's a funds loss bug or the software becomes unusable due to feature bloat for example.

The difference is that a human cares. The same thing that makes us reluctant to admit mistakes, is what motivates us to avoid them "harder". Ego? Reputation? Responsibility? An AI doesn't have that, for good and for bad.

That's why AI produces sloppypasta:

quoting naddr1qv…nsa5

Things are insane right now, and maybe I'm too. Be that as it may, I feel like I've been a broken record for the last 18 months. Maybe longer. Some nights I wake up at 3am, startled, sweaty, and I hear myself screaming "VERVAEKE!" before I cry myself back to sleep.

He's right about everything, you know, this John Vervaeke. Well, maybe not everything. But most things. I believe he's right about the AI stuff and the coming thresholds and the paths we must take. (I've shared this video so many times with so many people now, it's not even funny; I think I've shared it almost as often as his Awakening from the Meaning Crisis. You should watch both. And yes, the 2nd one is ~50h long. You should watch it twice. Maybe even three times, like I did.)

The reason why I'm tempted to produce this stream-of-consciousness that you're reading right now is that I stumbled upon a new word today: Sloppypasta. It's perfect. Chefs-kiss perfect.

Here's what LLMs can't do, and unless we give them bodies (and true pain, and real suffering, and proper death) it's a thing they'll never be able to do: care. And that's also the reason why people hate Sloppypasta: it's done without care.

That's why your eyes will glance over anything that has LLM smell. Be it text, or images, or infographics, or posters, or stock "photography", or tweets, or videos, or whatever. Humans are pattern matching machines, and the pattern you've identified is a lack of caring. Prompt goes in, slop comes out. Zero effort. Zero care. Sloppypasta.

That's also why open-source maintainers are locking down their repos, straight up blocking outside contributions. It's so easy right now to create a drive-by PR for a project, fixing a supposed "issue" (that might not even be in an issue in the first place, but you wouldn't know because you don't care enough to dig into the project to truly understand what's going on, and why things are built in the way they are built) that maintainers are flooded with supposed "fixes" and other vibed things.

Don't get me wrong. I'm not a Luddite. Far from it. If you've ever listened to the wind you'd know that I've been vibing from Day 1. It's how I've built Boris, it's how I've built ants, it's how I've built Castr, it's how I've built anything in the last 18 months or so. Including UvS.

Vibe-coding is what brought me back to programming, precisely because it's fun and easy and painless and quick. It's low-effort. It can be done on the side.

However, if you do do it on the side, it's by definition done without care. All the projects that are worth anything took thousands upon thousands of iterations, vibe-coded or not. Lots of thought. Lots of care. Human thought. Human care.

That's what LLMs can't do yet, unless we turn them into humans. With human bodies, human problems, human desires, and human tastes and values. That's what Vervaeke is talking about. Deriving an ought from an is.

LLMs are smart,¹ no doubt about it. But they lack wisdom, deliberation, and care. The right amount of care. The thing that can't be put into an algorithm. The thing that isn't a statistical average. The thing that makes us human.

This article first appeared on dergigi.com.

There's a reason why there's such a thing as a "smart"-ass. And there's a reason why a person can be PhD-level smart but still incredibly stupid in other ways (COVID, anyone?) - or incredibly knowledgeable and still an uncaring asshole. LLMs are even worse, because they're insufferable without being an asshole. ("You're absolutely right! I'll keep that in mind from now on.")

Pieter Wuille (fun) Facts http://pieterwuillefacts.com/ ...

2026-06-08T19:37:37Z

Pieter Wuille (fun) Facts
http://pieterwuillefacts.com/

https://stacker.news/items/1504789

sad :(

2026-06-08T18:42:21Z

In reply to nevent1q…n7va
_________________________

sad :(

Am I Being Pwned? - Browser Extension Scanner ...

2026-06-08T15:14:17Z

Am I Being Pwned? - Browser Extension Scanner
https://amibeingpwned.com/

https://stacker.news/items/1504617

DeepSWE: Measuring frontier coding agents on original, ...

2026-06-08T14:51:15Z

DeepSWE: Measuring frontier coding agents on original, long-horizon tasks
https://deepswe.datacurve.ai/

![benchmark](https://m.stacker.news/144105)

> Today's leading public coding benchmarks are starting to saturate at the frontier: top models cluster within a narrow score band where adjacent configurations often overlap on confidence intervals. DeepSWE is a long-horizon software engineering benchmark built to separate them. It delivers four advances over existing public benchmarks:

> **Contamination free**: Tasks are written from scratch, not adapted from existing commits or PRs, so no model has seen the solution during pretraining.

