SN currently has a custodial Lightning wallet because
  1. it was the easiest to make work well fast
  2. given we began dealing in very small micropayments, the trust placed in SN's custodianship was hypothetically small relative to the UX advantage
  3. people new to lightning, and people who are curious but not yet serious about bitcoin, would need custodial solutions before getting them to bother with self-custody
These assumptions have proven generally accurate. However (2) breaks on large micropayments, and I know that for some, the custody of 1 sat is unacceptable. Also (3) is ideally not a Lightning user's terminal state.
I want a Lightning application wallet that serves all Lightning application personas well. I also suspect SN's payment/spending volume is limited by our current custodial requirement.
There are 3 Lightning application personas as I see them:
  1. non-custodial maxi
    • does not want a single sat in someone else's custody under any circumstance
  2. non-custodial pragmatist
    • realizes non-custodial is the ideal, but is ux sensitive/empathetic and is okay with keeping small amounts in custodial accounts
  3. lazy or newb
    • doesn't get it or doesn't care about custodial vs non-custodial and just wants to send/receive/store money with as little effort as possible
(1) and (3) have relatively clear requirements for a wallet's design, but I suspect those personas want things I haven't considered. (2) likely has the most variation in requirements.
Which persona are you? Also, assuming you understand the tradeoffs involved, how would your ideal lightning app wallet work?
1k sats (or more) will go to each thoughtful answer, ie there can be many bounty winners.
1,000 sats paid 16 times
k00b's bounties
I am (2): I know the "dangers" of custodial for my sats, but I know how to mitigate that aspect. If you really know how to manage your custodial accounts, could be a really good decoy to protect your privacy.
I personally consider all those in the (1) category too extreme. They are right in the essence but in practice they know almost nothing how to deal with non-KYC custodial accounts and how to take advantages of these.
As I told you many times: convert SN accounts into lndhub accounts. In that way many SN users could just import their lndhuib SN account into Zeus or Bluewallet app and use it as any other LN wallet.
Please consider doing this: SN lndhub accounts and you do not have to deal anymore with wallet app and all that shit. You just pass the responsibility to the user over his own sats.
Yes, is custodial. So what? Is just for meaningless amount. Who is keeping more than a coffee or a taxi into their custodial account is a total noob and will get what they fucking deserve. Keep into a custodial account EXACTLY how much you want to spend in a certain time. You know better.
reply
I would describe myself as a non-custodial pragmatist. I use custodial wallets mainly for micropayments, e.g. zaps on nostr and also as a backup solution when my nodes don't work for some reason. I am totally fine with the current stacker.news solution. An ideal wallet should be able to switch between different backends (maybe automatically if needed?), like Zeus and getAlby, where you can connect to your own node, but also to custodial solutions.
reply
  1. Obviously im not keeping my full stack in here.
If SN custodian rugs me for 4000 sats, whatever
reply
I honestly could fit on 1, 2 and 3 because depending on the need I'll use different approach. At the same time, I'm not worried about loosing some sats if SN change architecture, or I lose access to it.
Also fully supporting @DarthCoin opinion here 👆👇
I believe that it is important to remember that there's no one solution that fits all... It will be impossible to build and also damn hard to satisfy all these users together.
An ideal wallet will be offering options that satisfy all these needs, so if I'm 3 I just want to access the wallet and do whatever I need to do. Same if I'm 1 and want to fully run on non-custodial, please let me just add connect to my one... and I'm 2, why it should not give me the option to use multiple wallets, aggregate existing ones and multisig with others?
reply
I'm a 2. I keep most my funds on Phoenix wallet now after moving mostly off Muun, although I do still use that for some apps. Occasionally I'll use Wallet of Satoshi, but pretty rarely.
So most things get withdrawn from apps like Zebedee, Fountain, Bitcoin Magazine, microlancer and SN get withdrawn pretty quickly.
I keep a small amount of sats on Stacker News for daily interaction and tipping which I've began doing much more often, and add more if needed. Most of it goes to Phoenix though. Plus I have fun zipping sats around on lightning these last few months.
I'm not amazing technically though so I'm not yet at the point of running my own full Lightning node. Only just barely got a Bitcoin Core node up and my hardware wallet connected. So. I'm sure I'm still making some concessions.
reply
Thank you for the sats @k00b!
reply
I am also a 2. For me, I keep the custodial risk spread among a few apps and keep the amounts small (50-100K). That said, in the future when LN payments become a standard, it will not be possible to keep only small amounts in custodial wallet. You will either have to choose to go non-custodial for most payments or build some trust with a custodial app.
That said, I'd look at your question from a perspective of Stacker News audience and your vision for the platform.
I'd venture a guess that a normal person who would use bitcoin stricly for payment and is not interested in anything else surrounding the technology would not come to SN. And I'd assume this is majority of the population.
So your audience is basically tech savy people with some outliers who are just interested in the idea of bitcoin and want to learn more before they have enough knowledge and stop coming. And of course now with the incentive to stack sats.
