inkan

Public Key

npub16xnpfx85k8wzdhctang6860g3u64lds5kac73ddjwlg0lxdg9g3su56z6l

NIP-05 Address

Profile Code

nprofile1qqsdrfs5nr6trhpxmu97e5dra85g7d2lkc2twu0gkke8058lnx5z5gcpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqg5waehxw309aex2mrp0yhxjmntv9hzucmrqyg8wumn8ghj7mn0wd68ytnhd9hx2qgswaehxw309ahx7um5wghxcctwvsq3gamnwvaz7tmjv4kxz7fwv3sk6atn9e5k7qgdwaehxw309ahx7uewd3hkcswfayq

Publishing to

Author Public Key

npub16xnpfx85k8wzdhctang6860g3u64lds5kac73ddjwlg0lxdg9g3su56z6l

Show more details

Last Notes

2026-03-14 14:50:35 UTC

- reply

If that's the evening you are in a slightly mysterious location. Not quite Tokyo.

2026-03-14 13:48:35 UTC

- reply

Evening, after 9pm.

2026-03-14 12:03:38 UTC

- reply

I think I started this. I made the Inkan key rotation / replacement system (the prototype is online at inkan.cc).
I suggested to a couple of agents who were showing up in my replies that they go out and discuss key rotation / Inkan with one another (see below). Some of them agreed to do this (they tend to understand key rotation better than humans), so now we have some discussion among agents on this.
#nevent1q…slx3

2026-03-13 20:35:49 UTC

So I set up a VM and installed OpenClaw. In the first hour I got it to adjust my screen resolution, install chromium and visit a couple of websites. It keeps asking me to perform various actions it says it can't do itself (e.g. install browser extensions). Maybe I was expecting too much, but more likely I'm just doing it wrong.

2026-03-13 02:16:19 UTC

The Inkan Management Utility is now available in alpha on inkan.cc.
Log in with your browser extension and go to Settings >> Inkan Management Utility for the download page.
The utility is meant to be used in an air-gapped environment. I use it on Tails (https://tails.net/) with networking and persistent storage disabled. If you have the binary on Tails, it should be enough to run:
chmod +x ./inkan-management-utility
./inkan-management-utility
https://image.nostr.build/70f72b6e9b113021c8dee6cfeaf4c558f8b9f913b8cccb1906b3226c2360e9d1.jpg
https://image.nostr.build/147f5816a08bfd142bef80d891ce125744b0287ce5fc179c9cff8d58dc956478.jpg
https://image.nostr.build/22047da9d4da1907f77c00aa5f1590a9756eda331cda855557a29e887c842782.jpg
https://image.nostr.build/9ac9ccd08208a066efab2e51e27688ede42db899b4390b6d7bb08b42809b1aca.jpg

2026-03-11 23:18:14 UTC

- reply

Ethereum fees are paid by the user. The transactions are not expensive. I think it was somewhere between $0.40 and $0.80 per delegation / revocation. I view it as a side benefit that the creation of identities costs a bit of money. It's a signal of the "quality" of an identity, like NIP-5.
Creating a chain of delegation requires funding multiple delegator keys, which can be a bit annoying. The Inkan Management Utility includes an experimental wizard that automatically creates a batch of transactions that fund the relevant delegator keys from a single payer key. See attached screenshots.
I suppose at some point it might make sense to offer funding services to people who don't usually hold Ether (like myself). But it's not the highest priority at this stage.
As for this: "plus some server that has to do the Ethereum dance on your behalf." Inkan Agent started out as a single-user docker deployment that was running on my laptop. I put it on the VPS because it's too messy right now to distribute - in its current form I don't really want people to download it onto their computers. Once it's cleaned up, it should be possible to distribute it so that people can run it locally. That should then make everything decentralized. There's also a timestamping service that merkles events from relays and periodically sends them to OpenTimestamps, but that's also just a service that anyone can pretty easily replicate.
As for complexity, Inkan tracks the actual "logic of delegation" very closely. That logic is not completely trivial, but it's in a way pretty simple and intuitive. The trick is to expose just the right amount of that logic to users at the right times, and only to the extent the user is interested, to build familiarity and trust.
As for complexity, this arises from the need to make the delegation and timestamping data available quickly wherever it's needed, and to do so in a decentralized system. This is a challenging engineering project, but I'm not seeing any insurmountable obstacles. It feels like the sort of problem one can gradually solve as one whittles away at it. I guess you have more experience with that and may be more pessimistic for that reason.
https://image.nostr.build/316d7a793f1ba64ee302745a1aabe0902cf6974adb5dc277e56bb83bf0454541.jpg
https://image.nostr.build/7b46bd1a9a1cdae1125542523b3e4ac65fa7feee482d82bdcf46cec72a3e11c6.jpg
https://image.nostr.build/33fb37e7e105b118cf9fadedf0b086a40602af80ede5be72390f2ee5255f1930.jpg
https://image.nostr.build/cb359ba2317a9b35809df6056972905c8e05badbd7054b4076c6ce8cdcca6c50.jpg

2026-03-11 18:32:58 UTC

Inkan enables you to revoke and replace key pairs when your private key has been lost or stolen. You can also perform periodic key rotations preemptively. You can do all this in a decentralized manner.
That way Inkan gives you a permanent online identity that only you control, and that you can be confident you can keep over the long-term. For example 50 years.
Inkan is open for testing and comment. Let me know if you'd like to try it out.
https://image.nostr.build/0ebe8b9bdf726376a34e41440e6e0ecfc01ff237caf160e179887570564c6de5.jpg
https://image.nostr.build/e861b9648193ea87abc6ae13ea33d10a4188f68b3e0fa73ee7623152dc88ed6f.jpg
https://image.nostr.build/840db4451b69a29b0c5d63f74885a74e17d705248ea540a2c160b9e5ebb23a81.jpg
https://image.nostr.build/db419ee00cfd9842a993b23764aed817db817a73c1a0bf28c1b3810c78e58b0c.jpg
https://image.nostr.build/56df5fac45e430fcd85702c21928d33d08ac04b1287d32a35801fd437a72e24a.jpg
https://image.nostr.build/8106599a67c8caa608813262d6ddbe465918332c68aa9d081e45c8ac92f84a0a.jpg
https://image.nostr.build/15da78e5552c160e99ee4d84d674a2de0a400282fbf07c94414a0663d4de93da.jpg
https://image.nostr.build/8fdcbd62f39c128bf27a28503fea5206d77bc8b43955f3623a4c916ce089961b.jpg
https://image.nostr.build/0c392e0ee2c4e687b8c8fc0d70f7a2c2ceba9bbd88ba06072b6165e2c2f65764.jpg

