I gave a talk on lichen at bitcoin++ in Seoul. Here's the slides:

https://btcpp.nyc3.digitaloceanspaces.com/seoul/presentations/lichen-l2sinchannelsenforcedinbitcoin-1793924417028220841-873d1893b609.pptx

Forst person to person lichen payment successful!

Here's my address if you want to test sending somewhere:

8a8a116a8987ccd2e584b16c60a71df087c2eea4cafbcb98673d1e060ecc8b7d

Thanks to Daniela Brozzoni (and bitcoin red team) for a scan that picked up several bugs in the old joinmarket-clientserver repo, one in particular that was an especially big deal - a direct funds loss of a sender in the old payjoin implementation (bip78). Thank god very few if any people really used that but we did include it directly in the client software, so I'm sure at least a few tried it here and there (with btcpayserver e.g.). I've now changed the repo from "archived; don't recommend you use this" to "archived and DO NOT USE THIS and BIG SCARY WARNING about funds loss" and repeated the warning on every release page just in case.

Worth mentioning that both I and, to a larger extent, @m0wer , did a fair bit of scanning of the joinmarket repos looking for bugs a few months back; we found quite a few but nothing with direct funds loss like that one ^ . So this was extremely valuable extra scanning.

https://github.com/JoinMarket-Org/joinmarket-clientserver/blob/master/README.md

I recommend joinmarket-ng if you want to do joinmarket coinjoins, but bear in mind, I do not work on Joinmarket any more. The general meta comment is of course that "don't use unmaintained software in 2026" is good advice, because most things with any complexity are broken unless actively audited.

A lichen L2 hub, running on Mutinynet, open for testing:

Mac or Linux. You need, on your machine, a C compiler (Mac: xcode-select --install; Debian/Ubuntu: build-essential) and Rust (rustup.rs). About 1 GB of disk and a few minutes to build:

git clone https://github.com/AdamISZ/lichen-gui
cd lichen-gui
./build.sh
target/release/lichen-gui

It opens a page in your browser. Then:

To connect to my hub you need the urls for the hub and sequencer, paste them into the fields to connect:
1. hub http://193.168.140.156:18310, sequencer http://193.168.140.156:18300. Check the keys it shows:
sequencer fc760de0f51debf97c3a116c19180660b21661c8ddfd5f18bcbae304a01b5ea5
hub 0d530a46a469ecaf7abe72a89cf737c88bf7dcb1ce149f86e0b8ae6186af8665

2. Open a contract (defaults: 100k deposit, 100k inbound). It shows a deposit address and amount, about 151k sats: pay it from https://faucet.mutinynet.com, in ONE payment. Opening takes a few minutes (it checks the hub's BABE setup and downloads about 240 MB).

3. Pay someone on the L2, or ask to be paid: near-instant. I'll keep an eye on replies here, if someone wants to test paying back and forth with me, send me your address here. My address is 8a8a116a8987ccd2e584b16c60a71df087c2eea4cafbcb98673d1e060ecc8b7d

4. Withdraw: about 5 minutes (it waits for an anchor on Bitcoin and its proof). Withdrawals and your reserve go to the wallet's own Bitcoin address, so you can pay your next deposit from it: the faucet is only needed once.

It's an early prototype: expect rough edges, and tell me what breaks. Disputes and shielded payments aren't in the GUI yet.

If you hadn't read anything I wrote here before about lichen, check out https://github.com/AdamISZ/lichen there's a paper there.

the binance hack will be a huge setback for those people who want to push bitcoin and cryptocurrency as "legitimate". It hands a perfect argument to those in power who hate bitcoin and want to shut it down, much more so than say FTX which was just "a bunch of people getting scammed". Now they can say (they already did, but this hack is *huge*) that we are literally funding North Korea's entire nuclear weapons program.

I could be wrong, but what I'm noticing about these headline-grabbing (deservedly) results from frontier models on long-held theories of computation limits is, that they are more showing that the long-held theories are approximations, not so much that they are wrong. The "Exponential Time Hypothesis" and the square-root bound for ECDLP are in this category: both the results that show they are wrong, show they are wrong to an absurdly infinitesimal extent. Maybe it's more like saying that things are fuzzier than you think, not clean, not so much saying that the general rule you have in mind is wrong. Just a guess, though.

How have I only today learned that the plural of lemma is lemmata.