> **High diversity**: Tasks span a broad pool of 91 repositories across 5 languages.

> **Real-world complexity**: Prompts are ~half the length of SWE-bench Pro's, yet solutions require 5.5x more code and \~2x more output tokens.

> **Reliable verification**: Verifiers are hand-written to test software behavior rather than implementation details.

> The result is a benchmark that reflects how today's frontier coding agents actually perform in software engineering work.

https://stacker.news/items/1504607

How Did DeepSeek V4 Make V4 So Cheap? - YouTube ...

2026-06-08T14:28:05Z

How Did DeepSeek V4 Make V4 So Cheap? - YouTube
https://www.youtube.com/watch?v=gC76aeibdFA

Paper: https://www.alphaxiv.org/abs/deepseek-v4

https://stacker.news/items/1504595

Looking for feedback on ...

2026-06-08T08:15:55Z

Looking for feedback on https://github.com/joinmarket-ng/jmp/pull/3 (Taproot offers)

The Economics of Protocol Design: Proposing Lightning over Ark‬ ...

2026-06-08T06:04:56Z

The Economics of Protocol Design: Proposing Lightning over Ark‬ - YouTube
https://www.youtube.com/watch?v=hqKQ4Rky3tY

René Pickhardt at BTC++ conference.

Description: Venture Capital Money has a surprising influence over protocol design. Not rarely technical sound solutions won't find consensus due to market sentiment and selfish interests of participants. We investigate this with the example of the debate about ark vs. lighting network as the future payment system for Bitcoin.

Paper: [https://arxiv.org/abs/2601.04835](https://arxiv.org/abs/2601.04835)

https://stacker.news/items/1504446

yes! what's about to say that. And that did not save them

2026-06-07T19:57:10Z

In reply to nevent1q…0ckm
_________________________

yes! what's about to say that. And that did not save them

And if you set the minimum reserve to match those, there's no ...

2026-06-06T19:21:33Z

In reply to nevent1q…gypv
_________________________

And if you set the minimum reserve to match those, there's no security problem anymore.

It uses the underlying node default fee limits. Right?

2026-06-06T19:19:31Z

In reply to nevent1q…c3uc
_________________________

It uses the underlying node default fee limits. Right?

We're now co-signing JoinMarket NG releases! @nprofile…79aj ...

2026-06-05T10:29:44Z

We're now co-signing JoinMarket NG releases!

/dev/fd0 (nprofile…79aj) has reviewed up to 0.30.0 and that's the first co-signed release.

https://github.com/joinmarket-ng/joinmarket-ng/commit/9a6e8d4333ff716336c05671e07bd9454dc7f43d

He also fixed a docker image reproducibility issue.

Still looking for more eyes. If you can review the code and want to co-sign with us, please reach out.

do they? they look more like inflation "detection" more ...

2026-06-05T08:40:32Z

In reply to nevent1q…elu5
_________________________

do they? they look more like inflation "detection" more than explanations on why it would be unlikely for it to happen in the first place

What's interesting is how they manage to keep the TRY ...

2026-06-05T08:21:34Z

In reply to nevent1q…5pe8
_________________________

What's interesting is how they manage to keep the TRY exchange rate more or less stable against other currencies. Prices have rosen like crazy but the TRY/EUR pair has barely moved.

That doesn't solve the "censoring". Primal uses a ...

2026-06-05T05:44:04Z

In reply to nevent1q…argz
_________________________

That doesn't solve the "censoring". Primal uses a special caching relay for reading. Which is not a nostr relay but some custom thing. I guess you could self host that, but it's not as simple as configuring additional relays.

On top of that they have an npub blacklist which I'm not sure where it comes from and that might not be easy to avoid.

I have a problem. I only like Primal for its UX but I'm ...

2026-06-04T20:49:48Z

I have a problem. I only like Primal for its UX but I'm getting tired of "mentioned event not found". Did not come to nostr for censorship but for censorship resistance.

Any Nostr client recommendations? Have tried Iris and wisp already and was not convinced :(

Good luck enforcing that🤭

2026-06-04T18:10:55Z

In reply to nevent1q…hj5u
_________________________

Good luck enforcing that🤭

sad :(

2026-06-04T06:20:36Z

In reply to nevent1q…f632
_________________________

sad :(

sad :(

2026-06-02T08:18:08Z

In reply to nevent1q…27wq
_________________________

sad :(

Fast, efficient, and high-quality text-to-speech for GrapheneOS ...