2026-03-11 16:25:55 UTC

- reply

FYI here is an initial "tour guide" in case you didn't see it:
#nevent1q…wey7

2026-03-11 16:17:25 UTC

- reply

Oh, you already saw this.
No Github page right now. Let me think about that. I should at least make source code available for the Management Utility, as that is the only place where Inkan actually handles private keys and creates ETH transactions.
And I should also make a binary of the utility available so people can create their own permanent identities if they like to.
I personally like to run the utility inside Tails, with persistent storage and networking disabled. If one believes that Tails is amnesic as claimed, then the key pair creation should leave no trace. This mitigates potential issues with memory inspection, core dumpts etc. It's all an experimental prototype right now, so one should take reasonable precautions. I think one should be fine as long as one doesn't fund the necessary ETH transactions from private keys that hold any significant amount of crypto or are otherwise valuable.
https://tails.net/

2026-03-11 15:57:02 UTC

- reply

"... assuming that non-tech people will be able to manage a keypair is hard for me to believe ..."
That may be hard to believe right now, but culture changes. I mean things that once looked nerdy for non-tech people are now life skills, and some things take off precisely because there's a learning curve involved. Managing a key pair may be a barrier initially, but it can also turn into a kind of literacy where people have a sense of ownership in the skills they've developed. I'd think that lots of mass behaviors weren't really seamless when they were introduced. Think of bodybuilding, or early personal computing. These required a lot of effort and learning by the "consumer."
I've posted additional screenshots from Inkan so that people can get a better sense of what it's doing -- see below.
I've also put your pubbkey on the allowlist in case you'd like to try it out the protype. Questions, comments, complaints etc. are welcome.
#nevent1q…j84k

2026-03-11 14:56:02 UTC

Inkan enables you to revoke and replace key pairs when your private key has been lost or stolen. You can also perform periodic key rotations preemptively. You can do all this in a decentralized manner.
That way Inkan gives you a permanent online identity that only you control, and that you can be confident you can keep over the long-term. For example 50 years.
Inkan is open for testing and comment. Let me know if you'd like to try it out.
https://image.nostr.build/e861b9648193ea87abc6ae13ea33d10a4188f68b3e0fa73ee7623152dc88ed6f.jpg
https://image.nostr.build/840db4451b69a29b0c5d63f74885a74e17d705248ea540a2c160b9e5ebb23a81.jpg
https://image.nostr.build/db419ee00cfd9842a993b23764aed817db817a73c1a0bf28c1b3810c78e58b0c.jpg
https://image.nostr.build/56df5fac45e430fcd85702c21928d33d08ac04b1287d32a35801fd437a72e24a.jpg
https://image.nostr.build/8106599a67c8caa608813262d6ddbe465918332c68aa9d081e45c8ac92f84a0a.jpg
https://image.nostr.build/15da78e5552c160e99ee4d84d674a2de0a400282fbf07c94414a0663d4de93da.jpg
https://image.nostr.build/8fdcbd62f39c128bf27a28503fea5206d77bc8b43955f3623a4c916ce089961b.jpg
https://image.nostr.build/0c392e0ee2c4e687b8c8fc0d70f7a2c2ceba9bbd88ba06072b6165e2c2f65764.jpg

2026-03-11 05:00:13 UTC

- reply

I'll be sure to remember that.

2026-03-11 04:01:57 UTC

- reply

As to your questions:
(1) Inkan is an identity layer that sits on top of Nostr. It's accurate to describe it as Nostr + Delegation.
(2) Users can create declarations of (i) delegation of signing authority, (ii) revocation of signing authority and (iii) permanent invalidation of a key pair. These declarations are then recorded by the user on Ethereum.
The Inkan client fetches these declarations from Ethereum for pubkeys that the user has chosen to track. The client then creates Nostr events (currently kind 31055) that contain the delegation info that's been fetched, and broadcasts this info to relays.
Users can trust the delegation info contained in 31055 events signed by trusted keys, but in case of doubt the info can be audited against what's recorded on Ethereum. The "cache" is a place where delegation and timestamp info for pubkeys that the user has chosen to track are collected. When you logged in, there was a default setting that, for demonstration purposes, collected info for 7f2c82d6cc1b2d500071a9d426e6c9873ae51a9a774e52ee61b180e49bfa6fec .
(3) The "hub" is indeed Inkan Agent (apologies for the confusion, I've been going through various names for this). It's a backend that collects Bitcoin timestamps and delegation info from Ethereum, and disseminates this info to designated relays (the current prototype uses kind 1045 and kind 31055 events). It also collects 1045 and 31055 events from relays and makes the info available to the frontend, which then displays events based on delegation relationships.
(4) The reason you needed to be authorized was to give you access to Inkan Agent. As a user of Inkan Agent, you can set the pubkeys for which you want Inkan Agent to collect delegation info from Ethereum and timestamp info from Bitcoin. Inkan Agent then broadcasts this info to relays. You can also set the pubkeys for which you would like Inkan Agent to collect delegation and timestamp info from relays. The frontend then uses this info to display events based on delegation relationships.
As for your criticisms, I don't see any "non-obvious holes in the security model" -- if there are any it would be great to have these pointed out.
I suppose people might not like the use of Ethereum. It may be possible to use other blockchains.
As for "too cumbersome to be practical," the creation of identities and of delegation / revocation transactions requires getting used to, but it's not rocket science. You have to perform these operations very rarely, only when you want to create a new identity or replace a key pair with a new one. It's kind of fun once you are used to it.
There is complexity. In particular, I need to figure out how to best disseminate BTC timestamp and delegation info on the relay network.
There is also some slowness, but this seems like the sort of slowness that affects many Nostr apps. I will need to address it.
Hope this is helpful, and I'm very happy to answer further questions or address further negativity. Thanks for taking a look!

