#

bosch Jun 16th at 7:24 AM
Thoughts? https://twitter.com/_Bosch_/status/1405026262628798464. @Evan Kaloudis
BoschBosch @_Bosch_
May be an unpopular opinion but lightning wallets should be just for lighting with no on-chain wallet. Channels should be PSBT funded from and closed to external “savings” wallet. #Bitcoin
TwitterTwitter | Jun 16th

40 replies

bosch 21 days ago
This is under the assumption that L1 is primarily for SoV - not payments - and LN is for MoE / payments. Which I think is were we are headed. You wouldn't want to store your stash alongside your day to day payments account and you wouldn't want to use an separate on-chain wallet with your payments for privacy as well as extra fees (savings -> on-chain lightning -> lightning instead of just savings -> lightning)

Paulo Sacramento 21 days ago
Having both LN and On-Chain adds a lot of complexity, but it may be worth it… I dont know…

bosch 21 days ago
A better way to frame is lightning wallets = payment account. on-chain wallets = savings account. Good to not distinguish them as separate networks to end users (adds confusing).

bosch 21 days ago
People just move funds between their accounts like they already do with their fiat bank accounts

Paulo Sacramento 21 days ago
Yeah, I get it. But what are the possible benefits of having both in a single wallet?

bosch 21 days ago
I'm advising against having both in a single wallet

Paulo Sacramento 21 days ago
Have you ever used Muun?

bosch 21 days ago
Yes. Muun is a very abstracted lightning wallet that has a lot of complexities on the backend that may not be viable in the future. I prefer solutions that don't have to make those trade-offs
:+1:
1

Paulo Sacramento 21 days ago
What is your favorite LN + On Chain separate wallets combo?

bosch 21 days ago
Muun still has the added confusion of both on-chain and LN interfaces. Things would be easier imo having just a LN interface in your payments app and just on-chain in your savings app.

bosch 21 days ago
Spectre for my savings / long-term storage and Zeus for LN / payments wallet
:fire:
2

bosch 21 days ago
Funding your 'payments account' from your 'Savings account' will be as as easy as scanning a QR code to fund a LN channel with LND's PSBT functionality

bosch 21 days ago
and you can make it so channels close to an external address (back to your savings account)

bosch 21 days ago
Still a lot of technical hurdles to overcome but I think in the future a similar version of this will become more popular than apps like Muun

Paulo Sacramento 21 days ago
Nice! Where can I learn more about LNDs PSBT?

bosch 21 days ago
https://lightning.engineering/posts/2020-04-30-lnd-v0.10/

bosch 21 days ago
Has since been improvements to this (in beta 0.13 atm)

bosch 21 days ago
For those who may be reading:
image.png
image.png

bosch 21 days ago
Zeus will have this functionality very soon :heart:

Paulo Sacramento 21 days ago
Wait a minute… Does this mean that it will be possible to fund nodes without an On Chain transaction?

bosch 21 days ago
A lightning node, yep!

bosch 21 days ago
Already can it just isn't in any easy to use apps just yet

Paulo Sacramento 21 days ago
Dude…………….

Paulo Sacramento 21 days ago
:mind-blown:

bosch 21 days ago
With Zeus you will be able to fund a channel directly from your external signer (hardware wallet)

bosch 21 days ago
no need to send it to your on-chain Zeus wallet then open a channel - hugeeee

Paulo Sacramento 21 days ago
SO. HUGE.

Johns Beharry 21 days ago
by the way @Paulo Sacramento you are still making an onchain transaction. the transaction just doesnt have to be created in the lightning application

Paulo Sacramento 21 days ago
Could you explain it a little bit more detailed?

bosch 21 days ago
Usually you would make an on-chain payment to send to your on-chain LND wallet. Then another on-chain payment to fund a channel from that LND on-chain wallet. Using PSBTs you can skip that second on-chain transaction and go straight from an external wallet to funding a channel

Paulo Sacramento 21 days ago
Ohhhhhh. Now I got it. Thanks!

Johns Beharry 21 days ago
Sure. So a PSBT as is simply a transaction file. The wallet is the editor of that file. You can perform a few edit actions on the file like add input, add output, sign, and edit various fields.

Once you create and prepare this file, you can share it with others and they can also edit their own versions of it maybe they need to add their own signature to the “contract” (like a multisig spending tx).

Once the transaction file is valid, it can also be broadcasted. Anyone can broadcast a valid transaction. Doesn’t have to be the same device/app/person who created the PSBT.

Johns Beharry 21 days ago
(I see @bosch responded but I’ll still do a braindump :sweat_smile:)

Johns Beharry 21 days ago
So if there is a specialized app or device for creating these transaction files, for instance some hardware wallets. You can now pass this to another app to broadcast.

