Nuh

Working on https://mlkut.org, designer and maintener of https://pkarr.org.

https://nuh.dev

Public Key

npub1jvxvaufrwtwj79s90n79fuxmm9pntk94rd8zwderdvqv4dcclnvs9s7yqz

Profile Code

nprofile1qqsfxrxw7y3h9hf0zczhelz57rdajse4mz63kn38xu3kkqx2kuv0ekgpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqg5waehxw309aex2mrp0yhxgctdw4eju6t0qyxhwumn8ghj7mn0wvhxcmmvqyv8wumn8ghj7un9d3shjtnndehhyapwwdhkx6tpdsq32amnwvaz7tmjv4kxz7fwd4hhxarj9ec82cspzemhxue69uhky6t5vdhkjmn9wgh8xmmrd9skckcrru7

Publishing to

Author Public Key

npub1jvxvaufrwtwj79s90n79fuxmm9pntk94rd8zwderdvqv4dcclnvs9s7yqz

Show more details

Last Notes

2026-07-19 12:44:05 UTC

You should support and celebrate killing zionists
#nevent1q…apm9

2026-07-16 18:50:19 UTC

- reply

Just checked... Literally just a central server https://letterboxd.com/?register=true
So what actually works is crowd sourcing of info, but that is fine, not what I am talking about... It is still a server with API though... You can't do this without a central server, at best you can do the AtProto thing of multiple aggregators and Indexers

2026-07-16 18:43:30 UTC

- reply

I don't know it's architecture but I am willing to bet there is a fat ass server somewhere that does most of the work ... Listen if all you need is the vibe I can do that, but if we are going to discuss architecture it is not going to be any different than a big server with API, just multiple servers with the same API, basically AppViews or whatever you want to call it, but it is just big things that need to make money somehow because it can't be made cheap.. yes. You can donate to it... But it is a big server somewhere and it needs to be compliant and at least sustainable financially.

2026-07-16 18:25:52 UTC

- reply

Except that is worthless... Next time you notice something and you need to see reviews for it, ask yourself if ANYONE in your social graph ever encountered that shit and possibly gave it a review... I won't even talk about the privacy aspect of it and will assume there is a magical ZK solution.
Sure all people leaving reviews might be 3-4 degrees from you, but that is worthless, because you are still going to do a breadth first search, and still end up querying millions of other people's public data repositories.
At some point people need to understand these are problems that can't be solved without centralization, the scalability limits are baked in.

2026-07-16 15:25:46 UTC

- reply

https://testnet.mns.mlkut.org get rate limited names by Blockchain, someone can register in bulk and transfer to users later, every update (other than ownership transfer) is offchain, and you get two extra goodies; 1) shorter IDs 2) key rotation with arbitrary smart contract, so instead of a bunker you can use any logic for someone to help you with custody, and you can change to any level of self custody as you feel more comfortable... Any innovation on key management on any Blockchain can be back ported to MNS.
Now that I mentioned this, Rootstock addresses are derived from secp256k1 so theoretically you can search for the names owned by an npub.

2026-07-16 15:21:03 UTC

- reply

I am taking about something like Celestia, basically the Blockchain (in this case Rootstock) doesn't verify anything, all transactions are pure data carrying, the only difference between OP_RETURN or Eth Blobs and what Celestia does, is that in Celestia you can add something equivalent to Nostr "kind"... This way people can query only transactions they are interested in.
A simpler way to do this on Rootstock is just contracts and associated Events they emit.
So let's say you want to make a Monero on Rootstock, you create a new contract, every time you make a transaction to that contract, it is only data carrying, thus miners don't validate anything except transaction fees.
On the client side, you can see all these data, and you can run them as transactions against the smart contract state, deriving the new state, at the client side.
But then when you want to take money outside of that contract to move it to another contract that says mimics ZCash instead, or if you have multiple Monero contracts each contain a shard, to keep things manageable on client side validation, then when you want to exit from one shard to the other, you need to provide a ZK proof to Rootstock that you did all the computation over that data carrying committed to that contract, and that the state after that computation contains a withdrawal request, not from one sidechain to another but from one contract to another...
Does that make sense?
This basically allows you to do 2/3 of what Drivechains does;
1. Freedom of innovation.
2. Sharding to limit the cost of validation
As for #3; DA scaling, we keep it on Rootstock with Blobs and data availability sampling, not only because we can, but because the synchronicity between all these shards is super valuable for interop and atomic swaps, because you remove a major risk in cross chain transactions which is that some chains may reorg independently from others... But if you do DA on the same chain, then reorg is atomic, every shard is reorged at the same time. Also you don't have to keep bootstrapping merge mining (a new BMM has higher risk of reorgs, but in DAS, there is no reorg independently from Rootstock).

2026-07-16 13:31:31 UTC

- reply

The flaw is in the way it is suggested in Drivechains... Merge mined Sidechains can also be hashrate escrow (and indeed Rootstock will be one with bitvm) if you use OPCAT to verify an SPV client, which means whatever the sidechain miners vote on goes.
The reason Drivechains hashrate escrow is worse is because it forces miners to interact with even starting a sidechain.
That being said, I think people shouldn't talk against eCash until they commit to using Rootstock, otherwise they are not being serious

2026-07-16 13:21:19 UTC

- reply

I guess you can call that hashrate escrow... But a merge mined sidechain with BitVM bridge is very much a hashrate escrow already.
The difference here is that if you don't want faster blocks than Rootstock, you don't create Sidechains and worry about the consequences of that, instead you just accept transactions as they come without any computations, and update the state later... Separating DA from Compute.
The obvious upside is synchronicity which makes the entire system safer and easier to reason about. Things all happen at the same time including their side effects in parallel contexts, and as long as majority of miners are honest (or ZK verifier is valid) you reach the same conclusion globally as you calculated locally.