2026-06-02T08:14:14Z

Fast, efficient, and high-quality text-to-speech for GrapheneOS
https://github.com/GrapheneOS/SpeechServices

Speech Services provides fast, efficient, and high-quality text-to-speech for [GrapheneOS](https://grapheneos.org) using state-of-the-art models running completely on-device.

https://stacker.news/items/1500823

The relay is not the problem. There are many relays already. But ...

2026-06-02T08:10:07Z

In reply to nevent1q…s42q
_________________________

The relay is not the problem. There are many relays already. But Primal is designed around their caching relay which is not easily replaceable.

Sure. But there's a big difference between who has access to ...

2026-05-31T19:16:20Z

In reply to nevent1q…3ecf
_________________________

Sure. But there's a big difference between who has access to it and what it stores. Selectively excluding notes sounds pretty intentional.

Is Primal censoring stuff? https://stacker.news/items/1499789

2026-05-31T11:40:26Z

Is Primal censoring stuff?

https://stacker.news/items/1499789

F

2026-05-31T11:23:24Z

In reply to nevent1q…pem0
_________________________

F

Cashu Mint Auditor https://audit.8333.space/ This app sends ...

2026-05-30T20:48:23Z

Cashu Mint Auditor
https://audit.8333.space/

This app sends random Lightning payments between mints. Forever. The sats go in a circle until all is lost to fees or rugs. If a mint fails to pay an invoice, it is marked with a scary red label.

https://stacker.news/items/1499492

can't zap you! but thanks :-)

2026-05-30T15:43:07Z

In reply to nevent1q…72cp
_________________________

can't zap you! but thanks :-)

do you have a link to it?

2026-05-30T15:01:28Z

In reply to nevent1q…8zcn
_________________________

do you have a link to it?

Wouldn't it be nice so receive Silent Payments and CoinJoin ...

2026-05-30T12:50:59Z

Wouldn't it be nice so receive Silent Payments and CoinJoin them after?

Or how else could a developer receive Bitcoin donations with some privacy?

PR for Silent Payments (BIP352) support for JoinMarket NG: https://github.com/joinmarket-ng/joinmarket-ng/pull/513

Have been using the now available Android app for a while now and ...

2026-05-30T11:39:17Z

In reply to nevent1q…x98k
_________________________

Have been using the now available Android app for a while now and unfortunately will have to go back to WireGuard configs. Mainly the missing features are:

- Widget to temporarily toggle it off. And that still works with the "Always on" VPN setting. So that you can actually use the Internet connection with the VPN temporarily off through the widget.
- Exclude addresses on local networks.

Other things are not that important. But can't live without these two!

Super happy with the service in general though.

How does the Lightning channel management work? Is the Bitcoin ...

2026-05-30T11:15:16Z

In reply to nevent1q…53mf
_________________________

How does the Lightning channel management work? Is the Bitcoin node within the TEE or external?

Privacy-preserving license plates using blind signatures License ...

2026-05-30T10:55:57Z

Privacy-preserving license plates using blind signatures

License plates are crazy from a privacy standpoint. We'd riot if the government made us wear a badge with our name on it every time we left the house. But bolting a permanent unique identifier to our car? Totally normal.

Now there are ALPR (automatic license plate reader) networks everywhere, public and private, logging everything. Check https://deflock.org/ for the USA picture. From plate data + camera feeds (public or insecure private ones) you can infer where someone lives, works, when they're not home, who they meet. Some countries even let you query the vehicle registry directly from the plate.

I don't think accountability for cars is necessarily wrong. If you rear-end someone and flee, it's fair they can identify you. I just don't think we need to trade mass surveillance for that. Cryptography exists.

A basic scheme:

Every day you generate a random string r, compute H(r), and get it blind-signed by an issuing authority (they sign without learning what they're signing: Chaumian blind signatures, same as ecash). You put r on your e-ink plate as a QR code, along with the signature. Anyone can verify: hash the preimage, check the signature. If you cause an accident, the witness notes r and can prove they saw your plate. To find out who you are, they go to court, the authority looks up who requested the blind signature for H(r). No public link between your plate and identity. Already way better than today.