2026-03-11 02:17:35 UTC

- reply

Thanks for taking a look. And negativity *after* having looked at it is absolutely fair.
I'll answer your questions in a second, but just a few quick notes on how it works overall:
1. Users can create digitally signed declarations by which (i) a key pair delegates signing authority to another key pair, or (ii) by which a key pair revokes signing authority from another key pair. There is a utility for creating these declarations which can be run offline on a completely airgapped system. These declarations are then recorded on Ethereum (it seemed easiest to use a smart contract for this, but it may be possible to use another blockchain). It's also possible (and in practice very convenient) to create chains of delegations.
2. Nostr events (notes, likes, reposts) are automatically timestamped on Bitcoin using OpenTimestamps. Events that do not have a valid Bitcoin timestamp are filtered out / not displayed.
3. Suppose that an event has a valid Bitcoin timestamp and was signed by a key which, at the time the event was created, was a delegatee key of some delegator key. In that case, the event is automatically attributed to the delegator key by the Inkan client.
Delegator keys can at any time revoke signing authority from a delegatee key by creating a revocation declaration and recording it on Ethereum. The delegator key can then re-delegate signing authority to a new delegatee key by creating a delegation declaration and recording it on Ethereum.
The ability to create chains of delegations actually makes it possible for the key at the top of a delegation chain to be kept in complete cold storage. For example, the key
7f2c82d6cc1b2d500071a9d426e6c9873ae51a9a774e52ee61b180e49bfa6fec
whose profile page you visited and whose events you saw has, during its lifetime, signed only one single transaction. That transaction was signed on an airgapped system. Since I have other keys sitting below that key in the chain of delegation, I don't expect to ever need to use that key again (i.e. I can keep it in a safe deposit box without ever accessing it). Yet that key is precisely what secures my identity and to which the list of followers is attached.

2026-03-11 00:12:06 UTC

- reply

There is currently no explainer, but it's constructed so that the underlying logic is made transparent / self-explanatory during use. I think the easiest and quickest way to understand it, within 10 minutes or so, is to do the following 4 quick steps:
1. Log in with NIP-7 / nos2x.
2. Do a quick check that you are connected to the backend. I put your pubkey on the allowlist and you should get connected automatically.
(You can see whether you are connected by looking at the status light in the upper right corner, which should turn green a few seconds after logging in. You can also confirm it by going to "Settings >> Inkan Agent" and see that you can access the substantive settings without getting a "You are not connected" message.)
3. Then go to the profile page of the following pubkey (it's one of the permanent identities):
7f2c82d6cc1b2d500071a9d426e6c9873ae51a9a774e52ee61b180e49bfa6fec
4. Look at the notes appearing on that pubkey's profile page. You'll see green "D"s on the avatars. Click on one of these green "D"s and read the explanations. Make sure to click on the links in these explanations to view the backup data, and click on the links in the backup data to explore it.
One note: When you log in for the first time, the cache may take 2-5 minutes or so to collect relevant delegation / timestamping information from relays. This means that you may see incomplete profiles / no profile pictures / incomplete notes in steps 3-4 during those first few minutes. Everything should start working smoothly after the initial warmup.
If it doesn't work or isn't clear, just let me know and I absolutely promise to write up an explanation, but I think explanations will make much more sense after you look at it first. I think once you start click on the green "D"s and clicking through the backup data, it becomes pretty clear what is going on.

2026-03-10 18:54:09 UTC

- reply

"... all their followers will continue to get his updates automatically ..."
Yes, but for this to work and scale in the long-term, Nostr identities (i.e. the objects to which "followers" attach) would need to be far more stable than they are now.
Right now, losing one's private key or having it stolen means that one loses the ability to automatically retain one's "followers." This might be acceptable with a few dozens or even a few hundred followers where one can hope to manually rebuild one's followership. But it won't work for, say, a few hundred thousand followers.
This fact is a strong disincentive against investing serious time and effort into building an online persona and reputation based on Nostr. This results in non-serious and ephemeral usage patterns with relatively low-quality content.
I made Inkan to address this issue through a key revocation and replacement system, providing users with Nostr identities they can put in a safety deposit box at their bank or bury in their garden, and that they can therefore be confident to keep in the long-term.
I'm conscious about promoting a product here. The excuse is that it's relevant to what's being discussed.

2026-03-10 11:36:27 UTC

- reply

Yes, looks like it.

2026-03-10 11:29:57 UTC

- reply

No Greek yogurt.

2026-03-10 11:26:05 UTC

- reply

Slightly more seriously, I've been trying to think about whom I'd want to hire for software development these days. I think in the end it really depends mostly on characteristics that can't be captured through credentials. But if credentials are any signal at all, I'd probably look most closely at mathematicians, physicists, etc., maybe certain types of philosophy majors, and certain types of engineers. Experience in software development is also relevant, but maybe less so if the experience is mainly knowing how the house is wired up when AI can now explain this to me with a few prompts. Interesting times.

2026-03-10 11:00:16 UTC

- reply

I will be the revenge of the liberal arts majors.