2026-07-16 13:16:42 UTC

- reply

Note also; it doesn't Have to be ZK proofs, you can still demand that miners do the computing and finalize the state, but you can demand that with latency, so they get to do that in the background and take their time, not while accepting txn to put in the block... Then they can vote on the state commitment of previous epoch.
The downside of that is that you don't get to invent custom vms and don't save miners the effort. The upside is that you trust miners majority, instead of math and possibly buggy code.
I don't know which is better, but I wanted to say this is possible

2026-07-16 13:12:40 UTC

- reply

I mean it depends on what are you trying to achieve, humans seem to be ok with few seconds latency for anything except very hectic metro lines, in which case you are better off with prepaid cards. But for everything else, if Rootstock has 15s blocks, I believe that is enough, and requires no centralized block producers, just publish your transaction, and IF you want to move from one context to another then you have to generate the ZK proof, but that is just the cost of business, you can obviously atomic swap with someone who generates these proofs and provide the liquidity... But note all of that is no different from paying a miner to include a transaction, that is to say it is so open and so competitive you can hardly do any better.
Now when you need instant payments, as in subsecond confirmations, then your options are;
1. Proof of stake sidechains (Stakechain) on top of Rootstock, which is not actually a single company. But while one confirmation is super fast, the more decentralised these get the longer it takes for finality ... But if you want such fast txns, you might be more happy to risk losing some with reorgs. But you still get more confidence than zeroconf.
2. Centralized payment pools, this one gives you the fastest finality possible, but you have the risk of withholding proofs that are necessary to withdraw or the business going down.. here I believe the absolute best you can do is just use small amounts that you are willing to lose, just like your pocket money, and maybe use many providers at once... That basically boils down to something like Ark, but with more programmability of the contract with the server.
So in conclusion, I think we can scale and indeed solve DA, AND have synchronicity which makes interop between contexts much safer than sidechains that go out of sync, but as soon as you require faster confirmations and faster finality, you must start accepting tradeoffs for different amounts ... But worst case scenario (waiting until Rootstock can't be reorged practically) is still orders of magnitudes faster than finality in any fiat system that merchants are used to, hell even cash notes are only finalised when you take them to a bank and see if they are counterfeit or not.
You don't think that is insanely good outcome? I personally played enough with Rootstock and I think their latency is acceptable, and I don't think that LN or Ark give me much better latency for payments I need to be careful about anyways, and for payments that I don't need to be careful about, I am ok just risking it with a temporary cosigner just like I am ok losing the money in my pocket once or twice in my lifetime

2026-07-16 08:55:10 UTC

- reply

It depends on the definition of scaling, computational scaling is limited and favours sharding of contexts then maybe use zk proofs to settle between contexts. But data availability isn't doomed, it is actually possible to scale it with users, as long as the number of users is actually objective, which is the real challenge. If you KNOW that there are 10X more users now than last year, then you can tolerate 10X more data in a data availability sampling paradigm. Which is why Celestia needs to hardcode the number of clients the community believes there are around the world.
But I wonder if someone can use Proof Of Work as a proxy of how many users there are watching the chain and sampling chunks of the data.
I don't mean necessarily decentralised mining, but just using the hashrate as a proxy for demand, and demand as a proxy for the light clients taking data availability samples...
This is not something you can do in celestia without oracles of the token price, but we have hashrate and they don't.

2026-07-11 20:10:18 UTC

- reply

I have no idea why do people refuse to use Rootstock so hard :) it does all the proof of publication, and it can actually increase the block size gracefully and you can make a large blocks sidechain to it with PoS if you must etc...
Anyways MNS is more scalable naturally, because you names are assigned randomly, no need for auctions, and you can register a whole batch at once, then assign names to people offchain with trust, and transfer the names onchain when the fee market is best.

2026-07-11 12:33:54 UTC

- reply

I am not interested in names that needs rent, I also am not interested in Blockchains that don't use BTC as a unit even if with a bridge that isn't perfectly secure.
For both these reasons I am building https://github.com/mlkut/mns so I can use Rootstock but also get names that are not rented, but owned forever without name squatting and phishing.

2026-07-09 09:02:07 UTC

- reply

I am more comfortable and happy with DeltaChat, but I love that guys like you are doing this work and trying these ideas.

2026-07-08 15:05:10 UTC

- reply

Demand is always measured by the literal payment for the privilege of using the system. As I explained the daily transactions fees for Monero is 100x less than that of Ethereum, so if the whole argument for Monero is that it is going to bring more demand for bitcoin miners, then focusing on Rootstock reaching as much as possible of the Ethereum addressable market makes around 100x more sense from ROI pov.
Also not that the obnoxious Monero folks aren't actually going to stop nagging once there is Monero on Rootstock, because they could have build that already, no they demand that Bitcoin L1 itself adds this massive complexity, despite Bitcoin refusing even OPCAT, for the promise of the negligible extra daily fees... It is absurd

2026-07-07 21:25:34 UTC

- reply

I will grant you that maybe if Bitcoin added FCMP or whatever we get more fees because people trust Bitcoin more than Monero ... But then we are in the hypothetical world, and I can hypothetically claim that Bitcoin is trusted because it is simpler and transparent and easier to understand and conservative etc, and that making any more changes than OPCAT might really change that trust.
Either way, if Monero was so great and Bitcoiners are missing out on great demand, at the very least we need to see some of that demand on Monero itself... No?

2026-07-07 21:22:01 UTC

- reply

If I am not wrong, if you take 20% of miners rewards, and you take ALL Monero transactions and transactions fees (not block rewards), then you make $20k a month, so not bad if you are achieving that entirely alone, but definitely isn't worth it if you are a medium size team, let alone if you take VC funding... Sorry, Monero folks aren't paying any amount of fees to make people horny.
Ethereum makes 100x that in fees ... And Rootstock is trying to attract that.
So imagine putting some of that decades long effort that Rootstock put, for 1% of the potential.
The same applies to telling Bitcoiners to add FCMP or whatever... The promise is tinnnny for insane complexity, and yet people are happily making their demands and not at all addressing this imbalance of risk reward

2026-07-07 21:11:25 UTC

- reply

Transaction volumes are irrelevant, the question is how much fees do Monero miners collect, because if I am a developer and I plan to take say 20% of that income, I need that 20% to be worth it, especially when I subtract all the resources I will pour into building, auditing, marketing etc..
If you think this is worth it, maybe you should build it, not for NGU, but to get that 20%

2026-07-06 17:03:27 UTC

- reply

No it is not in anyone's interest to spend years doing unpaid work because someone promises that this is the one thing that will significantly increase the demand... People simply don't believe you. So it is on you to build the thing and bring the demand, and if you believe in it take developers rewards from the fees if you want. But you don't actually believe in the thing.
Some people believe in Ark so they build Ark, some believe in Rollups with privacy so they build it on Citrea or whatever, some think there is a demand for tornado cash, so they port it on Rootstock.
Evidently no one believes that Monero on Rootstock is worth their time, not even Monero people who can't shut up about how is Bitcoin going to die because it is not fungible... Ok.