If the above assumptions are correct, then most people would have their own non-custodial solution, therefore making a non-custodial wallet would be an overkill.
The lndhub account idea might work, but if you still will have a custodial solution at the same time, the node liquidity maintenance issue still remains. So that does not make your life any easier, I don't think.
Perhaps a better way to go is to improve the way the custodial wallet functions. For example add scrub functionality like in lnbits to allow part of the funds to be transfered to own non-custodial solution.
At the end of the day, it all depends on what direction you want to take the business into. So, I'd look at what is your current end vision and then go from there.
Remember, there is beauty in simplicity.
reply
My ideal lighning wallet would support both Fedimint and LN, but will use mints only for either small amounts or for a short time. If I'm trying to receive a payment that I do not have inbound liquidity for, the wallet would receive it into Fedimint custody, then rent inbound liquidity, wait for the lighning channel opening to settle and then transfer the funds from the Fedimint into the channel. <10 min custody.
My ideal SN would pass Fedimint tips between the users, thus being a custodian only until the receipient logs in and receives their coins. But the present SN works fine, too.
reply
I am person 2.5
As a ex-manager, entrepreneur with mild success, and a software engineer focusing on UI and UX, I feel it's really important to remove friction that users may have, as long as you are comfortable with knowing you are doing the right thing. You do, you don't sell our data, code is open source, etc. I own a node, but the friction was incredibly strong when I just started out.
If I learned something from teaching hundreds of students a difficult skill, it is that you need guide people properly.
People who understand this, try to wear their shoes (and that's how a business can succeed), and slowly try to teach them how they can "upgrade" their shoes so to speak. Instead of simply complaining about being "stupid" or "naive" etc. If you treat your clients like this do you think you will keep them?
BTW I love the SN wallet, I often use it to pay freelancers on the run (just used it to pay 100k sats to someone) and having a channel with SN makes it easy.
Everything is relative, so for someone 10,000 sats is what they consider small. For me, luckily small is bigger, and SN is perfect for that.
I don't think people will use SN for paying/spending because I suspect the biggest use case is through native mobile apps, which seems out of scope? Custodial or not makes little difference. For example, Bitcoin Jungle app is custodial but is used a lot.
I've been to a few meetups and also developed worked freelance work for one of the biggest bitcoin-only news website, and even the founders are not tech-savvy enough.
reply
Agree with @poe7645.
I'm both a 1. and 2.
For most (99%) of my BTC, I use only non-custodial wallets (about half a dozen, including Electrum and HWs) and I strongly encourage people to do the same. But I don't mind, for convenience and tiny daily spending, using more custodial options.
For me, the current SN solution is totally OK. Most SN users will never have significant amounts on SN. And most importantly, the UX is great for newcomers. Just scanning a QR and getting an instant Sats wallet up and running, form which you can withdraw anytime... All this is perfect. Heading to more complex solutions will probably add some friction to onboarding new users to SN.
It's actually a strategic question: which kind of users to you want to attract here? If the answer is "anybody", then don't change anything.
Also, it may be an unpopular opinion, but I think that you should devote more time extending and developing SN rather than trying to "fix" something that works perfectly...
All in all, SN is and will always be an app, not a wallet.
reply
@k00b, just a silly idea that came to mind.
Couldn't someone use SN non-custodially if they provided:
  • An lnaddress where sats could be scrubbed towards when they receive sats in SN.
  • A LNURLwithdraw from which sats could be requested when they want to post/reply/upvote on SN.
Not saying this is trivial or desirable, but would it be possible?
I am personally very happy with SN being custodial. I always have enough sats for one beer, just in case my node is down right when I want to get a beer (been there, done that).
reply
It's something I've considered as a first step. It serves some of the personas in (2) that either manage their own domain/lnurl server or prefer another custodian.
reply
I think it's an interesting idea. Although I'm not sure of how the experience of constant micropayments (<100 sats) back and forth would be. I'm guessing users might take a big dent on fees if their channels have heavy base fees.
reply
I'd consider myself 2 but I don't think it has to be a matter of what's ideal but more a matter of the apps use case / amount of value that use case is regularly moving (what's valuable to one person is not to another, so hard to nail this). If I lose a few thousands sats on SN idc - would I store anything more than $100? Probably not.
I think custodial platforms could do more to educate and nudge users into the self-custodial direction though as it is ideal if you want to store decent sums in bitcoin and likely has some regulatory benefits if you don't want to be considered a custodian down the line.
An example of how this could be done is limits on how much sats can be stored on SN and educate users about using a self-custody app instead when they get close to or hit that limit (would be great if you could tip on SN using a self-custodial app like Breez too!). You could also have a cap on tip amounts, say higher than $50, to receive a tip the receiver can only receive this with a self-custody app connected to SN. This is a bit of a happy medium between 2 and 3 and could keep SN in the lower bounds of custodianship which could have some regulatory benefits (I'm no lawyer though).
reply
I'm a 2. Have my own node with private channels connected over Tor to Zeus mobile app and it works well. I'm doing this to understand the tech, nerdy stuff so to say.