Problems and improvements:

Daily rotation is too slow: Cameras still build a full-day profile. Rotate every 5–10 minutes instead. Eric Rescorla arrived at a similar scheme: https://educatedguesswork.org/posts/license-plates/ and explores the tradeoffs well, including how the authority can precompute a lookup table for all vehicles and time windows cheaply enough that de-anonymization doesn't require your cooperation.

The authority is still a chokepoint: Whoever registers you and creates the H(r) -> identity link already knows both sides. Threshold encryption of the stored record is a governance improvement (requires k-of-n parties to cooperate, auditable) but not a fundamental one. The registrar still has the information at creation time. The real protection is legal process + the fact that passive surveillance is broken. That's already a big deal compared to today, but it's not zero-trust.

The cryptography is production-ready. The gap between "solved" and "deployed" is almost entirely political.

Posted to https://stacker.news/items/1499201

Rewrote the "JoinMarket Maker Clustering and Taker ...

2026-05-30T10:10:38Z

Rewrote the "JoinMarket Maker Clustering and Taker Anonymity-Set Reduction" article with a clear example as introduction, a glossary, and trimming most sections.

Published at https://joinmarket-ng.github.io/coinjoin-simulator/mainnet-deanon.html (same URL, old version preserved in git).

The discussion about possible mitigations is ongoing at https://github.com/joinmarket-ng/joinmarket-ng/issues/508

BTW the JM protocol is not broken, just improvable.

JoinMarket NG 0.31.0 release is out. More improvements about deep ...

2026-05-29T15:27:47Z

JoinMarket NG 0.31.0 release is out.

More improvements about deep wallet scanning. Also security hardening, TUI improvements, and bug fixes. Automated Win/Mac/Linux install tests.

https://github.com/joinmarket-ng/joinmarket-ng/releases/tag/0.31.0

Max Hillebrand: The Praxeology of Privacy - YouTube ...

2026-05-26T13:38:14Z

Max Hillebrand: The Praxeology of Privacy - YouTube
https://www.youtube.com/watch?v=A4_1E2kJ5_w

Timothy Allen speaks with Max Hillebrand.

Max joins me to explore freedom, privacy, and property rights in the free market of ideas. We discuss Cypherpunk strategies, the appeal of Bitcoin, Free Cities, and digital nomadism as ways to resist creeping socialism and surveillance. Drawing on Lockean ownership theory, Austrian economics, and praxeology, Max makes the case for combining decentralization, community, and technology to defend liberty in both physical and digital realms.

TIMESTAMPS:
0:00:00 — Coming up
0:00:41 — Veritas Village preview
0:01:27 — Start of conversation
0:10:02 — Triangular Interventions of the State
0:11:26 — The banning of VPNs
0:16:27 — Mean Time To Harassment
0:24:43 — The Cypherpunk Ideal: Increasing the Cost of Attack
0:30:38 — Dragnet Surveillance & Privacy
0:36:48 — Free Cities and The Importance of Freedom in Meatspace
0:44:04 — There are No Frontiers Left
0:53:08 — Conscription is Coming Back
1:00:57 — There are Many More Good People Than Bad
1:06:53 — AI and Robots of Convenience in the Dystopian Future
1:18:25 — Bitcoin Proves John Locke's Theory of Property Rights
1:25:29 — Proving Economic Reasons
1:31:40 — The Cypherpunks Don't Know How Fundamentally Correct They Are
1:36:47 — Freedom is Correct. The State is Evil
1:49:22 — AI Will Remove Scarcity from the World
1:53:50 — Keynsian Bullshit
1:58:08 — Tik Tok Will Eventually Generate 100% AI Content
2:10:35 — Rally Cry for Freedom Lovers

NOSTR:

Max Hillebrand: npub1klkk3vrzme455yh9rl2jshq7rc8dpegj3ndf82c3ks2sk40dxt7qulx3vt
Timothy Allen: Search '[timothy@nostr.com](mailto:timothy@nostr.com)' on your Nostr app
Free Cities Foundation: npub1lsj8pmgedqqamt89c27tzjjnlf0wn7q7udjm7j2cl9xxz97eacns2mwpee

https://stacker.news/items/1496628

Please no! xD Take everything as open for discussion, and expect ...

2026-05-26T13:34:52Z

In reply to nevent1q…ycst
_________________________

Please no! xD

Take everything as open for discussion, and expect only optional changes that users will decide if they use or not.

Well you don't need the lightning node to run within the ...