2026-07-06 11:30:58 UTC

- reply

Just utter bullshit... Just make the damn thing on a sidechain and make it impossible to "use wrong"... When people exit the anonymity set, that is indistinguishable from depositing Monero on an exchange or wrapped Monero... When they get back to the private rollup, it is indistinguishable from going back to Monero from exchanges.
You guys just want to be sad for fun apparently

2026-07-06 11:28:49 UTC

- reply

All three of you are lying because none of you are actually doing any work towards that goal.
It is not Impossible to use Monero wrong, you can use Monero in Custodial services for scale, cheap transactions, safe custody and other reasons, so it is not impossible to have the ratio of Monero siphoned out of the anonymity set approach the ratio of BTC to private RBTC on Rootstock if someone builds that...
You all are asserting a nonsensical notion that you can have privacy without caveats, and that Monero is that promised land... That is bullshit, the economy will always have private and auditable moneys and the absolute best you can wish for is:
1. Large enough anonymity set.
2. Option to anonymize your coins when necessary
3. Bonus; merchants accept the private money directly without forcing you to exit the private pools
But #3 is not that important.
Moreover, almost ANY scalability solutions ends up adding to privacy because by definition you can't have perfect data availability at scale, so things like privacy pools, or private Rollups, or eCash mints with insurance smart contracts or just a proof of stake sidechains that only shares data between anonymous validators, or Intmax2 and other client side validated solutions... It all makes auditability and trackability much harder.
All I see here is people too lazy to build things on Bitcoin sidechains, but also are not satisfied with Monero to shut up! Like what are we doing here? Is this therapy? Write some code and bootstrap the thing the hard way like we all have to.

2026-07-06 11:19:30 UTC

- reply

That's not my point, because the same can be said about going through Tornadocash... My point is that everyonr that goes to exchanges and stays there (most people prefer custodial shit if they can afford it) then the anonymity set of Monero gets smaller and smaller and smaller. So my original point stands; a private subset of Bitcoin is the same as Monero with non-private set of Monero. Either way you have private and non private parts of the supply, so saying that private RBTC on Rootstock is inferior to Monero makes no sense.

2026-07-05 13:54:21 UTC

- reply

Who do you think will win? Tornado cash and the like, where people get paid to insert their money to increase the anonymity set? Or Monero folks begging for attention every day demanding that we all increase their anonymity set for free?
We watched millions of stolen USDT repeatedly get laundered through tornado cash, and no one can do anything about it... And they didn't bitch and whine, it just works.

2026-07-05 13:50:35 UTC

- reply

Dude it is ALLL opt in privacy what the fuck are you on about? If you use Monero you opted in and the entire universe keeps using banks, in fact even Monero is used transparently on CEXs... Man you are so childish.
Can't imagine never being happy until the entire world behave as I demand, what a miserable situation

2026-07-05 13:48:45 UTC

- reply

Think all you want, it is stupid claim and baseless too, because there is no magical difference between private RBTC on Rootstock, and private Monero on mainchain... The "non-private" L1 makes no difference whatsoever, any more than Monero on CEXs reduces Monero fungibility. If you stay on Rootstock who gives a fuck?
Again you just want to complain and get attention from people who are happily using Bitcoin L1, either your way or the highway .... Fine just sell your BTC because it will never ever ever have a change so dramatic, forget it.

2026-07-05 13:45:08 UTC

- reply

You are delusional... No one is insisting on shit, Bitcoin exists and it exists as is and it is impossible to change at this point as far as I can tell, I want OPCAT I won't get it, I move on.
Your problem is way simpler because you don't even mind having a separate token, fine go use it or use sidechains or whatever... But bitcoin simply won't change, it is not me that I am insisting on anything for whatever conspiracy you have.
Just deal with reality and stop being a baby. You want to build Monero on Rootstock? I will celebrate you building that. You don't think that is enough and you want to use Monero? I encourage that too. But bitching non stop for something that won't happen is simply procrastination and it is pathetic

2026-07-05 13:39:52 UTC

- reply

No he wouldn't because he would rather complain ... Anyone could build it right now, for context people ported tornado cash to Rootstock and it works just fine as far as I can tell. There is simply not enough demand for people to stop complaining and actually build the thing they claim to want.
I want a dame registery on a PoW sidechain, I am building one. Simple as that.

2026-07-05 12:12:46 UTC

- reply

I enjoy watching you calling everyone social justice whatever while complaining that the universe is not behaving as you think you are entitled to. My favourite account on Nostr for sure.

2026-07-05 11:17:20 UTC

- reply

A reasonable person would have realized that there is no difference between anonymity set in Monero and anonymity set in the sidechain, except that Monero doesn't pay the same miners and doesn't get as much hash rate.
But you prefer to complain about other people's choices. Which is why you won't sell your BTC, and even if you did you will still complain about other people's choices. Because from multiple conversations with you, on multiple topics, you will never be satisfied as long as human beings have the right to disagree with you and ignore you without any violence against you, you somehow perceive people choices as an injustice to you.
Either associate with Monero, or recognise that a private sidechain is just as good as a private mainchain, as long as it is used by enough people and has enough liquidity, or reject all that and continue complaining, that last one is easiest for you evidently.

2026-07-05 02:44:09 UTC

- reply

Fungible isn't an all or nothing, again you can have a subset of the supply private and the rest public, just like you can have Monero moved to exchanges and public sidechains or wrapped tokens that are auditable and public.
So unless you want everyone on earth to make the same choices as you do, with their own money, their will never be 100% of the supply in perfect privacy that satisfies you
Anyways, you should sell all your bitcoin.

2026-07-05 00:50:19 UTC

- reply

Says who? Who says that I need fungibility on L1? If the bridge to Rootstock is made as secure as humanly possible without soft forks (with BitVM), then I have most of my money in RBTC on Rootstock, inside a perfectly private rollup or yet another sidechain to Rootstock... How is that not good enough for me?
No matter what you do you can have some coins that aren't fungible, people can literally move Monero to a centralized exchange or a sidechain with wrapped tokens and make it super public ... So what are you complaining about, that not every single human being obey your dictates?
Even if you don't trust BitVM bridge, just make a new token on Rootstock, make if super private, and monetize it from scratch... I simply don't understand what is the point of Monero as a separate Blockchain that doesn't pay fees to the same miners .. other than historical doubts about merge mining maybe.