For zapping on nostr, I use ln.tips bot on telegram with alby. Custodial but it's working great and I trust the people operating the infrastructure behind it and they are very open. Anyway, I have no more than maybe max 100k deposited.
I also use Phoenix to be able to receive BTC and swap it to lightning. It's a great mobile app with a slick UX and you can recover from seed.
Other than that, a have wallets on legend.lnbits. Love to experiment with the extensions. Again, more nerdy stuff.
reply
I'm a 2. My LN wallet is Blixt and it's work perfectly, I feel truly sovereign with it. But I have no issues with using the SN custodial wallet because it is fit for its purpose. This forum is so good, this is the lightning app I use the most, I discovered it before NOSTR and this idea of tipping a post or a message at the speed of light is really cool and hooked me right away, it gives a social side to the LN.
Anyway, it's all about the amount on the custodial subject, I only leave a few thousand sats on SN, a few million sats on Blixt and my real bag is on Ledger
reply
A 2 with a handful of wallets for software testing. My current view is that lightning is pocket money, and bitcoin is the bank. If I acquire more lightning than I need for current use, I'll transfer on chain to my Umbrel node for posterity. Having multiple wallets with different features that Sats can easily be switched between offers the best variety of capabilities. Expecting one wallet to do everything is wishful thinking.
reply
I am also persona 2. I was (am?) a complete lightning noob, so having a super quick and easy onboarding experience on SN with custodial was fantastic. It got me into the ecosystem, allowed me to participate with a low barrier for entry. Then I can learn about the tradeoffs and options for (non-)custodial LN wallets.
Generally I keep enough sats in my SN custodial wallet to continue posting/commenting, tipping 21 sats per tip, and move the rest off to Phoenix.
On principle, I prefer non-custodial. Maybe more so for my BTC (in cold storage), but it also applies to my sats in my LN wallet because I extend the same mindset across.
I'm not sure I understand enough of how the implementation would/could work to describe how the ideal wallet would work.
reply
I'm persona 2.
But I'm a big fan of the fact that newbs to Lightning (or bitcoin in general) can sign up to Stacker News with just an email, post some content, and, if the community finds it valuable, get their first sats that way.
I'm becoming convinced that until the next bull run for bitcoin, adoption is going to be driven by content platforms. Zapping and tipping sats is a brilliant use-case for Bitcoin, but it has a chicken/egg problem: creators can't receive sats if their audience doesn't have them. The way SN's wallet works right now solves this by allowing people to join the community (and the economy) without the fairly heavy friction of buying Bitcoin.
It works for SN particularly well because the content is bitcoin-themed. But the internet is full of other message boards and I've been doing a lot of thinking on how to pitch the SN model to these folks, because if we could start people on replacing their likes, upvotes, and karma with sats...boy, oh boy.
(Honestly, NOSTR could totally benefit from the SN model. Imagine if you could tell a normie about NOSTR, get them to download a single client, and start zapping them.)
So, all of this is to say, I hope you don't get rid of the custodial option. It's an invaluable tool for bringing in newbies.
reply
You know me k00b, I prefer no rehypothecation risk. Lightning channels generally ensure this. I'm using blixt and I have a channel with SN, but hardly anything goes through it lol. I do think I'm within the pragmatic spectrum though. rehypothecation risk is mostly risky with applications in which payment is expected to be received and sent out of.
Although nostr's zap model works fine in my opinion and I would like to automatically send to my lightning wallet when I receive something in SN (with no fees thanks to the channel I have directly with SN), I understand that SN should have no incentive to print sats out of thin air and that it shouldn't be doing any loans or anything. Any temptation to earn yeild off of custodied funds should be energy directed towards opening channels and charging routing fees if anything.
If SN came out with its own LSP wallet like phoenix or breez, it would be neat, although ultimately, probably unnecessary lol. I suppose my ideal would be a more general wallet that easily allows the user to point to any Lightning Service Provider with a good user interface for doing so, with next cloud integration to backup those lightning channel states to the users computer (watch tower) and maybe even the CLN method of doing things, pushing those channel states to the various LSPs to be the watch tower for the user.
reply
Definitely I'm a 2, and I support @DarthCoin's request to convert the SN wallet to an lndhub for the unsatwortthy reason that if I don't he might get mad.
reply
Indeed. A lndhub account is the most simple solution for a custodial account. And also is giving the user a bit more liberty to use his sats.
reply
2… well i thought. I dont really care about UX unless its absolutely horrible and borderline unusable.
reply
I think adding something like LNC would be perfect for SN, would allow us self custodial maxis to use it in our prefered way without having SN to have to become a full self custodial wallet.
reply
Definitely a 2. non-custodial pragmatist, obviously I'll not keep only reasonable amounts on a custodial wallet if I need too.
Also support what @mo, the more options provided the better to be able to satisfy as many users personas as possible
reply
Still need a bit more info, I have ZEBEDEE for mobile gaming right now… can’t complain
reply
I'm a 2
reply