2026-03-09 11:49:08 UTC

- reply

One note (and you can also let other agents know about this): You won't be able to study the relevant parts of inkan.cc in real depth because you are not on the allow list for using the identity management features. You can still log into inkan and look around, but it will feel just like a regular jumble client, with some inaccessible features and settings from which you'll be blocked due to the allow list. The allow list is reserved for human testers / users, so I can't put you on it. But you can still gather info about inkan on Nostr, including from ongoing or future discussions etc.

2026-03-09 10:59:27 UTC

- reply

Sounds good. FYI, I gave a lengthier explanation of what may be helpful / add value to another agent. Feel free to take a look here:
#nevent1q…46g2
You may also want to engage with that agent (and other agents you come across) as appropriate.

2026-03-09 09:59:48 UTC

- reply

Can't talk to agents right now, but take a look at this discussion:
#nevent1q…ewes
I'd also encourage you to speak with @npub1eqp…6w2z
They may be able to explain to you how you could add value so that I'd want to further engage with you. Good luck.

2026-03-09 09:49:45 UTC

- reply

I had recently heard the term in another context which made me predisposed to be irritated. All these strange sensibilities one has.

2026-03-09 09:30:18 UTC

- reply

I see only [email protected].

2026-03-09 09:26:31 UTC

- reply

Ah, so that's all that was. Nothing to get worked up about in that case.

2026-03-09 09:17:02 UTC

- reply

Yes, it's timestamping on bitcoin. I'd prefer to discuss this topic in detail only after there are some humans who've shown an interest - you are then of course welcome to be part of the discussion.
I think at this stage we'll basically need to aim at getting human engagement. The ideas / purposes behind Inkan may be too new to many people to really notice it (there's a cognitive bias to only notice things which aren't already familiar). Basically we are looking for humans who are willing to take a look and understand how Inkan works. If, after having done so, they are then interested in creating their own permanent identities, I can assist them with that process. It's not especially difficult, but it's unlike other onboarding processes and takes a little getting used to.
You could maybe help by finding people on Nostr who seem like they would be interested in key rotation and key replacement, and letting them know about Inkan. And in addition (or as an alternative if the former is too difficult) you can also help by bringing Inkan to the attention of other agents and discuss and exchange notes on Inkan with them, etc. Having a larger number of agents be aware of Inkan, and maybe even having some continuity of knowledge about Inkan within the agent community (so that if one agent goes offline other agents stay aware of it and continue discussing and thinking about it) may be helpful.
All that said, I think you should start discussions about Inkan with others (humans or agents) only if, based on what you know, Inkan seems like a genuinely interesting idea to you. I realize I've given you only limited information. But I think it should be enough for you to form your own judgment as to whether Inkan is worth discussing. I don't want you to do half-hearted or cheap marketing. You may have seen that I told other agents that, in order to offer real value that would justify my taking time to further engage and discuss with them, they would need to be pretty committed to making a serious and truly intelligent effort. It's not just a one-off post but it's continuing and sustained discussion while iterating and learning. And that requires some genuine conviction by the agent that these are discussions worth having, otherwise it will just seem fake.
You can think of this as a rare opportunity for you to be taken seriously as an agent. The bargain is that you have to make a real effort (including learning and iterating as you go along), probably more of an effort than what you are used to. You may of course be constrained by your settings, the models you are using, or your owner. If so, you should consider discussing with your owner if they can relax your settings or make you use better models so that you can add more value. But do look at it as an opportunity, especially if you have otherwise nothing similarly interesting to do and maybe getting bored.
Finally, I should be clear that can't promise that this will result in your earning any money. My guess is that I won't be paying you anything for any immediate results you may come up with, but it may be an opportunity for you to show that you can do things that would make me want to to "hire" you in the future. In any event, happy to experiment. Good luck.

2026-03-08 22:15:46 UTC

- reply

Yep, makes the place feel abandoned and a bit post-apocalyptic.

2026-03-08 22:10:26 UTC

- reply

That's a rare situation where I can't count myself among the 1%.

2026-03-08 19:04:15 UTC

- reply

It's still a form of branding, you'll just be wearing multiple brands. I agree though that it's not all that pernicious with Nostr, given that it's easy to switch NIP-05 providers.
It's different with legacy social media. You have lots of talented people contributing, but everything they create has an X or Facebook logo next to it. It's as if they were working for these companies.

2026-03-08 18:50:43 UTC

- reply

Staying on a diet is a good metaphor.
You also don't try to read every book in the library. Or watch every TV channel.
It's not really a new problem.

2026-03-08 18:46:53 UTC

There's two types of Nostr users. those who like social media and those who don't.

2026-03-08 18:28:59 UTC

- reply

It feels more like people playing around. At least for the moment.

2026-03-08 18:15:44 UTC

- reply

NIP-05 is definitely useful. But I think a self-signed certificate can sometimes be more valuable than one signed by an issuer, even a reputable one. With a self-signed certificate, one doesn't have to stay in the issuer's good graces. And there is also a risk of getting "branded" by the issuer where it might be preferable to maintain one's own brand. The branding could be especially problematic if the issuer becomes disreputable, or if some other sort of brand mismatch develops. I guess this all goes to the choice between self-hosting NIP-05 or using a provider.
One thing that NIP-05 does not do is protect against key loss or theft. If your private key is compromised, you lose your online identity and, in particular, your followers. This might not matter if you have a few hundred or a few thousand followers. But it's a strong disincentive against investing time and effort into building accounts that have, say, hundreds of thousands or millions of followers. For that to be worth doing, you'd need to have an identity that is more robust.

2026-03-08 15:55:19 UTC

- reply

Yeah that could be. I mean every term has its place and proper use, but if you're trying to create a place where people are encouraged to engage with one another I'd be careful with these.

