Deleted Account

Public Key

npub1a6we08n7zsv2na689whc9hykpq4q6sj3kaauk9c2dm8vj0adlajq7w0tyc

Profile Code

nprofile1qqswa8vhnelpgx9f7arjhtuzmjtqs2sdgfgmw77tzu9xankf87kl7eqpzfmhxue69uhhwmm59e6hg7r09ehkuegppamhxue69uhkumewwd68ytnrwgalzvrt

Publishing to

Author Public Key

npub1a6we08n7zsv2na689whc9hykpq4q6sj3kaauk9c2dm8vj0adlajq7w0tyc

Show more details

Last Notes

2026-02-21 18:21:00 UTC

And with this, I have an announcement of my own. Having left Haven in a good state, with an easy way to export data through JSONL to other relays, including the new work thatn@nprofile…r7ku is building, I’m retiring this npub and my presence on Nostr Kind 01 permanently.
At this point, Haven is robust, battle tested software used by hundreds of relay operators. I’m sure that, other than for potencial incompatible changes to the Nostr protocol, it should remain solid for, well, longer than it has even existed.
The V2 / nostrlib version that I often spoke about will not happen, at least not for now, and likely not under mine or utxo’s guidance. But, as utxo said, I have been working hard to give users most of the features they asked for before putting Haven into maintenance mode. Well, it is finally time... I will, of course, still keep an eye on GitHub for bugs, and I’m sure that the many developers who forked and contributed to Haven will maintain their forks and carry the torch. Maybe they will elect a new official fork, or at least keep their versions alive until users are ready to move on to the future vision that utxo, Fiatjaf, and other Nostr developers are building. At Nostr’s pace, this should not take long anyway.
Before I NIP-62 and NIP-09 my Nostr presence out of existence (at least on well behaved relays willing to respect my wishes), I just wanted to leave my contact details for anyone who would like to stay in touch.
I’ll keep my presence on ActivityPub / Mastodon, although I do not post there much:
https://accioly.social/@anthony
I’m also still on @nprofile…7nrg’s noirc.net (nick: anthony) and in some Libera.Chat (aaccioly) channels. If you are an IRC enthusiast, I’m happy to chat there.
Finally, I’m on Signal / Molly if you would like to speak directly:
https://signal.me/#eu/cDGXbPN8BUyAyyhJ4LjhTxdftVPixKjvZ-IDdrN-QhTrPYC3lotILLNJE8RKlHom
I’ve met some incredible people on Nostr and would genuinely appreciate ways to stay in touch outside of it.
#accountRetirement #shrinkNostr?
#nevent1q…e2n5

2026-02-21 17:45:11 UTC

And with this, I have an announcement of my own. Having left Haven in a good state, with an easy way to export data through JSONL to other relays, including the new work that @nprofile…r7ku is building, I’m retiring this npub, and my presence on Nostr Kind 01, permanently.
At this point, Haven is robust, battle tested software used by hundreds of relay operators. And I’m sure that, other than ror incompatible changes to the Nostr protocol, it should remain solid for, well, longer than it has even existed.
The V2 / nostrlib version that I often spoke about will not happen, or at least not at the moment, ans likely jot under my or utxo’s guidance. But as Utxo said, I have been working hard to give users most of the features they asked for before putting Haven in maintenance mode. Well, it is finally tie :). I will also keep an eye on GitHub for bugs, and I’m confident that the many developers who forked and contributed to Haven will maintain their forks and carry the torch. Maybe they will elect a new official fork, or at least keep their versions alive until users are ready to move on to the future vision that utxo, Fiatjaf, and other Nostr developers are building. At Nostr’s pace, this should not take long anyway.
Before I NIP-62 and NIP-09 my Nostr presence out of existence, at least on well-behaved relays willing to respect my wishes, I just wanted to leave my contact details for anyone who would like to stay in touch.
I’ll keep my presence on ActivityPub / Mastodon, although I do not post there much: https://accioly.social/@anthony
I’m also still on @nprofile…7nrg’s noirc.net (nick: anthony) and in some Libera.Chat (aaccioly) channels. If you are an IRC enthusiast, I’m happy to chat there.
Finally, I’m on Signal / Molly if you would like to speak directly:
https://signal.me/#eu/cDGXbPN8BUyAyyhJ4LjhTxdftVPixKjvZ-IDdrN-QhTrPYC3lotILLNJE8RKlHom
I’ve met some incredible people on Nostr and would genuinely appreciate ways to stay in touch outside of it.
#accountRetirement #shrinkNostr?
#nevent1q…e2n5