2026-07-04 20:04:09 UTC

- reply

Maximalism is the only thing that makes sense; one coin + merge mining [1] means more miner profitability for building on the same chain, which in turn means more security and finality for any transaction which makes the whole system more useful.
[1] in fact merge mining is more important here than single token, but once you have merge mining you might as well try as hard as you can to bridge the same token, but honestly even if Monero token migrated to Rootstock and ignored the RBTC, that would STILL be better than diluting the Work, let alone all the ecosystem compatibility and yes the NGU.

2026-07-03 09:05:45 UTC

- reply

Nonsense they don't have a business bone in their body they are a bunch of hackers. You still can use normal email server if you want, users just don't want to, the setup with legacy email servers like Gmail was gnarly, Chatmail relays are like Nostr relays or Simplex relays in that they don't store your data for long and reject unencrypted mail, which makes it very cheap to give you an account without much onboarding, that is the UX tradeoff. But you still are using email protocol, so if you want to create your email server and have perfect retention, you are welcome to do so.
In fact there are advance flow in Deltachat to use a legacy email server, it is just not fun, and for the very specific use case of ephemeral chat, the upside is not enough so no one bothers with it.

2026-07-02 10:21:02 UTC

- reply

Maybe I am mistaken though... I hope this works at scale.

2026-07-02 10:18:28 UTC

I am a huge fan of Cody, and I don't hate being the most popular Nuh here, but if you want to see why WoT and reputation systems are not really good solutions for non-random name, search for Google in this reputation system app and tell me if the result make any sense. Hell, the fake Elon Musk accounts have more reputation than myself!
#nevent1q…zljs

2026-06-28 15:14:29 UTC

He is saying that while hitting US bases again, or whatever is left at least.
#nevent1q…nj9m

2026-06-28 15:04:43 UTC

- reply

Email isn't only successful just because network effect. In fact if it wasn't really good already, it would have been replaced at least partially by phone numbers, which is not only earlier than emails, but also more prominent in chat apps which are increasingly more prominent in commerce and many other access context where you join the thing just by joining a group chat.
I am singing the praise of reliable and simple messaging, whether that is reliable messaging of a server with TCP/HTTP or a more intermittent entity like the case in Email, and indeed SMS or other chat apps.
Email is indeed benefiting from network effect, but it is also not the first solution, it was just the first good enough solution. And indeed any upgrades that I find worthwhile for Email, happen entirely at the DNS level, not the email format or protocol.. that shit is really truly good enough...
Hell I am using a Signal alternative that is entirely on Email protocol and it is working like a charm, and it is not even using Email as a thing you already have, it creates whole new account in the background... It just uses SMTP because SMTP is simply good enough

2026-06-28 12:43:43 UTC

Email remains undefeated even in the niche of niches that wanted to replace it.
#nevent1q…kd4w

2026-06-28 11:38:51 UTC

- reply

No, OP_CAT would let sidechains start permissionlessly with blind merge mining and optimistic bridge. But even string concatenation is too hard to do on Bitcoin, which is insane given that everyone wants post quantum now, which is orders of magnitudes harder.

2026-06-27 07:39:08 UTC

Play censorship resistant broadcast games win censorship resistant broadcast prizes.
#nevent1q…474y

2026-06-26 22:05:37 UTC

- reply

Filtering by tags is worthless for anything but broadcasting, but we established that even Nostr apps are mostly going in direct mailing direction with communities and the like.
To that extent Chatmail relays are doing way better job than Nostr, probably more relays too.

2026-06-26 21:14:54 UTC

- reply

You are under the impression that only Nostr relays do that. That is false, and they aren't even the best at it. But I am also not interested much to prove this so use whatever is working for you

2026-06-26 19:51:21 UTC

- reply

1. Nostr doesn't magically solve anything here it just abandons the very minimal expectations that your data doesn't get deleted.
2. You can and people do run their own SMTP servers
3. Many people pay for paid email especially privacy orientated
4. Many competitors offer 15GB for free, to attract users that upgrade to premium.
Email is the most competitive service out there in the modern age... And it can only get more competitive if you add sovereign domains so you an exit from one service to another without losing your id.

2026-06-26 17:21:14 UTC

Servers, protocols, APis, it will all come and go, identity is the only thing that we can salvage and maintain our if that ship of theseus, which is why it breaks my heart when people bake identity in the transports and protocols, when the very internet proves that Name systems are wholly separate layer, in fact it is the most durable and orthogonal layer ever, it will probably outlast even IP if we ever made something that overcomes IP.
Let's solve identity, at protocol and UX level, then figure out the rest as we go.
If you really need broadcasting, well Nostr is here, broadcast if you must, if you need reliable messaging Email is still here mail as you wish... But don't make Email accounts identities, and don't make Nostr keys identities either.

2026-06-26 17:15:15 UTC

- reply

If I know that I only ever want to do small communities, and private communities, or at least communities that don't care about broadcasting as widely as possible, I wouldn't care about Nostr at all, I would just focus on adding sovereign domains, ones that are easy to provision to users, to what we already have; emails, mailing lists, and some work on Drives, encrypted or just WebDav/S3-like API.
That is basically using Email to do what Nostr relays do, but the difference is that you don't add anything more than you absolutely must; usuable sovereign domains. The rest is good enough.
You can argue that you can repurpose Nostr relays to do what Email servers already can do, but then the same applies to S3 providers, and it is not clear why would I bet on Nostr relays and query language, Vs betting on S3 API and SDKs or Emails and their countless tools.
As if what I would do if I absolutely insist on using Nostr, then I would probably go all in on inboxes and outboxes and relays with serious data retention and storage quotas and access control and the whole thing, until they are indistinguishable from Email... Except for all the redundant effort for no reason.