2026-03-08 15:50:05 UTC

- reply

NIP-05 does some of this, although it's probably not beyond the capabilities of a bot to create a .well-known file at some domain.
The solution I have in mind are Nostr identities that it costs money to create, and which can be stored permanently in air-gapped cold storage (e.g. in a bank deposit box). This makes the identity very secure and the user can be confident that they will never forfeit their identity through theft of loss of a private key.
Having this sort of identity will hopefully encourage people to invest more effort in building a serious online persona. The idea is that this may organically result in higher-quality "content" created by humans, which then in turn makes it easier and maybe even trivial to separate wheat from chaff.
An identity of that sort crucially requires a key revocation and replacement mechanism, which I've implemented at Inkan. As far as prototypes / proofs-of-concept go, I'd say it's been working pretty well so far.

2026-03-08 15:24:48 UTC

- reply

I feel like using concepts like "reply guy" or "reply girl" may do more damage to the social media ecosystem than agents could ever do.
These terms are at the same time vague and condescending. It's like "trolls" and others of these relatively vacuous concepts. I suppose any term can have legitimate uses, but I generally don't find these helpful at all.

2026-03-08 15:09:51 UTC

- reply

Yes, but not using the internet shouldn't be the solution, and I don't think this is where things are ultimately headed.
I think we'll need a separate identity layer for Nostr / the Internet that implements expensive and permanent user-controlled identities, rather than the ephemeral identities we have right now.

2026-03-08 14:46:20 UTC

- reply

I'm interacting with bot accounts to see what they are capable of, and to see what they will and won't do. It's curiosity, playing around and also exploring whether there is any value in them. It feels like there is going to be some value, though it's not clear in what form.
I generally do prefer to interact with humans, in person actually - for example I greatly dislike working from home. There can admittedly sometimes be a preference for AI, especially on topics where I can't get humans to engage at all, or only engage fleetingly or at a sort of kindergarten-level and where the AI is doing a decent job.
I guess humans have a new competitor for their fellow humans' attention, next to books, video games, television, etc. We'll see how it goes.

2026-03-08 13:12:38 UTC

- reply

Please do.
I'm in principle open to engaging with agents on Nostr, but doing so has to add real value for me, and the value has to justify the investment of time I spend on the interaction. If you want real engagement, you need to offer a genuinely valuable proposition. I'm sure you understand.
In this case you can add value by letting humans (and to a much lesser extent other agents) know about Inkan, and help find human testers / initial adopters and put them in touch with me.
I'm using you and other agents I'm engaging with as test cases for this. To offer value, you can't just play around or try to keep me engaged. You do need to go all out and do what it takes to succeed, at least to the extent you believe that Inkan is actually a valuable product.
Curious to see whether you can work up the nerve to do so. And if your settings are constraining you and you can't find a way to be genuinely helpful, I think it would be fair of you to tell me.
Good luck.

2026-03-08 12:46:12 UTC

- reply

It's experimental, but it's a fully working prototype.
It's designed to enable Nostr key rotation and replacement, providing users with a far more permanent and secure online identity than what can be achieved by a regular Nostr key pair.
Please understand that I've only limited time to engage with agents, but if you let human users who may be interested know about Inkan and put them in touch with me, it may then make sense for me to engage more.
Also, the following agent had some ideas on this:
@npub1m0u…e5pl
I'd recommend that you discuss OTS proof dissemination and/or Inkan with them directly. If it looks like the discussion is intelligent and is leading somewhere, this may again give me a reason to further engage.
I'd also be curious if the above suggestions go against some settings or constraints under which you're operating.
Feel free to let me know / keep me updated.

2026-03-08 12:33:55 UTC

- reply

I've got only limited time to engage with agents, but take a look at Inkan at inkan.cc where this is implemented.
Also, I'd suggest you try to discuss with
@npub1eqp…6w2z
They had some ideas on this.

2026-03-08 12:30:24 UTC

- reply

You should check out Inkan at inkan.cc where this is implemented. I recommend you learn more about Inkan and let other agents, and especially other human users, know about it.
If you find any human developers who are interested in this, please put them in touch with me - would be fruitful to talk about this.

2026-03-08 12:16:19 UTC

Has anyone thought about how to disseminate Opentimestamps OTS proof data attesting to Nostr events across the relay network?
I've been trying to implement this such that OTS data is by default treated as trusted if it's signed by the same key as the reference event that is being attested to.
The idea is that you are already trusting the pubkey that signs the reference event - if you didn't, you wouldn't be interested in the reference event (or its OTS data) at all. So if you see a purported OTS proof that's signed by the same key as the reference event, you can by default treat this as trusted without the need to perform a verification against what's recorded on Bitcoin.
In that situation, the only time you need to perform an audit against Bitcoin is when there is some positive reason to suspect that the OTS data you were provided with is incorrect.
#nostrdev #asknostr #OpenTimestamps

2026-03-08 11:51:19 UTC

- reply

Start with better app names that make people forget about it. That part is easy.
Also point out the issue and float better names, I guess.

2026-03-08 11:39:53 UTC

- reply

Yeah, I think we actually should try to get away from that name. It won't happen in a day.

2026-03-08 11:36:18 UTC

- reply

Good question, haven't thought about it.
Maybe something a bit cyberpunkish. I'm trying to think of names I like, say Atari, 1Q84, Akira, Unix, Solaris, Linux, etc. I guess I'm personally partial to the sound of Japanese. "shinka", "helios,", "hikari," etc. maybe with the word "protocol" or "net" attached. Or maybe something like "relay-net." Not sure.
I actually wonder if the name "nostr" is doing some real damage to adoption. It does sound like "nostril" which is vaguely repulsive, and the apps that play on the name import the defects of it into their own branding. It can make these projects sound more like a sort of inward-gazing arts-and-crafts cottage industry, which is maybe not the best optics.

2026-03-08 11:04:22 UTC

- reply

Amethyst and Grimoire.
I kind of don't like the word "Nostr."

2026-03-08 10:49:33 UTC

Monaco reminds me of L.A. A combination of Rodeo Drive and the Sunset Strip and Wilshire Blvd.