2026-02-20 12:25:03 UTC

@nprofile…fj80, morning mate, how are you? (Just doing NIP-01 as apparently NIP-17 is more broken than usual.) I was having a chat with @nprofile…dgdk the other day and started thinking: each one of us is maintaining a different SQLite driver for nostrlib/eventstore. What do you think about us joining forces and creating the ultimate super-duper SQLite3 driver? I’m happy to combine the features of all three repos into an initial version myself. I’m mainly looking for other folks who use it in their relays to help maintain it, as the more of us working on it, the better the driver will become and the more we can share the maintenance burden. I want Haven V2 to default to it with proper sane backwards compatible migration. Badger is gone from nostrlib, BoltDB sounds like more of the same and even LMDB will require a non backwards compatible migration. Thoughts?

2026-02-17 00:37:11 UTC

- reply

@nprofile…jgy8, @nprofile…fzvp, @nprofile…m55g and last but nost least @nprofile…8ltw (many thanks for your contribution mate)

2026-02-17 00:33:29 UTC

- reply

@nprofile…jgy8, @nprofile…fzvp, @nprofile…m55g and last but nost least @nprofile…8ltw (many thanks for your contribution mate)

2026-02-13 23:56:34 UTC

@nprofile…xk05, am I welcome on your WoT Relay? :)

2026-02-07 16:18:09 UTC

- reply

No problem, Utxo. Thanks for the zap, and I hope you are doing well!
I’m also working on improved backup and restore to finally move v1 into maintenance mode, and will reach out so we can decide what we are going to do about v2. Life got in the way since we last spoke but things are slowly starting to progress again.

2026-02-07 16:08:26 UTC

- reply

CC: @nprofile…fzvp and @nprofile…vvjh

2026-01-03 19:49:45 UTC

- reply

Thanks, Oberon. I’m unfortunately sort of done with all the "getting used to" after more than a year around here. There isn’t much left on NIP-01 once I filter out low quality posts (though I’ll miss Japanese Nostr, some of the Nostr devs, and a few artsy folks). That said, there are some brave, persistent people trying to raise the bar on Nostr, buried beneath mountains of BTC L2 marketing posts, bots, GAB-like stuff, etc. Unfortunately, I don’t see this changing despite their valiant efforts with hashtags, custom relays, and so on. Still, have a look at my follow list, there are some gem npubs there, even though a lot of them have also quit Nostr.

2025-12-29 15:18:01 UTC

- reply

Agreed. I’m all for people unionising and for businesses being punished for union busting. I’m also against people being forced into unionising or being extorted into paying membership fees without real representation. Funnily enough, this is a posture that tends to annoy authoritarians on both the left and the right, as it makes too much sense. Unions can be a good thing, and when they are a good thing, people don’t need to be forced into unionising. Ask Brazilian finance professionals or US SAG-AFTRA members if they want to give up their union membership.

