Thank you all for the contributions here ! We are still developing the protocol and the app and we appreciate all the feedback added here.
We have tried to build an extensive library about Fedimint (https://fedimint.org/docs/intro), but we understand this is a large concept that can be difficult to communicate. We’re looking to expand on this and welcome your input via our telegram group (https://t.me/fedimint).
We added below some points that might help to answer some questions you raised. Again, we are super open to learn from you and we might have questions we won’t be able to answer right now, as we are still in the development stage. But they are key for us to build as much bulletproof product as possible. With that said, let’s go to some points to consider:
  • Fedimint is a community custody protocol that allows you to more easily store your bitcoin with the people that you trust, whilst still being able to transact in a private way.
  • This isn’t a new concept, it already happens either with physical money in communities around the globe or in Bitcoin - some of you might have the experience to assist friends or family members with custody.
  • We know the best custody is self custody, but we believe that 2nd party custody with a simple product is better than 3rd party custody. We believe removing bitcoin from exchanges into the hands of communities will be a significant win for bitcoin.
  • We understand this is a first step on the journey of many to understand the importance of custody in general and adopt Bitcoin in a safer manner. That’s why guardians are assumed to be trusted parties in the community otherwise you would be unwise to place your bitcoin with them. So it isn’t like a bank or an exchange, but a collaborative form of custody.
  • A key point with Fedimints is to understand that we are building for a world where there will be thousands to millions of collaborative Fedimints - decentralising the custodians. Each will have different trust models based on your relationship to the guardians of those federations. Larger federations may be useful for more private transacting and smaller federations would be more appropriate for custody of funds, we expect there will be a design space to explore here.
  • Current custody models used in bitcoin provide the custodian with full visibility of the individual's balance and transaction history. We believe one significant improvement which can be made here is to use Chaumian eCash to ensure that the custodian does not have visibility of the balance or transactions of its members.
  • When referencing backup the guardians only hold an encrypted shard of a master key which can be used to get access to funds. The guardians cannot reconstruct this key. They can, however, assist the member of the federation in retrieving this key so that they can decrypt it. Once decrypted they can recover their funds without further assistance from the guardians.
  • Fedimint isn’t Liquid, but has some similarities. One way to think about this would be that Fedimint is a software platform that makes it simpler to deploy many federations as opposed to one federation in the liquid model, decentralising the custodians.
  • Finally it is worth describing the relationship between Fedi and Fedimint. Fedi is not Fedimint, itself but a company committed to building on and educating about Fedimints as a platform. We don’t take any custody of federations funds, these will always remain with the communities supported by a federation.
Hope this clarifies some questions. When we have the app ready we would love to have an AMA, so you can see this concept translated into reality. But as a whole Fedi will be a safe place where individuals will communicate and move their money freely and privately with the ones they trust the most. We are getting there… ;)
reply
This feels like a pivot to possible fractional reserve in the future.
reply
Like if fractional reserve is 100% bad in Bitcoin. Might be 90%.
With regard to LN fractional reserve may be good since it improves security of the liquidity provider which has basically hot wallet.
More on that here Liquidity abstraction in Lightning Network.
reply
what prevents a federation from issuing more tokens relative to reserves? can't all federations become fractional without any transparency?
reply
reply
yes they have a image that depicts fractional reserve. how is anyone ok with this trust model? it is a recipe for people to get completely wrecked
reply
Damn, that makes me lose a lot of faith in this project. Nearly all faith.
Why would it be impossible for user fedi wallets to show their guardian's total amount of issuance outstanding? It's like the fedi devs don't like transparency.
reply
I have all these same questions about audibility too.
It might be possible to do that but I’m sure there’s still an element of trust with displaying a balance of issued notes.
I do like the idea of automated runs though.
Maybe there might develop a market for ‘authority guardians’ or some way for ‘independent third parties’ to be signed up as a guardian (like an independent director on a board). But kinda can’t believe Im writing those words on a bitcoin message board lol!
Listening to Fedi team chat on podcasts it sounds like their pitching it, to start with at least, at local communities so people will personally know their guardians which may be biggest deterrent? Personal reputation risk.
reply
I guess, as it mentions right at the bottom, this issue exists in existing custodial relationships like exchanges 🤔
reply
That also what i'm wondering.
reply
Great podcast episode with Obi Nwosu, co-founder of Fedimint here https://fountain.fm/episode/11174134811
reply
It would be funny if Fedimint provides yet another avenue for shitcoinery. My prediction is that Fedimint software will be used to make at least one shitcoin that is backed by a basket of the top 10 cryptocurrencies. It will be sold as a fully-anonymous stablecoin.
And then the markets will dump even harder and all shitcoin founders will rugpull at the same time. Or, just as likely, the Fedimint-based shitcoin founders will rugpull.
reply
I think they are still doing a bad job at explaining it in a few simple sentences.
Maybe that is because Fedimint shouldn't exist. It's not in the spirit of Bitcoin just like Liquid.
reply
Exactly, because it's just another affinity scamming credit token, a shitcoin that has nothing to do with Bitcoin.