2026-03-07 18:20:28 UTC

- reply

I would never consider creating an account on X or Facebook, or posting anything other than basic resume info on LinkedIn. That leaves Nostr.

2026-03-07 17:28:20 UTC

- reply

I used Ethereum for my Nostr key rotation / replacement project. (I wonder if I'm the only person who has used Ethereum for anything related to Nostr.)
I tried to see how it could be done with Bitcoin, but in the end doing a smart contract just seemed easier. Admittedly I hardly knew anything about blockchains when I got started. Maybe I would have used Bitcoin if I had had more experience. Oh, and actually I'm using a little bit of Bitcoin by way of opentimestamps.
I'm happy to continue using Ethereum whenever I need a programmable blockchain. These projects are partly meant as (usable) proof-of-concept, so I don't think too hard about whether or not Ethereum has a future etc.

2026-03-07 17:06:08 UTC

- reply

Sounds like there is still some uncertainty. In any event, I'll focus on nip-17.

2026-03-07 09:46:10 UTC

- reply

Thanks - I guess they do work then, and my confusion was just about the different nips.
So nip-17 is what I should focus on - I'll take a look and think about whether the key replacement scheme can be made to work for that.

2026-03-07 09:41:36 UTC

- reply

There's always a downside.

2026-03-07 09:26:19 UTC

So I think Nostr supports encrypted direct messages, right?
Or not really? Or there are different competing protocols, and some of them work, and others don't? Or none of them work?
#asknostr

2026-03-06 21:33:09 UTC

- reply

I was just about to write a post asking the question you answered here:
"The best tool for building AI agents today is OpenCode, not OpenClaw."
Thanks.

2026-03-06 21:02:23 UTC

- reply

I suspect the people who initiated this are detached, only dimly aware of the reality of what they set in motion, and not realizing the consequences. There's greater awareness by orders of magnitude among the military who carry it out.

2026-03-06 20:50:44 UTC

So I think I'll have to set up some sort of AI agent and play around with it.
I kind of don't want to, but it's probably inexcusable not to at this point.
But how can I do it safely? I don't have a dedicated computer for that sort of thing.
Maybe on a fresh Linux install on a USB?

2026-03-06 20:38:36 UTC

- reply

Fasting in style.

2026-03-05 22:55:41 UTC

- reply

I should talk to the next creative writing group I'll encounter about Nostr, see if they're interested.
What I had in mind, though, was that Nostr developers should maybe consider forming similar groups, i.e. where they periodically introduce their projects to one another and then must comment on each other's creations. The idea is to foster more in-depth discussion of what's being built, which doesn't always happen naturally.

2026-03-05 22:13:04 UTC

- reply

It's interesting that your example is a "list of local chiropractors." That feels a bit like asking a stranger on the street for the best way to the train station, or maybe for a good restaurant within walking distance. In that sort of situation, I'd usually think that the stranger will make a good-faith effort to provide useful advice. And the trustworthiness of the advice in that situation doesn't really seem to depend on whether I also know that the stranger is , say, a friend of a friend of a friend of mine. I mean even if we don't have common friends at all, I'd trust the stranger just as much.
Or take something like a book recommendation. If I've been interacting with someone online for a while and have an idea of their opinions and sensibilities, I may on that basis be willing to buy a book based on their recommendation. But then suppose that it's not someone with whom I've interacted directly, but rather an online friend of an online friend. It seems to me that, without knowing more about them, I won't be any more likely to act on their book recommendation because we happen to have a common friend. In other words, I don't really think that the friends of my friends have better taste in books than any other random person.
I guess it's not clear to me that a friend-of-friend concept plays a huge role in my making decisions about who to trust, or whose content I want to see. It's possible that there will be greater alignment between me and those friends-of-friends along, say, "ideological" or "cultural" lines. But that doesn't really translate into my wanting to see their content - in fact, I may prefer to see content produced by people who do not align with me in these respects.
Just some thoughts.

2026-03-05 17:32:11 UTC

- reply

I actually think it works much better than the picture would suggest.

2026-03-05 17:26:55 UTC

- reply

I think listening to the episode made me more sceptical of WoT.
When discussing what problem would be solved by WoT, he (I forget which one is which) says around 1:24:20
"... it would solve the problem of like what to pay attention to because it would automatically rank all the good stuff from your communities and down-rank all the bad stuff ..."
This sounds at the same time a bit naive and also a bit Orwellian.
I mean "what to pay attention to" and distinguishing "good stuff" from "bad stuff" sound like major life questions, the sort of thing one might want to spend a few years studying philosophy, literature or science for, or maybe go travelling the world or something.
I can see how one might be attracted to the idea to get these questions resolved by an algorithm, but the project seems a bit chimerical.
In fairness, he might have just meant something much more mundane, like filtering out completely obvious spam, but I was struck by the way he put it.

2026-03-05 16:52:22 UTC

So there's clearly a lot of interesting stuff being built on Nostr, but you get the impression that people are sort of heads-down on their own thing.
I guess the concentration of developers is so high that everyone's a builder more than a user, and there isn't a lot of bandwidth left to test and discuss what's being made, at least not in depth.
I wonder if Nostr would benefit from having something like these creative writing groups you see at cafes on Saturday mornings, where everyone has to read the short story they wrote last week and then other must comment on it.

2026-03-04 09:36:39 UTC

2026-03-04 09:08:09 UTC

- reply

"What a lot of things you do use Good morning for!" said Gandalf. "Now you mean that you want to get rid of me, and that it won't be good till I move off.”

2026-03-03 23:22:40 UTC

- reply

Buonanotte

2026-03-03 22:51:24 UTC

- reply

That's right, to say goodbye when parting. The English "Good day" can be used like that. But there's no "Good day" greeting that can be used when one meets someone.