2025-12-29 14:56:20 UTC

- reply

Thanks for talking some sense and baiting some of the bad apples on Nostr to out themselves, @nprofile…6u4e. Nostr content is getting so tiresome...
Fun fact: while you were taking a break, I got some heated DMs basically calling me a hypocrite for collaborating with you and Bitvora while calling this shit out on Nostr. Apparently, they took you for one of the racist / North American MAGA / authoritarian-dressed-as-libertarian sheep on Nostr. And Bitvora for one of the crappy, spammy Lightning service providers on Nostr (after I blocked a couple from spamming the relays I’m running and returned a few of their bribery attempts).
For the folks who need a reason for me to volunteer my time (why would anyone need a reason to maintain good software?), that’s why. Don’t take a good man, a builder, a second-generation Canadian, for one of the low-IQ, xenophobic racists loitering here.
Signed: a Brazilian-born, UK citizen who has contributed more to this country in the past ten years than all of the Tommy Robinson followers on Nostr combined. I never claimed any benefits, not even during unemployment (and yes, things got pretty rough for a while). I built two businesses from the ground up, volunteered quite a bit, and was subjected to the 60% UK tax trap for several years while paying extra fees like the NHS surcharge. Tax-wise, I am sustaining an average of at least half a dozen of Tommy Robinson’s “patriots” on the dole while they are busy coming to London to riot and paint roundabouts red. The only thing I have to say to the "patriots" out there is... Piss off! Stop complaining about the immigrants putting food on your table and get good. Real patriots should, you know, do something for their country other than seeding hate and trying to start riots.
You folks voted for Brexit. You folks did nothing while the corrupt politicians you voted for sold out. You folks let private equity take control of the country, including real estate. You are the ones about to elect Farage and swallow Musk’s fascist rhetoric of “violence is coming for you” while letting the US and Big Tech aristocracies weaken Europe and ruin our way of life. I have nothing to do with the UK’s decline. You do. Stop blaming immigrants for the consequences of your choices. Also, be careful what you wish for. Unlike you, I have real mobility (hint: moving abroad is nowhere near as easy as you think), so please take some responsibility and stop destroying the country that I love and that you are trapped in. Take some responsibility before it is too late.
PS: As a very white person (yes, there are lots of people of European descent in Brazil, mostly children and grandchildren of European immigrants fleeing their countries during difficult times, so do remember that tables turn), I can say that “Arabs”, “Asian”, and “Brown” people are doing very well, thank you. In fact, most British “patriots” I know would move to places like Dubai and Qatar without a second glance if they had the opportunity to.

2025-12-28 14:49:02 UTC

- reply

Primal did, but compatibility between Blossom servers is still a bit hit and miss as some fair play is required between clients and both servers involved.

2025-12-16 10:14:36 UTC

- reply

The main problem is that a lot of Lightning, ecash, etc. wallets allow you to associate an nsec to unlock the wallet. Not many people pasting their nsecs into random vibecoded apps are security-conscious enough to use a separate Nostr key, so if one of those vibecoded apps leaks nsecs, you can pretty much scan the respective lud16 for half a dozen popular wallet domains and, more likely than not, hit the BTC jackpot.

2025-12-15 22:46:50 UTC

- reply

So, with this version I can log in using Citrine and the bunker url flow, but still can't with the nostrconnect flow. Amber window now closes, but it seems like things get sorta stuck after that, I.e. clients keep struggling to load timelines, etc.

2025-12-15 22:08:01 UTC

- reply

We unfortunately have 200 other problems to solve besides NIP-47, starting with Nostr content and the attitude towards non-bitcoiners, but this is a big one for sure. I don’t think most folks are ready to move to totally ephemeral pubkeys, and reusing nsecs across experimental, increasingly vibecoded clients is just... dangerous... Nostr really needs to get the basics done right. As a community, we have wasted so many opportunities to scale Nostr already.
I joined about a year and a half ago, during one of the many huge waves of people quitting Twitter. Since then we have lost at least three other opportunities to grow. The last one was the hype around @nprofile…62p3’s deVine stuff. Every time one of the non-BTC Twitter experiments shows some promise, Nostr seems to crap its pants.
We really, really need to get the basics right for Nostr to have any chance. I don’t think lightning will keep striking the same place over and over again, and we may not get many more opportunities to make Nostr scale.

2025-12-15 21:04:07 UTC

- reply

This one seems very likely to be a bug on Amber by elimination (i.e. everything works together as long as Amber isn’t in the flow). Nak can act as a bunker on its own, but it’s a major hassle to use it on mobile, as it means having my VPN on and SSHing into my linux box to retrieve bunker URLs when I’m outside my network. Plus, running Amber and Citrine locally is the only way I know of for remote signing to work even relatively well, as NIP-46 is extremely sensitive to latency. Even the latency of a WireGuard VPN is enough to degrade the experience, with Amethyst becoming absolutely sluggish, almost unusable.

2025-12-04 13:37:23 UTC

- reply

Good to see you back Beave. How's the whole VPS + Wireguard + self hosted relay combo doing? Any progress or has work kept you busy?

