This week we saw two new grants going to Lightning builders!
- Superlunar issued a grant to Dusty Daemin to support his work on Lightning, and notably splicing.
- Okcoin is now sponsoring Johns Beharry’s work of promoting Lightning projects and experiments through BoltFun.
Legends Of Lightning Update
Speaking of BoltFun, I thought it would be interesting to see how the LegendsOfLightning tournament has evolved since we last took a look. There are now 25 different projects competing, and each of them is sharing their story along the way! Go take a look!
Seems like TabConf was lit! 🔥
yep, @tabconf was *def* the best #bitcoin dev conf I've been to in years 😭— Olaoluwa Osuntokun (@roasbeef) October 18, 2022
if you develop any sort of Bitcoin software n didn't go, you should be feeling a tremendous amount of FOMO, self-doubt, and inadequacy
but ofc, don't be too hard on yourself, there's always next year!
André Neves gave the keynote, detailing how Bitcoin can thrive through developer education, mainstream adoption and open source software, and what they’re doing at Zebedee and NBD.
There was also a cool presentation on where we’re at with Eltoo.
Wallet & Tools
Impervious Browser Is Live
Impervious Browser is live! The browser features encrypted messaging, peer-to-peer group video calls, collaboration features and native support for Lightning.
And Keet Now Supports Lightning
You can now send and receive sats in a Keet Meeting. In the settings, you can now configure your wallet by connecting to an existing node backend (either LND, Eclair or Core Lightning). Once it’s done, you’ll be able to transfer value P2P with other people in meetings.
Alby 🤝 Nostr
It was a bit teased in Latest Strikes #5, but here it is: Nostr integration in Alby. Users can easily generate new keys from the extension, and said keys are then managed by the extension itself. It will allow users to sign messages on web Nostr clients with the click of a button (no more copy-pasting of private keys!).
Centrality in BoltObserver
Betweeness and closeness centralities are both based on the analysis of shortest paths between two nodes on a graph. The betweeness centrality of a given node takes all the shortests paths that exist between all nodes in the network and count those where the studied node is on the path. The closeness centrality of a node is the multiplicative inverse of the sum of the lengths of the shortest paths between this node and and all other nodes.
On the other hand, eigenvector centrality is a measure of how powerful your direct neighborhood is: your own score is proportionnal to the sum of all the scores of your neighbours (eg, of the peers you share a channel with).
After NFC cards, it didn’t took long for us to be blessed with the first Lightning-enabled ring! It works basically just like the cards, but the NFC tag is hidden inside the ring. Very useful (and beautiful) for when you leave home without a phone or wallet (as the latter increasingly tend to be included in the former).
Regarding security, while it is easier to protect a NFC card (simply by keeping it in a NFC/RFID blocking card holder), one could imagine an equally secure ring where you have to actively press a small button on the ring to close the electrical circuit and allow the NFC tag to be read from.
Spec & Implems
Error Attribution Upon Payment Failure
Joost Jager sent a proposition in the Lightning-Dev mailing list aiming to improve how nodes can pinpoint where a payment failure comes from on the route.
Currently, when a payment fails, the hop where it fails is responsible for writing the appropriate error message and propagate it (from hop to hop) back to the sender. The issue is that the error source (the hop where the failure happened) could simply write random data, and the sender would have no way of telling either what the error was, or who wrote the random data. Indeed, in the current form, only the error source commits to the error message via an HMAC, which means that any node on the route can replace the error message with any gibber they want.
Unable to determine where the error comes from, the sender would then penalize (in terms of internal reputation score) either no one or the whole route, which frustrates the whole point of propagating error messages back to the sender.
The idea is to instead have every hop of the route authenticate the message via an HMAC as well. This way, if the error message is corrupted, the sender can always pinpoint a pair of nodes where the corruption happened and be sure that one of them is responsible.
The issue is that, in order to preserve the privacy guarantees that Lightning’s Onion Routing provides, each node on the route must provide several HMACs, one for each potential position it is sitting in on the route, from 27th hop (the highest possible number of hops) to 1st hop1. Even with some optimizations, this still leads to an error packet of 12KB. However, Jager notes, this should not be so problematic if the frequency of errors do decrease as expected, and this size could be brought down by setting the maximum number of hops in a route to a lower amount.
Entourée de ténèbres aveugles,
Peuplée de moines-bâtisseurs ayant fait voeu de silence,
Et parcourue par la douleur sourde des innocents qui en trouvent la porte
C’est une bien étrange cathédrale.
Elle scintille là où il n’y a pas encore de lumière
Et les photons diffus semblent jaillir de ses vitraux-mêmes
Comme s’ils étaient l’émanation d’une explosion ancienne et singulière
Dont l’onde, en passant, cadancerait les ères.
C’est une chose à voir que ce scintillement!
Prêtes-y un regard, voyageur inconscient
Tandis que tu glisses, sans bruit et sans tourment
Vers le prochain lieu de ton asservissement.
If you enjoyed this recap, you can show your appreciation by throwing a few coins to fanis @ lnmarkets.com, by scanning or clicking the QR code.
Or, to put it another way, as we want hops to ignore where they are on the route, or even how long the route is, for privacy reasons, we have them provide an HMAC for any position they could be in. ↩︎