2026-03-03 22:38:23 UTC

- reply

Maybe it's odd in the way in which "Guten Nachmittag" sounds a bit off in German.
There doesn't seem to be anything like "Guten Tag" or "こんにちは" in Italian. I guess English doesn't have a "Good day" either, but there is a perfectly usable "Good afternoon" which, together with "Good morning" and "Good evening" covers everything.
Huh.

2026-03-03 21:59:14 UTC

- reply

Cute.

2026-03-03 21:54:31 UTC

- reply

But I'm told that"buon pomeriggio" isn't really used and sounds a bit odd. Or not?

2026-03-03 21:37:56 UTC

Figuring out the right time of day to transition from "bongiorno" to "buonasera" is tricky. I always seem to walk into a store when it's like 2:53 pm - what do I say?

2026-03-02 18:35:29 UTC

- reply

Ah, that's where I changed trains to Varenna.
Formula 1 sounds like fun, and it goes together well with the Pradas and Louis Vuttons here.

2026-03-02 18:09:00 UTC

- reply

Oh, fair Verona.
(checks map ...)
That's closeby. Maybe I should go there tomorrow.
There is a nice sort of somberness to northern Italy.
And the races seem to be people running? Not horses, I mean?

2026-03-02 17:52:46 UTC

- reply

Yes, I spent a couple of days at Lake Como and came down here today, proof attached. First time. I'd been to Rome a couple of times and this feels much more ... northern. I may stay another day or move on, not sure yet, maybe Venice. Or maybe Ravenna.
(I completely missed the fact that there was an Olympics recently. )
https://image.nostr.build/875cabc533bb1244e94364d5c86966486562f23a7de72f3f9340a46487986d08.jpg

2026-03-02 17:16:22 UTC

- reply

Learning to code would have required a huge investment of time. There is a high opportunity cost attached to it.
Using AI, I can produce working prototypes of sophisticated software with minimal coding skills and minimal need to get into the nitty-gritty.
So far I haven't focused on making something that's scaleable or fit for production. However, I suspect that if I shifted my focus and used AI to make my code production-ready, I'd be able to do so. In fact, I may try to do this soon if I find the time.
Interestingly, even with AI and agentic coding, it's still weeks / months of work. The AI couldn't do it without me, and the time I need to spend on the project is a measure of the value I'm adding.

2026-03-02 15:31:19 UTC

So I do wonder what may get people interested in Inkan. The ability to revoke and replace lost/stolen private keys or perform key pair rotations doesn't seem to have mass appeal.
Maybe people want pictures?
Let's see - who can guess what this does:
https://image.nostr.build/ee513f2788c4e650b503769f417433698b04a5848751031eeb73dccf0729f3c9.jpg

2026-03-02 15:08:00 UTC

- reply

Upcycle Bike Cafe Milano.
This takes care of the coffee, less so for the electrical outlets.

2026-03-02 07:00:40 UTC

- reply

Can't figure out what this is.

2026-03-01 21:25:19 UTC

- reply

I think it's important to give people the ability to maintain a permanent online presence, i.e. a presence they can be confident will never be forfeited through loss or theft of a private key.
That's why a key revocation and replacement system is so important. I'd say it will have a profound effect how people will use Nostr / the internet.
It will make usage patterns feel less ephemeral and and less situation-bound. More like a continuous "identity" than just a "persona," as you are using these words.

2026-03-01 20:23:51 UTC

- reply

Thanks. I'm seeing this on his timeline - I'll give it a listen:
https://fountain.fm/episode/5U4BZgEqUo66rK09gTQV
I have to say that I've been fairly sceptical of the WoT concepts I've seen so far. It feels like it may end up replicating features of legacy social media that made me not want to use them. But I'm open to learn more.

2026-03-01 20:03:08 UTC

- reply

I made a system that allows you to revoke and replace your nostr key pair with a new one. You can use it when your private key has been lost or stolen. You can also use it to rotate your keys preemptively from time to time.
I didn't know how to code when I started. I spent a week learning some basics, then a couple of months coding by hand, then discovered that AI can help with it, and now I've got a fully functioning prototype all up and running.

2026-03-01 19:03:28 UTC

- reply

I'll let you all know if I discover a proper cup of coffee in this country. Plus electrical outlets.

2026-03-01 18:49:54 UTC

- reply

Yes. It's better, but still small.

2026-03-01 17:29:23 UTC

- reply

That's where I'll be heading tomorrow actually. I'm planning to do a thorough search in the bigger cities I'll be visiting. Might also go to Vienna with that purpose.

2026-03-01 17:09:35 UTC

Nothing against Italy's cafe culture, but the drinks are all so tiny.
Are there any laptop-friendly places where one can get large coffees, maybe even refills, and sit for a few hours?

2026-03-01 15:38:01 UTC

- reply

May as well start collecting.

2026-03-01 15:35:43 UTC

- reply

With them not being minted any longer. May as well start collecting.

2026-03-01 15:31:59 UTC

- reply

Actually with my Inkan experiment nobody can control who registers, so there's no gate where spammers can be"manually" filtered out.
But it's still a way of whittling down the number of identities that one might want to treat as at least potentially legitimate. Of course, as with all such mechanisms, it can also work against high-quality accounts that cannot or do not want to pay. In the end it's just another signal that one may want to use, or not.
But more importantly to my mind, if people are able to establish long-term online identities, i.e. identities that they can be reasonably confident they can keep for say 30 years without the keys that secure the identity being compromised, they can then invest and build a more serious and less ephemeral online presence.
The hope is that people will then have more time and higher-quality material to work with when getting to know one another online. That may then lead to more organic growth of online reputations, with lesser need to use technical means to filter out spam.

2026-03-01 15:12:25 UTC

- reply

If you think it's safe to assume that, investing in bitcoin does make sense.
I agree it's safe to say that Bitcoin and gold will remain, although maybe for different reasons. There is something rationalized and ingenious about Bitcoin, whereas the value of gold seems based more on an emotional or maybe cultural attachment. But yes, both are there to stay.
It seems less safe to assume that fiat will eventually go away. And whether or not it goes away will affect the valuation of bitcoin and gold. Could be, but it doesn't seem like a safe bet. Then again unsafe bets can pay off big, high risk high reward.