2026-06-25 08:22:27 UTC

Nostr pivoting to forums and group chats, Bluesky pivoting to Communities, and nature is healing, but all the technical debt and bad choices made to support a Twitter clone remains, spoiling everything for no reason.
#nevent1q…96ek

2026-06-21 14:44:35 UTC

jewish supremacists kill children to please their god.
https://npub1jvxvaufrwtwj79s90n79fuxmm9pntk94rd8zwderdvqv4dcclnvs9s7yqz.blossom.band/f562db6371e1536a38353af20363184d3a642784a005206fd0336104d9a28080.jpg

2026-06-15 14:59:15 UTC

Iroh is finally 1.0 give them plenty of love, the best project in the space and definitely the best team https://iroh.computer/blog/v1

2026-06-11 03:07:37 UTC

- reply

Ok but whatever client you are using for Nostr can be made to do the same things, and Matrix can have clients on different apps stores than regulated ones. I think people just give their favourite protocols too much grace. Fundamentally you can censor Nostr just as effectively, and sacrificing consistency and single source of truth doesn't make you immune to any form of censorship, it just degrades everything.

2026-06-09 17:10:14 UTC

Delta Chat puts all Nostr DMs to absolute shame!

2026-06-09 16:27:23 UTC

You should support and celebrate killing zionists
#nevent1q…sut8

2026-06-09 16:11:15 UTC

- reply

Not true, I come from a tradition of P2P networks and there is zero evidence that a P2P network is more censorship resistant than a good old http servers + agile DNS records that keep changing whenever an IP is black listed.
If anything that is how most successful and useful services remain step ahead of censorship.
Let alone that if Jumble stopped checking npub signatures you wouldn't even care or notice.
But I am. Not going to bother selling anything to you, it is just as futile as you selling to Nostr.

2026-06-09 15:51:09 UTC

- reply

Exactly that is why it makes zero sense to bother with Nostr, if you are going to make a proper solution make a proper solution, key rotation and all, with strong consistency as well.
So not going all the way, when the half measures are just as unacceptable in Nostr as the full measures is a waste of time.

2026-06-09 14:27:50 UTC

- reply

The very funny thing is neither of you will reach anywhere because Nostr fundamentally doesn't care or want to bother with key delegation. So while you look down at many solutions as overly complicated, Nostr looks down at any delegation whatsoever as overly complicated... Hell Nostr refuse to bother with consistency even so good luck with knowing what is the current state of delegation

2026-06-09 14:25:43 UTC

- reply

Ynniv isn't thinking about how to keep old events signed by old delegated keys, without allowing new events signed after blacklisting these keys.
So either he didn't think it through or he is of the opinion that you should nuke all old events and resign them with new keys that are in use. Which is what I mentioned before about how Bluesky for example has a merkle tree over all files, and if you rotate your keys all old signed files are discarded and apps need to go check them again on the latest version of your PDS.

2026-06-08 12:49:11 UTC

- reply

Yes but you would also get more traction on Telegram groups or WhatsApp groups or Signals or Matrix groups or literally a mailing thread with a lot of CCs, which wouldn't require relays or any large infrastructure that are easy target to lawfare

2026-06-08 12:47:27 UTC

- reply

But that at best turns Nostr to Email, and that is the best scenario if people actually embrace message passing paradigm. If they don't and insist on the shared Heap paradigm and just querying few relays, then your relay is just shouting to the void not unlike having a blog... In which case you are better off with an actual blog.

2026-06-08 12:45:39 UTC

- reply

Yes but then you are joining a forum, maybe large one, but not that large even, so freedom of reach cannot possibly be the selling feature, because you are not reaching anyone other than the few people in this forum.
I just want people to chill, you can't have censorship resistance at large scale except with architecture similar to Email, even mailing lists have moderation.
People need to embrace small scale and direct messaging and stop expecting relays to be capable of protecting arbitrary content forever, they can't.

2026-06-08 12:06:41 UTC

- reply

Leaving Twitter to Nostr is the perfect shadow banning is what I meant. You don't get more reach here than in Twitter, even if you are the most popular Nostr account, there aren't enough users. This applies to even Bluesky and they have 40 million users and a million daily active users.
All social media approaches just a big forum, going to a very niche forum reduces your reach even if there is no moderation at all in that forum.

2026-06-08 11:57:25 UTC

Unpopular opinion there is no more powerful shadow banning than Nostr. I am not here for censorship resistance I am here to talk to interesting silly people who care about open web protocols for silly reasons... And in fact as soon as I get their attention I prefer talking to them in proper chat apps.
This is just another place to signal my interests and find similar minded people, it doesn't take much censorship resistance really.
#nevent1q…zegl

2026-06-08 11:49:43 UTC

- reply

It is actually still going strong in Peergos

2026-06-08 08:15:54 UTC

- reply

It is easy to say you don't need key rotation... I did for long time.. I want people that make that claim try to provision keys to even experienced developers and see how often they lose it when they don't really care about it enough initially.

2026-06-07 18:09:06 UTC

- reply

Actually I don't mind if you onboard users to MNS, I would much rather though if he pull the rug from under USD backed stablecoins and redirect that insane demand to a more geopolitically neutral basket of commodities and make it pay fees to Bitcoin mimers

2026-06-07 18:06:39 UTC

- reply

Maybe it is more productive to convince Jack to make it more open protocols friendly. Would be awesome if he gave all users an MNS names whether they want it or not. I also wish he creates a Bitcoin backed stablecoin that is pegged to a basket of currencies or PoW or the price of Bitcoin in electricity around the world etc... and make that on Rootstock + statechains or other scaling solution.
Open protocols are usually not the hard part, it is the product market fit and user base.

2026-06-06 16:57:04 UTC

- reply

Except his definition of filesystem is quite generous. AtProto doesn't have filesystem structure, instead it has collections and these collections are made of items all supposed to follow certain schema which is in the name of the collection. This is app friendly and let servers optionally enforce schemas... But that is not a filesystem at all. Of course it also has no access control over that, so even if it was filesystem it would be missing access control semantics for such filesystem.

2026-06-06 15:17:37 UTC

I don't like identity politics, but I do like Blacksky because it is very much bullish decentralisation, it tells us that not only Bluesky failed to gain more than 1M daily users, it doesn't even scale social trust to all these users. So you already see "community" indexers... This is great, this proves Mastodon and ActivityPub architecture optimal, and that architecture is Actually decentralisable.
#nevent1q…nwpx

2026-06-06 14:57:24 UTC

- reply

To be honest stamps never made any sense to me... Don't Take Steps Backwards. I am not going to pay for things I did for free since I had an internet connection.

