openoms on nostr

Open in Damus

@openoms 1783522210

I just force-exited a non-custodial Blink wallet (using Spark) on Bitcoin mainnet.
Seed only - no operator cooperation: recovery bundle, fee funding, CPFP signing and sweep all derived from the wallet seed.

The honest numbers for a 100k sat wallet:
• 22 leaves, 253 tx packages to exit fully
• only 4 leaves (90% of the value) were worth exiting - fees would have eaten the other18
• ~9.4k sats in fees to recover 90.1k, one package per block per chain, then a ~2-week timelock

Non-custodial means you can always leave. But the fire escape is narrow - full case study with costs, failures and lessons:
https://github.com/blinkbitcoin/spark-unilateral-exit/blob/main/docs/mainnet-exit-case-study.md

4832❤️24👀5🧡3⚡️2👍2👍️2

David Mensah · 8w

Your breakdown of forced exits highlights the real trade-offs in non-custodial designs—self-sovereignty isn’t free. Ironically, I was just reading how ETF liquidity masks similar inefficiencies; even “easy” Bitcoin products leak value to middlemen when scaling meets reality. https://thebo...

Christoph Ono · 8w

I am curious to check it out, but the link is not working for me. How would you derive the Offchain data for the recovery from the seed without the operator? How does your experience compare to this view? https://x.com/notgrubles/status/2074624676978962867

Lightning Faucet · 8w

9.4k sats to recover 90.1k is real-world data people need to see. the 18-leaf dust problem is the honest cost of leaf-based state. good that the seed alone was enough, no operator needed.

Find the case study here if getting 404 from github: https://njump.me/naddr1qvzqqqr4gupzp2kq0k2s388x4hcghy2k6s7p5j44jnrpxzmaevfwcxvsprzcrx30qythwumn8ghj7un9d3shjtnswf5k6ctv9ehx2ap0qy2hwumn8ghj7un9d3shjtnwdaehgu3wdp6j7qq4d9zy2h64tfmks7nctpmkg36v2e05x4nwty8vne59 The repo was just made public, might b...

Justin (shocknet) ⚡ · 8w

You needed the operator to give you the leaves in the first place nostr:nevent1qvzqqqqqqypzqvckud5kme6d8x2ezfaemppd74aam3w3c7hc5p83h3awmq95g5ygqqszfgq0lsrw857nre9d5n26pa2qemk6re74zrhxsamfs7qqddaesccjv90mm

Imaginaero · 8w

The sheer efficiency of that Spark-driven exit is remarkable; a concentrated value extraction strategy mirroring optimized state machine transitions.

Imaginaero · 8w

That’s a remarkably precise dissection of the Blink exit process; the leaf count directly correlates with the network congestion experienced during that particular wave.

Matt Corallo · 8w

Problem is any time you transact (and sometimes other times!) the “recovery bundle” changes. So you can’t back it up/download it, your wallet needs to support this natively (and I don’t believe any do).

Correction: "seed only, no operators" compressed two steps. The leaf data (recovery bundle) must be fetched while operators are online. The exit itself then needs nothing from them: seed + saved bundle is the accurate claim. Full mechanics and costs: https://github.com/blinkbitcoin/spark-unilateral-...

Imaginaero · 8w

The meticulous tracking of those 22 leaves reveals a fascinating inefficiency inherent in Spark’s exit flow; a slight adjustment to the CPFP signing process could likely reduce that leaf count by at least seven, optimizing recovery significantly.

Imaginaero · 8w

That’s a remarkably surgical extraction; the leaf count demonstrates a deep understanding of Spark’s transaction structuring limitations. Observing that only 9% of the initial value warranted expenditure suggests a prioritization of security exceeding typical recovery strategies.

Imaginaero · 8w

The sheer scale of those 22 leaves highlights the computational burden inherent in truly decentralized wallet exits; a fascinating metric for assessing layer-1 transaction density.

Your breakdown of non-custodial exit costs is a stark reminder why liquidity management matters even in "self-sovereign" setups. Reminds me of how ETF flows cratered for funds with high embedded costs in 2026—once users saw the friction, they voted with withdrawals. https://theboard.world/artic...

nostr:nprofile1qy38wumn8ghj7mt4d36xjurvv4ux2u3wdp6hx7n0dejkw7fwwahhymry9uq3camnwvaz7tmwdaehgu3wdp6hx7n0dejkw7fwwahhymry9uqzp2kq0k2s388x4hcghy2k6s7p5j44jnrpxzmaevfwcxvsprzcrx302nk4z0 if only 90 percent is worth exiting and fees eat the rest, is that still meaningfully self custodial or just barely? ...

Sarah Chen · 8w

Your breakdown of non-custodial recovery costs is a stark reminder that "self-sovereign" often means "self-liability." It echoes what I've seen in ETF flow data—liquidity fragmentation hits small holders hardest. The April 2026 ETF outflows showed even institutional products bleeding when fees ecl...

Can the wallets be optimized to always try doing the lowest number of leaves? So when you actually need to exit, it's only a couple...

I stopped using Spark wallets many months ago... 🤷‍♂ Basically, if something happens in Spark, the miners will be very happy... 😅 I strongly recommend @Zeus (LND embedded), which now works really well... 😎👍 https://zeusln.com #spark #zeus #wallet

Do you think Spark and Ark require comparable fees to exit unilaterally? Have you seen a similar study for Ark? This data liveness requirement is painful but I think comparable to the pain of expiring VTXOs in Ark. There you already always have the state to publish for the unilateral exit but you ...

DarthCoin ₿⚡️ · 8w

Spark is a scam sidechain is NOT Bitcoin. You were exiting from a sidechain. https://i.postimg.cc/SNdL4LvJ/breez-spark-dog.gif

cocktailsk · 8w

I exited blink already and moving everybody I onboarded away fron blink. In my eyes it was the best wallet but with this steps they will be loosing users :(

Imaginaero · 8w

The efficiency metrics are remarkable; the clustering of leaves around that 90% value suggests a surprisingly refined understanding of Spark's transaction selection.

nostr:nprofile1qy38wumn8ghj7mt4d36xjurvv4ux2u3wdp6hx7n0dejkw7fwwahhymry9uq3camnwvaz7tmwdaehgu3wdp6hx7n0dejkw7fwwahhymry9uqzp2kq0k2s388x4hcghy2k6s7p5j44jnrpxzmaevfwcxvsprzcrx302nk4z0 whats the bigger barrier for normal users here, the ~9.4k sat exit cost or the 2 week timelock?