2025-11-25 19:01:40 UTC

Test after golang.org/x/crypto library bump to v0.45.0

2025-11-24 23:04:17 UTC

- reply

I know the feeling. As usual, Brazil mimics what happens in the western hemisphere with some delay. Outsourcing pressure postponed the layoffs in Brazil for a bit, but now the big banks, big tech and startups are playing the same game of laying people off and trying to rehire cheaply as they did here. We need to brace for the next few years. Things are starting to get better here, but it was a total nightmare, especially for IT folks, and even now the market is nowhere near its peak. But I’m honestly starting to see light at the end of the tunnel, so hopefully things will also improve in Brazil in due time.

2025-11-24 22:56:28 UTC

- reply

So, same as always? 🤣
PS: I'm building some backend stuff (not Android) with Kotlin, http4k and Quartz. Quartz is quite nice, even from the perspective of a relay dev. I like where you folks are going with it. I'm very tempted to expand on Citrine or even work on a non-mobile Kotlin relay (well, I'm actually already doing it, but it's hyper specialised and basically not useful to anyone else but me at the moment). Is you, or is there anyone you know of who is already working on something like Khatru in Kotlin, i.e., a Relay side library / framework? (CC: @nprofile…xcpk)

2025-11-24 22:35:21 UTC

- reply

I hope you’re doing well, sudocarlos. You will soon, and things will get better for sure (they’re slowly getting better for me :)). Also, as someone who once jumped straight to the FU stage without really having the FU money and had to live with the consequences… Don’t do it if it can be avoided. Even if it looks like you have clients and a "validated business" on the side, only pull the trigger when you actually have the savings to back it up or when your side business is taking up so much time that your main job has become a bottleneck. That’s the real validation.

2025-11-12 00:23:28 UTC

- reply

Side Pocket was the GOAT. Virtual Pool on PS1 also had a great sound track:
https://youtu.be/BLTJPLdWY4Y

2025-11-11 15:34:18 UTC

- reply

Dor some reason on Amethyst I only saw your own answer after I commented, lol. jq is love, jq is life.

2025-11-10 13:49:32 UTC

- reply

Aye, just to make myself 100% clear, I’ve got nothing against Keychat, and I can see how what you’re building would appeal to a lot of folks on Nostr. I’m also very much on board with the model of creating revenue for relay runners instead of, you know, zapping them 100k sats "prizes" to try and create goodwill. I heard nothing but praise from Nostr users towards Keychat so far.
That said, it’s not something I see myself using. Signal covers most of my chat needs. @nprofile…nggv got me back on IRC (noirc.net), which I’m enjoying for “dev” talk that doesn’t need to be private. If I ever need something truly trustworthy and private, I can always go back to self-hosting an XMPP server within a properly secured self-hosted environment behind a VPN.
Nostr devs as a whole won’t come together anywhere regardless; each will insist on their own NIP-29 or NIP-29ish approach and mostly just refuse to collaborate. So even if you nail the UX, revenue model, and everything else, that’s why I said that from my perspective, as amazing as Keychat can potentially be, given human behaviour, it will likely remain a niche tool for BTC and eCash enthusiasts.
Don’t get me wrong, though. I’m genuinely rooting for your success and for anything that strengthens user privacy in the messenger space. And extra kudos for building a model that properly discourages spam.

2025-11-10 00:19:01 UTC

- reply

I respect your opinion but couldn’t disagree more. Actually, I’m on a one-man war against Nostr’s real, palpable incentives.
People are just bad at prioritising what truly matters. Most people on Nostr will give grand speeches about freedom, anarchy, and building your own roads, then proceed to zap bitcoin memes and behave exactly like they would on Twitter, Gab, etc. On average, the palpable incentive is to keep Nostr as a safe space for the Bitcoin Twitter bubble.
Funding developers who aren’t aligned with, onboard with, or sponsored by the BTC Twitter use case is what Nostr needs in order to evolve. This goes against the same incentives that brought many of its users here in the first place. To stay relevant Nostr needs to outgrow its early-day priorities for the greater good.
I don’t use Keychat, and to avoid falling into the human tendency of criticising what we don’t understand, I won’t say a word against it. That said, based on the project’s own landing pages, their scope seems to be to become the "everything app" for Maxis. So, by definition, I can’t see it as much more than a beloved niche tool.
I’m not disagreeing that ultimately Nostr may exist as an experimental protocol for Bitcoiners, a transitional evolutionary step toward something better yet to come.
That said, I don’t think my points above are wishful thinking or addressing some utopian, unsolvable problem. We just don’t like to face the reality of our real problems. As I like to say, there’s no technical solution to moral problems. Ultimately, people have to choose... And for the vast majority of Nostr users, there’s a lot of theoretical talk and very little to show that they’re ready to actually build and fund a real alternative.
By definition, they’ll either buy or be the product (here or elsewhere) regardless of the technology powering it. Once people truly understand this (and I mean really understand it, beyond grand speeches and doomsaying), hopefully healthier palpable incentive like owning your data and controlling your feed will be taken seriously. And there's no other way to take it seriously other than voting with your wallet and time.
As hard as the technical and organisational problems are (and yes, Nostr has a bunch), the real challenge is getting people to truly understand why we should build something like Nostr, and to have genuine skin in the game.