2026-05-26T13:15:57Z

In reply to nevent1q…5ejt
_________________________

Well you don't need the lightning node to run within the enclave right? Just a "smart" signer that signs HTLCs given the redeemed cashu token. For rebalancing maybe some rule of "not loosing funds" and someone external has to pay the rebalancing fees. You can prove to the enclave that the rebalancing succeed or that a LN invoice for which it only revealed the payment hash has been paid.

I'm sure there are lots of challenges here, but the main point is that it doesn't need to run "everything".

Been digging into JoinMarket maker clustering on mainnet. the ...

2026-05-25T21:02:04Z

Been digging into JoinMarket maker clustering on mainnet. the short version: yes you can cluster makers by fee fingerprint from onchain data alone, and yes it reduces taker anonymity sets. But JM holds up pretty well in practice. mean anonset goes 7.6 -> 6.9, and the mitigation (fee policy homogenization) *can* be a client default change, no protocol surgery needed.

The counterintuitive part: some "obvious" countermeasures like makers avoiding change as input actually make things worse.

Rough draft, not peer reviewed, happy to get feedback: https://joinmarket-ng.github.io/coinjoin-simulator/mainnet-deanon.html

I do! It started as an LNbits instance but felt uncomfortable ...

2026-05-25T20:59:08Z

In reply to nevent1q…rjxj
_________________________

I do! It started as an LNbits instance but felt uncomfortable with seeing their balances and all transactions!

Now I'm able to give them that service with the same custodial "risk" but at least with some privacy.

myself for friends and family

2026-05-25T05:58:35Z

In reply to nevent1q…ut6u
_________________________

myself for friends and family

The Ultimate Lightning Network Privacy Explainer ...

2026-05-24T16:02:16Z

The Ultimate Lightning Network Privacy Explainer
https://voltage.cloud/blog/lightning-network-privacy-explainer

https://stacker.news/items/1495707

Why meshcore over meshtastic? Have heard contradictory arguments.

2026-05-24T13:26:14Z

In reply to nevent1q…rqyg
_________________________

Why meshcore over meshtastic? Have heard contradictory arguments.

Blockchains don't scale.

2026-05-24T09:05:27Z

In reply to nevent1q…63a6
_________________________

Blockchains don't scale.

Please name a place on earth where it's easier to make money. ...

2026-05-24T06:52:56Z

In reply to nevent1q…lck6
_________________________

Please name a place on earth where it's easier to make money. It might be true that building wealth is hard in general and that the USA is the easiest place on earth to do so.

Texas sues Meta, WhatsApp over encryption privacy claims ...

2026-05-23T21:00:47Z

Texas sues Meta, WhatsApp over encryption privacy claims
https://www.reuters.com/legal/government/texas-sues-meta-whatsapp-over-encryption-privacy-claims-2026-05-21/

Texas’ lawsuit cites news reports about a federal investigation into ‌claims ⁠that Meta had access to unencrypted WhatsApp messages and a whistleblower report to the U.S. Securities and Exchange Commission.

https://stacker.news/items/1495402

Nice to see you back!

2026-05-22T12:17:29Z

In reply to nevent1q…4f9e
_________________________

Nice to see you back!

How are you going to make any profits in a rigged system? Every ...

2026-05-21T17:12:21Z

In reply to nevent1q…7u9u
_________________________

How are you going to make any profits in a rigged system?

Every wall is just an added cost.

Not exactly Ocean IMO. You can choose your own blocks with DATUM. ...

2026-05-21T15:47:35Z

In reply to nevent1q…surl
_________________________

Not exactly Ocean IMO. You can choose your own blocks with DATUM.

Before, they used to have different template options for stratum miners as well. Not nice that they've removed this feature.

Lightning Network privacy is real. But "better than ...

2026-05-21T15:42:15Z

Lightning Network privacy is real. But "better than on-chain" isn't the same as "private."

quoting
naddr1qq…0xsd

Lightning Network privacy is real. And limited. Let me explain both.

LN is a genuine privacy improvement over on-chain Bitcoin. Payments don't touch the blockchain, they're not broadcast to the whole network, and onion routing means each hop only knows its immediate neighbors. If you're paying for something, your counterparty will learn little about who you are.

But "better than on-chain" isn't the same as "private."

A 2020 paper by Kappos et al. titled “An Empirical Analysis of Privacy in the Lightning Network”, did a thorough job of poking holes in four privacy properties LN “claims” to offer. Here's what they found:

1. Private channels aren't really private

LN lets you open channels without announcing them to the network. In theory, nobody knows they exist. In practice, the researchers used blockchain analysis (following "peeling chains" of Bitcoin UTXOs) to identify ~27,000 likely private channel openings. Of those, they could identify at least one participant in 79.3% of cases. Meaning that they could link it to a public node on the network.