This could have benefits as @bosch mentioned about separting your on-chain “savings” wallet from your daily spending wallet (lightning).

bosch 21 days ago
Yep! ty for that Johns. I've found newbs using lightning wallets having to send from a bitcoin wallet, to essentially another bitcoin wallet AND then opening a LN channel is a lot of friction. Seperating a lightning wallet and a on-chain wallets conceptually makes more sense imo. Muun almost achieves this though it still has on-chain aspect and if Muun the company disappears the seamlessness of the UX goes away (They do a lot of channel management stuff behind the scenes to make the UX so clean). I'd rather a solution that didn't rely on a company to keep on existing.

Steve Lee 21 days ago
@bosch this is the opposite way I think about it and what LDK is trying to achieve. IMO LN is just a technology that should be hidden from users. It is an enabler to allow many more users have access to bitcoin, and enables instant, low fee, more private payments. Users shouldn’t be marketed a “Lightning wallet” but instead simply a bitcoin wallet.
As for having 2 separate wallets, that is the world we live in today with LND, C-Lightning, and Eclair, because each of them embed a wallet within their LN node software, forcing the issue. LDK is different in that it can easily be integrated into an existing wallet thus not requiring a user have 2 wallets.
Two wallets are bad for users for numerous reasons:
It is already hard enough to secure and backup one wallet. Adding to the burden by requiring users to secure and backup two wallets is a bad UX.
It hurts scaling. We already face a situation where every person in the world cannot have their own UTXO. By promoting a two wallet approach, we are effectively cutting the number of users who can be serviced in half.
What I understand of your goals can easily be achieved with a single wallet. It is possible to have multiple spending policies and velocity limits can be put in place such that the same wallet can be both a secure, savings wallet and a spending wallet.
For example, imagine a 2of3 wallet in which one key is on a hardware wallet securely stored in a safe, one key is on a mobile phone, and the 3rd key is with a trusted 3rd party in the cloud. It is possible to then have a super slick spending wallet on the phone in which the user doesn’t even have ot think about keys, multi-sig, etc. They just open a wallet and tap to pay (with the 2 keys signing being the cloud key and the key on the phone). The cloud key could enforce spending limits such that a compromised phone couldn’t swiftly drain their savings account. If the user wanted to access their entire savings or needed to access their money without the 3rd party’s key, they can use the key on the hardware wallet with their phone. (edited)
:fire:
4

Evan Kaloudis 21 days ago
There are some weird limitations currently with closes on channels you didn't initiate but have the right to funds on but I think we can get there soon. Things are about to get very interesting this year! Stoked about what I'm seeing
:fire:
1

bosch 20 days ago
I agree on hiding LN from users, though I wouldn't market this as a separate "Lightning wallet" but you would have "Payment wallets" and "Vault wallets" - kind of acting like a "Spending account" and "Savings account."
Payment wallets would ideally be mobile apps and vault wallets a desktop app like Nunchuk. Yes these "accounts" could exist in a single wallet, but for security and privacy reasons I wouldn't want my bulk storage too associated with my day to day spending - that's just personal preference though. I need take a deep dive into LDK, I'm not too familiar with it as of yet.
Most people using lightning today use 3 wallets - their external bitcoin storage, their LN on-chain and their LN wallet. The LN on-chain wallet could be replaced with PSBT transfers between the external storage wallet and LN wallet. That's the main friction point I was looking to address here, I'm sure LDK has a much more elegant solution :joy:
Regarding backups with this system, "payment wallets" could auto back-up a SCB to the users cloud account during on-boarding removing that friction (these "wallets" would have no on-chain component hence just a SCB not a recovery phrase). "Vault wallets" would be the only wallet that would need a manual backup.
Regarding scaling with this system, I could see these "payment wallets" as the predominant on-boarding / where majority of users exist. Some people may never leave L2 to a "vault wallet." These "payment wallets" could also enforce some kind of 2of2 signing using a mobile device key and cloud key if extra security is needed / user does not plan on moving BTC to a "vault wallet." Even if you have these two accounts in the same wallet that doesn't really benefit scaling as you are still making the same amount of on-chain payments.
I'm not exactly sure where LN is involved in this setup? Just sounds like a standard co-signing wallet like Blockstream green? Paying directly from your savings account (or even having it visible on the mobile device) could lead to a lot security and privacy issues.

Steve Lee 20 days ago
The mobile app can choose to display whatever it wants. It could choose to only display a “spending balance” which is based on spending velocity. It could require extra effort to show the entire balance or maybe not even offer this at all if there are security concerns. The main point is that the design can accommodate a wide variety of options none of which are worse than having 2 wallets.

Nicolas 20 days ago
balanceofsatoshis from Alex Bosworth is a good cmd line tool to experiment going from cold storage (and even multisig) to a lnd wallet: https://github.com/alexbosworth/balanceofsatoshis#open-many-channels
alexbosworth/balanceofsatoshis
Tool for working with the balance of your satoshis on LND
Stars
157
Language
JavaScript
Added by GitHub