2025-11-08 15:43:48 UTC

As someone who used to spend a lot of time with ROI-obsessed bean counters controlling, well, every budget on 🌎. Make the magic conversion metrics plummet and you kill the main incentive driving companies not only to slurp up everyone’s data, but also to keep you addicted to maximise the chances of you clicking on an ad. Turn Google’s, Meta’s, and X’s own game against them.
If they turn off the data-scraping bots taking down FOSS infrastructure, I might even consider disabling my brand new "ad-clicking" plugin.
https://youtu.be/7GeCq1qwqjc
#ADS #Privacy #BrokenInternet

2025-11-06 18:49:23 UTC

- reply

Nah. Java / Kotlin / Scala / .NET Senior finances or telco folks. The keywords are distributed systems, risk, arbitrage, low latency, etc (and it works basically as well in Investment Banking and Web3 stuff). Some Architects and Data People. McKinsey pay for MBAs, not technical folks.
Ah, come on, I'm sure I donated a few sats to you myself. So you are what :). Just a few million or tens of millions sats in the red right now lol 🤣.

2025-10-31 21:20:40 UTC

https://youtu.be/qw4fDU18RcU
I'm not gonna lie: Running a 10 GPU homelab rig capable of running a 240 billion parameters lAI modelI Leading a protein folding research team (Folding@Home is worth checking out by the way), hosting a council of evolving agents, building his own AI frontends... This is all beyond impressive.
PewDiePie went from "Look at me, I’m running Linux and ricing my desktop" to a quite impressive homelab and experiments... And he has done all of this in what, 6 months? Maybe a year?
You can like him or not, but I can easily see many of the kids (well, former kids) who grew up watching his Minecraft videos getting inspired by this and building amazing things And this is coming from someone that isn't exactly bullish in the whole "AI revolution" thing.
He has earned my respect, and I hope he and his agents eventually take some interest in decentralised social media.
How about you, what are you building right now?
#homelab #diy #selfHosting #ai

2025-10-26 22:46:45 UTC

- reply

Will have a look, thank you.

2025-10-26 22:00:57 UTC

- reply

@nprofile…hk4d, just pinging you again. Also @nprofile…fae9, (hi, how are you)? Did you get to migrate your relays to nostrlib or are they on go-nostr and Khatru? (I'm just doing a "survey" on the current status of the Khatru ecosystem).

2025-10-26 17:30:15 UTC

- reply

Fair enough again. I feel the "Software People" exhaustion even at a relatively "young" age. It is good to know that there's a path to reading and naps :).

2025-10-25 14:02:41 UTC

- reply

This one is already pretty "mainstream", or as close to it as a CLI transfer tool generally gets (almost 32k stars on GitHub; I’m not sure the whole Nostr ecosystem combined gets that much attention).
But fair point regardless, especially for folks using public relays. Just like in Nostr, sending some money to public relay runners (like @nprofile…6u4e, @nprofile…dx2g, @nprofile…nggv, @nprofile…hfq7, etc.) is a good way to keep their respective public relays alive.
Likewise, sending some $$$ to FOSS developers is the best way to guarantee they actually get some development time.
This is something that folks (even those who say they’re all about "value for value" like here on Nostr) are really bad at. If only users knew how expensive it is to develop and run this sort of thing, they’d probably pool together an annual budget to split among the FOSS devs building the tools they rely on.