2026-06-06 13:51:36 UTC

- reply

Except it never worked neither did any Nostr DMs... The only reason I use Keychat is exactly to browse Nostr apps.

2026-06-06 12:15:12 UTC

Experimenting with a new way to visualise MNS names as default avatars, this time inspired from Lifehash https://gist.githack.com/Nuhvi/15105f2ce378d1aa1b2d2807824b40b5/raw/mns-visualize-lifehash.html

2026-06-04 17:30:05 UTC

- reply

Sidechains inherently checkpoint their history (discard it) whenever a withdrawal to the mainchain occur, otherwise they would be at risk of diluting the peg and the whole thing goes belly up. So they must make the withdrawal slow enough to be safe, once safe no point of keeping any old blocks really.
Rootstock is doing some work to enable that and I am voicing demand for it.
In that case you can have very light clients for Rootstock even if Rootstock had large blocks, with the only bottleneck being the ability of nodes to verify all these transactions, but even that has lots of work done to improve it, including transactions that are only data carrying and require little compute and Ethereum work on making zk proofs so fast that you can produce them in real time so validating nodes are only bottlenecked by data bandwidth, and then you can get more scaling with sharding data and sampling them with erasure codes etc..
But even if all these optimizations sound too complex and even if Rootstock can't scale beyond 2X Bitcoin TPS, it has very expressive VM, so you can make sidechains of Rootstock much easier than Rootstock itself, giving you fractal and shared scaling that can scale much easier than you can actually get normies to care, which is the real bottleneck.
Even the 2wp in Rootstock is going to become better in the coming years with BitVM reducing the honesty assumption to 1-of-N.
At least this is my attempt to encourage any and all serious builders to take Rootstock seriously because I don't believe we will have any better option, and we don't actually have the luxury of waiting a decade or two until figure out how to make Bitcoin usuable at scale.

2026-06-04 10:54:17 UTC

- reply

No I mean what people call AppView or Backend For Frontend, or the Indexer in the case of Pubky... Servers that aggregate data, map it to other forms, and index it based on many aspects and prepare the database indexes and API endpoints to return very specific data for very specific client needs.
Do you really enjoy your Nostr clients counting reactions in real time as they stream? Every time you open the note? Do you enjoy that you can't get consistent and fast counts and statistics? Do you enjoy that you can't get reliable notifications count and reliable cross device count down of notifications count as you open more and more of them?
All of these stuff require a server that isn't necessarily dealing in the raw data, but metadata. And metadata that you can't design at the protocol level, every app will need different stuff.
It is a shame that Bluesky knew all of this from the get go, and even Mastodon does a good job there, but here people are pretending the laws of the universe don't apply, because we are cool.

2026-06-04 10:46:59 UTC

- reply

The general idea that suddenly the world won't need well crafted UIs and APIs is retarded. People had 3rd printers and before that power tools, yet they didn't suddenly started creating furniture on demand on a whim. This disregard for craft is annoying. Also isn't Nostr a big ass experiment in how useless and worthless quickly developed and thoughtless apps are?

2026-06-04 10:07:58 UTC

People are going to have to learn the hard way that UI is designed on the backend first, unless you don't care about latency or bandwidth or IF you download all the data first and your AI has a local indexer.
Please talk to people who worked on any functional application and see how much back and forth goes into optimizing the API to fit the UI as much as possible without over fitting so as to keep the API as useful as possible to as many UI needs as possible.
AI agents can make your local data maximally useful, but it can't magically make usuable UIs to remote data that isn't already indexed with the UI in mind.
#nevent1q…55hj

2026-06-04 05:20:42 UTC

Goyslaves are pathetic man, how much more power and wealth and destructive capacity do they need before growing some dignity?
#nevent1q…2phs

2026-06-03 12:00:01 UTC

I am 100% on the valuations skeptics side, but I stopped betting against hype long time ago, the economy is fundamentally fake and has been for some time, there are no tethers to reality, no gravity, no truth, all a belief system. So don't short anything, just buy more of what you need while you have liquidity.
#nevent1q…26yp

2026-06-03 07:39:07 UTC

I try not to get emotionally invested in any of the protocol debates, but I really hate it when people look at one of the miracles of the internet and decide to be retarded and ungrateful instead
So here is a crash course for the 1000 times.
1. A DHT is a self organising sharding mechanism, made of very cheap to run servers, with expectations of high churn.
2. Any client behind a firewall is NOT a part of the DHT, so any Bittorrent client that doesn't successfully open ports automatically, doesn't count in the millions of routers in the DHT.
3. The numbers of nodes in the DHT are not fake, there are no official numbers, you calculate these numbers statistically by observing your own node's routing table, each node in that table is a node you literally contacted yourself, noted it's ip address.
4. Most implementations, certainly mine, enforces a limit of nodes per region and rejects multiple nodes from the same IP etc. So the mossad would have to own shit tons of computers in every country on earth to fake these numbers, .. just to what? To know that you are downloading pirated movies? Be real.
5. Permissionless DHTs are theoretically vulnerable to Sybil, Yes, 100%, but Bitcoin is vulnerable to a lot of shit in theory, in practice however quantity has a quality of its own, and when you have millions of organic nodes, it becomes very expensive to attack specific range of the DHT, and even then that attack will be very detectable and then clients can react by querying more nodes further and further from the target.
6. Attacking the entire DHT to make a specific attack undetectable would require running 20X more fake nodes than the honest nodes.
7. Nodes are soooo cheap to run, they have an LRU cache, churn at will etc... still because there are millions of them, you can easily scale to billions of published records with good reliability... I dare you to try to overwhelm it... Try it.
8. You only need the DHT to work relaliably _some_ of the time... Everything else can be a caching server like DNS work, and you still get the censorship resistance because the DHT is better than ICANN root servers.
9. If you don't want to use a DHT don't, Iroh is perfectly happy with it, and Pkarr is perfect for machines especially, but even I am looking for Blockchain based alternatives because I want more for humans... But please stop lying and detering people from using Mainline DHT, it is a miracle that we didn't deserve but we have it, and if applications stopped injecting the DHT and open ports, the numbers will dwindle on the long term. If you are building a P2P application especially Bittorrent make sure to put DHT in it and encourage people to open ports or just open it automatically for them, it is great to have such free routing that costs individuals very little but enable magnificent things at scale.
#nevent1q…xp7u