Uncle Jim custody doesn't benefit from a shitcoin gateway, this is a bankers project.
reply
I really want fedi to be a self custody solution, but in the most recent bitcoin magazine podcast about Fedi Obi just confirmed my suspicion. He said that if there is a dispute or someone dies, the guardians who are in charge of the mint can revoke your balance and gain access to your account. He doesn't say it directly but you can read between the lines fairly clearly. Once again, the rule of confusopoly applies. If you can't understand what is going on behind the scenes, even after looking and asking questions, you're probably looking in places and asking questions they don't want everyone knowing the answer to. Obi appears to be avoiding direct questions about Fedi; like what are its limitations, how do you transfer between mints and how private is it really? It seems all this talk about human rights and helping people who are oppressed by authoritarians is just another angle at fundraising akin to "feed my starving children". I prefer that Obi (who I know is occasionally on this platform) be an honest actor and believe he is creating a beneficial technology. However, after my cursory investigation of the, there are so many false promises and conflicts of interest built-in, it amounts to what someone else summarized in a comment here: its no different than a community bank. A small code change could cause the mint to operate on fractional reserves and nobody would be the wiser. If you're using a fedimint you truly are putting your funds in the guardian's custody, and the fedi tokens are nothing better than a banknote IOU. If I were a banker, I would be fully in favor of mass adoption of fedi tokens because it would mean I wouldn't need to change my business model of lending on more funds than I have on deposit and collecting interest on other people's money.
reply
its no different than a community bank.
Imagine running a fractional reserve and the bitcoin price spikes. That community bank would have a huge deficiency (in fiat terms). But this community bank could probably meet withdrawals because with the price rallying, there's probably more people depositing (jumping in) than withdrawing.
But thenwhen there's a bitcoin bear market there's going to be a lot of users cashing out / withdrawing ... it's only then that this community bank's insolvency is discovered (because attempts to withdraw cannot be serviced).
So ya, ... not your keys, not your coins, applies here as well.
reply
I should give credit where credit is due (no pun intended), fedimint does solve the custody problem for small communities wanting to centralize their custody with a group of 'trusted individuals' and it federates the centralized model into atomic communities.
What I would like to see is the tokens themselves having a similar cryptographic relationship between the owner and the mint as do sats on the lightning network between channel peers such that the mint can be punished for misbehaving and likewise the wallet holder for double spending. This simply is not what Fedi offers, however I believe it could evolve into this level of custody. I would also like to know if there are other E-cash token solutions exploring this possibility.
reply
how do you transfer between mints
This image isn't showing in the guide. I'll submit a PR so they fix that.
In the guide, it is described here:
reply
I remember seeing this in some other documentation, but I didn't read the Alice and Bob story before. Basically Alice offers Bob some of her Fedimint-1 tokens, and Bob's Fedimint-2 issues Bob some tokens after it resolves the transaction over the lightning network. I didn't realize it was seamless, they initially made it sound like your Fedimint-1 tokens could only be used on your local mint, and you had to jump through hoops to transact with someone on another mint. I'm glad that's not the case, so thanks for pointing this out.
My point still stands about potential fractional reserve and losing your sats if you become ostracized by the people running your mint or there is collusion between the guardians.
reply
This page in their documentation set describes the tradeoffs and dangers of Fedimint. https://fedimint.org/docs/TradeOffs/Trust-Trade-Offs They call out what I've been talking about. I must give many kudos for this level of transparency.
reply
I wouldn't go so far as calling it a "shitcoin".
But Fedimint sure shows that tools from trustless, low-trust and centralized projects are actually a spectrum instead of a black and white thing. Which is surprising.
I wrote something up about the Bitcoin ecosystem wasting to much time and effort on too many projects a while ago: #63518 This is exactly why should focus on making Lightning rocksolid first.
reply
By definition there is shitcoin and Bitcoin, since it's not the latter it's the former. This isn't even a shade of gray, it's a completely centralized project that works on trust.
You're somewhat right though on the scope of projects "in the space", this is not organic.
Bitcoin is disruptive to a lot of monied interests, and they will fund clandestine operations and projects to slow it down or steer it with a bottomless pit of money.
Liquid/Fedi are the same effective thing, having nothing to do with Bitcoin, and no one having asked for it. Uncle Jim's Pizza Parlor doesn't need a shitcoin layer to bootstrap communit wallets, yet it was bestowed upon us by all the astroturf accounts on Twitter/Podcasts with seemingly unlimited funding... even though no one has plans to use it. Not organic, Bitcoin is under attack as should be expected.
reply
By definition there is shitcoin and Bitcoin, since it's not the latter it's the former. This isn't even a shade of gray, it's a completely centralized project that works on trust.
Not trying to defend something I actually don't support but it is backed by Bitcoin and derives its weak security on multisigs...
You cannot seriously look at Liquid, Fedimint, hosted channels and submarine swaps and not tell me that these projects can be sorted in a spectrum of shitcoinery.
If you personally draw the line between any of these projects and paint one half white and one half black, that's fine. I do as well. But do you actually claim your line is right and everyone else is wrong and therefore it's not a spectum? I think that's not a good faith argument then.
reply
it's a credit token spec, not inherently backed by anything. The affinity scam is in the narrative "it could be backed by Bitcoin". Same with the Liquid "peg".
Hosted Channels / SubSwaps are a different animal, both are services not token specs, and in the case of the swaps are atomically trust-less.
Shitcoin vs. Bitcoin however is binary.
Custody solutions are grayscale to be sure, and for that reason I don't hold any strong opinion on hosted channels. I do lean toward their being a waste of time and unnecessary overhead, but not inherently bad.
reply
What is the issue with liquid please?
reply
I did not say issue but Liquid is basically a Fedimint
reply
Standard Sats against Fedimint Hype
reply
Listening to the podcast with Obi on Foundation here my take is that "Fedi / Fedimint" is a solution to get people off of CEXs who do not trust themselves enough to self-custody.
I am keen to understand more of the arguments against this as there seems to be some sentiment in the comments that Fedi / Fedimint is not in the spirit of Bitcoin, etc. Do we need to get an AMA with Obi perhaps or a definitive list of questions together for him to answer?
reply