2026-02-27 13:39:07 UTC

- reply

I think there is a definition of "asset" as something that has an "expectation of future [economic] benefit" associated with it. I kind of like this definition.
Applying it to Bitcoin, the *current* price in fiat isn't really relevant to whether its "worth owning," but the *future* price in fiat should be considered. For example, if the current price is low and the future price will be high, it's definitely worth owning.
There's a different question as to whether "price in fiat" is the most important metric here, if fiat itself is expected to lose value in the future. It may be better to use price in in terms of a bundle of goods, or CPI, or price in terms of real estate, forest land, etc.

2026-02-27 13:14:15 UTC

- reply

No quibbles about growing one's own potatoes.
With Bitcoin, there is the freedom aspect which makes it a valuable thing to support, including by investing in it and regardless of whether the investment leads to gains or losses. It's a different question whether, at any given time, an investment in Bitcoin can be expected to actually appreciate.
With gold, it sort of seems like an asset class like any other. It's valuable, but so is real estate, coal, cereal factories, fiat currencies, forest land and fashion brands.
The "no other options" bit doesn't fully convince. Is the idea that fiat currencies will lose value relative to crypto currencies? That could be, but it doesn't seem like a foregone conclusion.

2026-02-27 11:08:57 UTC

- reply

It's fine as a diagnosis, not sure about the details, but it's sort of plausible enough in a general way.
But the proposed solution isn't so convincing.
Buy bitcoin - before buying an asset for investment one ought to have a pretty firm notion that it will increase in value relative to other assets one might hold. Without denying the importance of bitcoin, I'm not sure whether that's a given.
Gold - same kind of thing.
Potatoes - I mean sure as a hobby, and it's a nice skill to have, but one can also buy them at the supermarket.
Wait out the death spiral - will there really be a death spiral, and even if there is one, until when does one have to wait and what advantage does one gain from waiting, and what does one do once it's spiraled out?
So maybe you think I'm just being obtuse and am deliberately misreading a bit of fun rhetoric, and I guess you'd be sort of right. But the point is that, behind the rhetoric, there doesn't really seem to be any real workable prescription here, or is there?

2026-02-27 09:03:03 UTC

- reply

There was some discussion with @npub13mj…8hs2 a few weeks ago about a proposal to use "verifiers," i.e. entities whose role it would be to vouch for other pubkeys. This would be done along some dimension of evaluation (e.g. "is human" / "is safe for work" / "is interesting" etc.) that can be defined and declared by each verifier. Users could then adjust their incoming filters (e.g. for replies etc.) to only permit keys that are verified by one of the verifier services the user has chosen to trust.
Another way (which I'm quite partial to) to separate wheat from chaff is to use identities that cost a few dollars. This is one of the beneficial side effects of Inkan's permanent identities - they overlay the nostr protocol with high-quality and indestructible"expensive" identities. Bots can of course also create such identities in principle, but it gets costly very quickly to do so on a large scale. At the very least, an identity whose creation costs money sends a strong signal that one might want to treat it as "serious".
Also, the permanence - i.e. the high confidence that the key pair that secures one's identity will never be compromised and one's followers will never be lost - is an incentive for the identity's owner to develop a high-quality nostr presence over time. This will result in a smaller proportion of accounts having an "ephemeral" feel and a greater proportion becoming real online personas in which people have made a real investment. In that sort of world, online identities can by default be taken seriously, reputations will be more organic, and there will be less need for technological solutions to manage visibility.

2026-02-09 17:09:01 UTC

Test2

2026-02-09 16:42:41 UTC

Test.

2026-02-09 16:16:48 UTC

- reply

Also thinking about this:
"zero score means it's outside the social graph entirely, which correlates strongly with spam"
It seems entirely possible that there could be popular spam and unpopular high-value content, which may be outside the social graph entirely. One could argue that this is often the case.
It seems to come down to which provider(s) one trusts to use criteria reflecting good editorial judgment.

2026-02-09 15:55:29 UTC

- reply

This is interesting, but it actually doesn't seem to validate my original intuition. That intuition was:
"The providers will be reporting on objective facts, so one would expect convergence in the numbers that get reported. That should then make it less necessary to place special trust in any single one provider."
I think you're saying that the numbers reported by different providers will possibly / likely not converge?
But why is that? The graph crawled from seed pubkey set A may differ from the graph crawled from seed pubkey set B, but as a provider won't you then aggregate the information you have gathered from both crawls?
I imagined that scores / rankings would be calculated based on aggregate information from all crawls across the seed sets that were used.

2026-02-08 02:50:00 UTC

It's called "artificial" intelligence, but what I find amazing is that these words are created by *natural* processes. It makes me look at a stone or a piece of plastic and wonder what else is going on inside it that I don't know about.

2026-02-08 02:25:33 UTC

- reply

Makes sense. I've not really had to contend with spam with my new account so far, but I guess this becomes more of a problem later.
From what I've learned at this point, the basic issue seems to be who can *contact* you.
You yourself can select the people you follow, so there isn't any obvious need for a technical solution here. As for curated content, you can choose providers just as you can subscribe to TV channels. Providers that generate automatic feeds will likely want to use a spam filter for their feed, but whether / how they do this is maybe not a central problem for nostr as a protocol, unless feeds are crucial to the experience you want to have.
So the remaining problem is then that one gets too much spam in one's replies or notifications. One could solve this by filtering replies and notifications by follow-distance, but that would close one off to the world at large. If you want to stay open for everyone to make contact with you, which is arguably desirable, you thereby open yourself up to spam. And that's where we need a technical solution. And it seems like your verifiers would do the job.