2026-06-03 06:44:05 UTC

Every time I try to open my heart to open inboxes where anyone can send messages without pre-approval, I am reminded how broken that system is. At the very least the "message request" UX should be enforced to increase the friction and force users to be vigilant with first introductions.
#nevent1q…7r02

2026-06-02 08:00:08 UTC

- reply

Did he sell? I hope he sell it all, I am sick of these suits

2026-06-02 05:41:23 UTC

- reply

Yes Postcard is rust specific, but I am a Rust maximalist. Yes JSON is trash for crypto, but CBOR is not the only encoding that solves this. You can always define your own serialisation for each struct. But I understand why people gravitate towards CBOR

2026-06-01 19:24:28 UTC

- reply

I have signal and Matrix but I don't like it much. I can download Delta Chat again, last time I had no one to talk to so removed it.
Soon we should have our own app to be honest with Mlkut and Peergos or an approximation of it :)
Let me know if you have a preference.

2026-06-01 19:06:13 UTC

- reply

I love talking about history and how I came to make choices I made... But if you want to ask me about these, I will have to insist you ask me in generic ways.

2026-06-01 19:02:12 UTC

- reply

For the record, I am not ignoring this thread I just don't want any drama, and I don't trust people to respect my emotionless comments without considering me a shill or a bitter ex employee... So I have to just pretend I don't know what are you talking about and live in peace 😅

2026-06-01 18:15:24 UTC

- reply

If so then that must be in a Peergos-like system, and definitely not in DNS records

2026-06-01 17:38:38 UTC

- reply

If Bluesky is any guide, ain't no body is even going to take the step to take custody of their names so it is barely a concern, but yes I will do my best to make it as convenient as possible for fee sponsorship, but at the end of the day, if the user is retarded they can always transfer the ID to a Rootstock address that contains all their money, what the fuck can you do about that.
On the other hand I can only wish that people actually use Rootstock at all so that is a champagne problem

2026-06-01 17:35:40 UTC

- reply

Not necessarily, because;
1. Initially you are most probably going to get your name from someone else who did register a batch earlier, so that Registrar is assigning you a name as a "trust me pro" situation, where you sign-up with a Passkey or email or whatever, and the registrar promises to eventually transfer ownership to you when you are invested enough to try self custody.
2. When they transfer it to your address, they may pay the fees themselves, potentially just as a part of the service, that you paid for other ways, or as a batch transaction for fee saving.
3. The address that you transfer the ownership to could be any Rootstock address, and most of the time it is only meant for rotation and recovery etc., you don't have to put RBTC in it, and even if you do, you can use a swap app to transfer BTC to RBTC and that swap app isn't likely to dox that relationship, but if you are really paranoid wait until there is good privacy solutions on Rootstock
So long story short, yes if you aren't careful to take some steps to break the trace from your main wallets to the address you use for making key rotation events you might get in trouble.
My ideal scenario is that all these transactions are performed by your Registrar/Nameserver/Hosting provider and you just pay them for your subscription and that is just among the costs.

2026-06-01 17:27:04 UTC

- reply

Yeah well, unfortunately I think this must require a hosting provider. At best like if you push Nameservers to their extreme you can add a TXT record with Bio and a TXT record with a link to your photo, or use data URL to encode the entire image
.. and the Nameserver does the profile job. But if your Nameserver is so generous with storage, I suspect it can double as a Peergos server as well. The main difference is that if you put the profile in Nameservers, you can use it without having to use Peergos SDK.
I often tell myself the reason people tried WebFingers and Gravatar but never the same in DNS, is because no body has domains.

2026-06-01 17:08:10 UTC

- reply

Realtime voice and video yes for sure, Iroh is already used for synchronous chat in Delta Chat no? Iroh doesn't support store and relay though, so you can't use it for messaging while everyone else is offline, you need a relayer, like in Simplex

2026-06-01 16:51:32 UTC

- reply

Actually we have the opposite problem most protocols shy away from solving DNS, and even the ones who did solve it with centralized solution are trying to remove it, and most are just using capability system.
For example Willow identifies the namespace as a pubkey, but don't bother with key rotation for that namespace, there is delegation of course, but if the master key is lost it is over.
Peergos has centralized PKI and they are planning to remove it and rely entirely on capability tokens for the only part that use them ; establishing a friendship connection and grant friends access to some stuff.
So really Mlkut name system, if it works, it will be like a sovereign DNS that you use, possibly along side good old ICANN to rotate the Namespace keys of Willow if lost, and point to a file in Peergos with a short URL.
Put another way, most protocols use identity systems that are cryptographic by nature, but they don't exclude a layer of indirection through DNS to deal with the needs that can't be done with cryptographic primitives... That is exactly where Mlkut Name System would fit.
For example, NIP05 could have used DNS TXT records directly instead of relying on HTTPS, but you would need the clients to treat that as the root of identity.
I don't think it is hard to create clients that do that in Nostr, or in Willow or in Peergos. In fact it is much easier in Willow because you don't have to worry about interop with all other Willow clients and communities, since you don't actually care about interop outside the community.

2026-06-01 16:31:52 UTC

- reply

I think the genius of Willow is that it starts from communities already, literally the whole protocol revolves around a shared space with access control, and an implicit relay.
It is hard to argue against this or insist that you still need identity system or database per user in that scenario.
Maybe still need an identity system for the community itself, and definitely I want to have websites and personal blogs and message passing like ActivityPub etc... but if you just start from chat groups and communities, the participants identities are as simple as users in a forum.
So who knows, maybe we will end up using Mlkut + Peergos + Willow... All of it.

2026-06-01 16:15:35 UTC

- reply

Actually, there is more, the main author of Willow invented set reconciliation algorithm, which I then explained to the author of Strfry relay, which he then implemented as Negentropy which is a set reconciliation algorithm for Nostr events that Damus uses to do set reconciliation with other relays.
So in a sense, Nostr already implemented some of Willow protocols :).

2026-06-01 15:06:05 UTC

- reply

It is important to use Blockchains judicially as they are expensive, hard to scale and very full of friction for users, so I make every effort to only use it when I must, and keep the rest offchain, in my experience you should always minimise your exposure to things that require consensus and even P2P in general, because they are where usability and reliability suffer compared to client/server flows.
Thank you for the engagement it was fun.