So the channel is quiet on the network graph, but it's still written to Bitcoin's public Blockchain.

Additionally , there's a point not covered in the paper: even without any "fancy" analysis, you would directly need to disclose your "private" node's pubkey and routing hints (channels a.k.a. UTXOs) to the payer. So the privacy of that opening channel UTXO does matter.

2. Your channel balance can be probed

This one is the most direct attack. An active adversary opens channels with nodes in the network and routes fake payment hashes with varying amounts (binary search). The success or failure of those probes reveals the exact balance of a channel. On testnet, they probed 1,017 channels and successfully determined the balance of 568 of them (~56%). No special privileges needed, just some sats (that don't even have to be spent) and patience. The attack is generic enough that even if LN clients stop returning detailed error messages, it still works.

3. Routing nodes can guess who you are

When your payment is routed through an intermediate node, that node isn't supposed to know if you're the sender or just another hop. The paper shows the simple strategy of "assume my immediate predecessor is the sender" works surprisingly well:

- For successful payments: the attacker correctly identifies the sender between 15% and 57% of the time (depending on network parameters)

- For failed payments: up to 83%. Because a failed payment leaks path length information

34% of LN payments fail on the first try (per measurements cited in the paper). That's a lot of free signal for routing nodes.

This is partly structural: LN's topology is centralized, paths are short, and clients are designed to find the cheapest (and therefore shortest) route. Short paths mean fewer hops hiding you.

https://blossom.primal.net/a57b13f8f1e98c9deb70a95a29822dd205b264f070b9b0d3f67716a6046d81ad.png

4. Balance snapshots reconstruct payments

Combine the balance probing attack with a periodic snapshot approach: an attacker checks channel balances at time t, then again at t+τ, and looks at what changed. If a path of channels shows correlated balance shifts, there was likely a payment along that path.

At a snapshot interval of 30 seconds (already feasible in their experiment), the attacker achieves:

- Recall >67% (correctly detecting payments that actually happened)

- Precision ~95% (most detected payments are real, few false positives)

The faster they sample, the better it gets. And the attacker doesn't need to be involved in routing the payment at all.

---

What helps:

None of this means LN is broken. But it means there are real mitigations worth using:

- BOLT12 / blinded paths: instead of your node's pubkey being in every invoice (BOLT11), blinded paths hide the recipient's identity. This is a big deal.

- lnproxy: wraps the invoices through a proxy node to further obscure the final recipient. It's trustless for funds but not for privacy respect the lnproxy operator (which will see the memo and final destination).

- Private channels: reduce your surface area on the public graph, even if not fully hidden (see above). Using "private enough" UTXOs for opening those channels and ideally having a private channels only node.

Even with all of this, the anonymity set is inherently small, since there aren't that many Lightning nodes.

---

The same lesson applies here as everywhere else in privacy: LN gives you real, meaningful privacy improvements over on-chain Bitcoin, and zero privacy guarantees. The threat model matters. Who are you hiding from, and what are they willing to do?

Use Lightning. It's genuinely better. Just don't use it as if it made you invisible.

---

Kappos et al., "An Empirical Analysis of Privacy in the Lightning Network" (2020)

arXiv: https://arxiv.org/abs/2003.12470

It's a very nice problem to have compared to the confiscation ...

2026-05-21T15:40:50Z

In reply to nevent1q…ywk8
_________________________

It's a very nice problem to have compared to the confiscation problem ;-)

The latest JoinMarket NG release comes with major improvements ...

2026-05-20T17:15:39Z

The latest JoinMarket NG release comes with major improvements for wallet performance, specially for exiting old wallets with lots of transactions. Also security hardening, TUI improvements, and bug fixes.

https://github.com/joinmarket-ng/joinmarket-ng/releases/tag/0.30.0

why not use Bisq or another P2P method?

2026-05-19T15:02:46Z

In reply to nevent1q…278l
_________________________

why not use Bisq or another P2P method?

Nostr event nevent1qqsg2ddpmq80q3q72hklzg2h76rz99qpsty9hl40f6c8nzz8ukf0txczyp69n5en4anqvmcxdnu809hxjrdn49hlg560nm05a2m5muhjqu5fkegzsv4

2026-05-19T05:53:16Z

In reply to nevent1q…uvvj
_________________________