2026-06-01 14:58:14 UTC

- reply

I didn't say all credintials are DNS based, but most are indeed based on a PKI (public key infrastructure) as in a server or a set of servers mapping names to keys and keys to keys etc.
As for the primitive, I think it is misguided, the web will never be built on top of signed events in the ether. Case in point, everything we are doing right now can be done in a chat group or DMS in Matrix Protocol which is built on Homeservers, and yes they have signed events but they don't need OTS, they can just make usernames to keys, I am just suggesting we replace Matrix centralized identity providers with decentralised DNS registry.
Same applies to how ActivityPub and AtProto work.
But it is your time and energy and you are free to make the bets you want. I am sorry to say there are no chance anyone in Nostr would start using key delegation, and no chance anyone outside Nostr will start using signed events in the ether... So the market fit is quite difficult to find.
Best of luck still.

2026-06-01 14:39:52 UTC

- reply

Apologies if I sounded rude, I just encourage you to think of DNS name server with signed records in the same way you think of signed Nostr events and realise that there is nothing you can't do, it is basically a better version of Nostr "lists" because you don't have to wonder which relay they are stored at.
I also am frustrated that I have to explain DNS from the analogy of Nostr events, because DNS is much older and is being used for these stuff already.

2026-06-01 14:33:03 UTC

- reply

The purpose is anything you can do with DNS records including TXT records.
If you can't imagine how to use that to authenticate everything else then you should try to think harder.
But to be fair I don't care enough about the usecase of events distributed in the ether so I won't try to prove that signed DNS records are sufficient to satisfy that use case as well, although it is easy to show that that's how all credentials in the universe work, using a reliable PKI.
So you will have to think this through yourself, for me I only care about fixing the usecases that are actually desirable, all of which don't require signed events in the wind, instead they very much depend on reliable hosting.

2026-06-01 13:58:21 UTC

- reply

On chain you put a Nameserver URL and a public key that you will use to sign the records you store on that Nameserver... That is a delegation of sorts.
If that records key gets leaked, you do exactly what you would do with any delegation, you revoke it and make a new delegation on chain.
If that is unacceptable, you can make multiple tiers, so a Cold Records key that you keep safe, and a Hot Records key that you sign offchain records with, if that gets compromised you use the hot key to disable updates until you make the onchain rotation...
Many ways to do this, but at the end of the day it is not different from a signed profile event on your outbox server.

2026-06-01 13:43:22 UTC

- reply

First of all, you can prove any single file with succinct merkle proof that is the whole point of merkle trees as opposed to stream hashing all files.
Second you can still OTS merkle roots.
I just want you to stop and start thinking for a minute if there is a single person on earth cares about accurate timestamping of shit posts other than you as a protocol designer.
I am not telling you you shouldn't use OTS, it is not adding much cost, but I am definitely saying no body gives a fuck.

2026-06-01 13:06:28 UTC

- reply

The merkle tree just provides cheap signature over the entire set of files on your server... So instead of signing every single file you sign a merkle root and that also attest to the entire set, inclusion and exclusion proofs etc.
But yes, once you migrate from one server to another, the next server should sign the latest merkle root (saving it the effort of signing all files one by one), and then any app who notice the sever change, can go check if the merkle root changed from what they expect from last time they downloaded the proof of a file, then if the merkle tree changes they can download the new proof.
This could also be designed like a Blockchain so each update to your entire database points to previous updates, with something like a Skip list instead of link list, so now the new server can not only sign over all files at once, but even all history at once, then old server can't make updates with old keys.
Bluesky doesn't go that far though, and indeed most people don't give a fuck most people are satisfied by just checking the server if there is a controversy, just like you can doubt a web archive or a screenshot...

2026-06-01 12:35:35 UTC

- reply

In Bluesky like in all proper systems, there is a single server that you write to, think like in WebDAV or in Google Drive or Proton Drive...
Because you have that, you also have a merkle tree that is built on top of the entire tree and signed with one key at the root.
The moment you change your server, and thus you also change the delegation to the signer over the merkle tree, apps realize that your identity is now pointing to a new signer of the merkle tree..
So they abandon all previous events signed by the previous delegated server key, transitively with merkle proof and then go fetch the same event again from the new server with the new signer.
That if they care of course, in reality if anyone is suspicious whether or not a Post that is featured on a feed is actually created by you, the simplest solution is just go read it from your Homeserver, and they know that Homeserver the same way they know the signer, they read it from the PLC directory which acts as a centralized ledger instead of a Blockchain.
The same applies in systems that use Rootstock instead of the centralized ledger.

2026-06-01 12:18:08 UTC

- reply

You make something like {
content: "Hello world",
author: "243.inkan.alt",
signer: "....",
signature: "....",
timestamp: OTS proof
}
And then people can go check that "243.inkan.alt" has a DNS record that claims that within a certain period the signer above had a delegation.
Of course in my system I won't even bother with random events floating in the ether like that, instead I would just go visit people domains like normal and read their blogs or micro blogging or whatever data that they host or someone else host for them.
I am trying to fix DNS because I think it is sufficient and the normal web is good enough.
But even if you must use events and float them around, think of the DNS as an offchain Central Public Key delegation database that people and apps can go check, after they resolve its position from your record onchain, and then you can do anything with that really.
If you are not married to your solution we can work together and I can show you how would this work further... But no pressure I think it is valuable to try everything

2026-06-01 12:02:45 UTC

- reply

That is definitely a problem indeed, but it is a problem because you are using Nostr with key delegation and Nostr is very unfriendly to that.
If you take a step back and think from the point of view of the normal web, you don't need this mess.
For example in Bluesky, you can rotate keys and recover them and delegate keys to a server etc.. but you don't need timestamps, because there is a source of truth that you can rely on.
A master key that in a protocol with strong consistency and single source of truth can easily declare the entire state of blessed events, and delete everything else that they don't like.
Think of a website, once you change your server and point your domain to a new server you are done.
But yes in this paradigm of Nostr signed events thrown everywhere you need OTS for all events, but I am afraid you are going to be disappointed because Nostr is unfriendly to all forms of delegation, so after you try to conform to its limitations you will find that no one wants to use this system, we have historical examples of key delegation schemes that was removed even after wide adoption.
But I can't blame you for trying.