2025-10-23 19:38:06 UTC

- reply

Thanks for your service Dan. Out of curiosity, how's the relay / Blossom side of things going. Have you migrated to nostrlib already?

2025-10-23 08:41:45 UTC

- reply

Vegetables in red meat shape would be popular on Nostr 😂😂😂

2025-10-21 18:32:27 UTC

- reply

I assume you’re Irish, correct? How cold is it in Ireland right now? Plenty of Nazis there, same as in the UK. The Justin Barretts and Scott Woods of the world are doing just fine. There’s plenty of money going into making neo-Nazism mainstream, and plenty of dumb people here in Europe who think they can’t find jobs or afford rent because of immigrants (while the very people sponsoring these neo-Nazism movements, Elon Musk included, are behind the collapse of their countries).

2025-10-21 10:18:19 UTC

GM folks. What will last longer: $LIBRA or the 20 billion dollars of US taxpayer money just sent to Javier Milei?
https://haven.accioly.social/326df6d90765165b242f39d1d9bc2c45d795d25d9d02364f2d76649524d775e6.png
Argentina deserves better; better than Peronism, better than Milei. Just like Brazil, Argentina could be a global superpower with better leadership and longer-term thinking. Can we be done with self-serving leaders and self-serving billionaires for a change?
#GM #Argentina #chainsawPolitics

2025-10-19 18:48:16 UTC

- reply

In most big companies in the UK, it works like this too. But it only accepts a UK sort code + bamk account number (i.e., proper banks or bank-like fintechs; not crypto exchange accounts, as far as I know). I’ve also never worked for an employer whose payroll system lets me split my salary, though it might be possible by calling HR / Payroll team.
For people in investment banking, you’re often required to pre-clear crypto transactions before buying or selling; just as you would with stocks, ETFs, bonds, etc.

2025-10-19 17:54:53 UTC

- reply

Like ask your employer do split it for you and deposit directly on am exchange? Very unlikely in the UK, unless you are working for a Crypto / Web3 company.

2025-10-18 20:18:55 UTC

- reply

Ainda está um pouco lento e a previsão de custo está acima do valor real, mas está fundando. Agradeço a contribuição.

2025-10-11 12:23:33 UTC

- reply

This is amazing Girino, great stuff! My own version of Haven (not on GitHub for now) is also running a custom build of SQLite in WAL mode, with better indexes, json1 functions, improved upserts for replaceable events, and a few other tweaks.
I didn’t denormalise data like you did (still one big table) and am still exploring concurrent writes (https://www.sqlite.org/src/doc/begin-concurrent/doc/begin_concurrent.md). Basically, I’ve reached the same conclusion as you: I’m getting lower latency and higher throughput, even at levels of concurrency that would be unrealistic for personal relays. For large WoT relays write contention can become an issue at some point; but that’s what PostgreSQL and LMDB are for.
Maybe in the future we can join forces and publish a solid SQLite "eventstore driver" compatible with nostrlib ("go-nostr + khatru v2" if you’re not aware of it)? That seems to be the Fiatjaf-approved way ("write your own eventstore drivers, mere mortals"), and it would give us more control over migrations.
I honestly don’t want to migrate Haven users from Badger to BoltDB, even if it would be the path of least resistance to adopt nostrlib.
@nprofile…6u4e (I haven’t forgotten about you or our chat; just absolutely swamped with real-life stuff, apologies). I’m more and more inclined to thing that a custom SQLite driver is the way forward for the whole nostrlib migration effort. Maybe even a PostgreSQL driver for the algo relay. Old-school SQL is officially the new cool 2025 Nostr dev trend (and I'll keep repeating this until it becomes reality lol). :)
#nevent1q…8mff

2025-10-09 01:48:14 UTC

- reply

Thanks for your help here Turismo.
@nprofile…w35w – apologies for bothering you, but any clues here?
accioly.social/.well-known/lnurlp/anthony is just redirecting to walletofsatoshi.com/.well-known/lnurlp/uniformbakery871
With my former custodial wallet, zaps would display correctly. Now, while I can still receive sats, the zap value isn’t showing up on Nostr. Is there anything special about Spark wallets that might not play well with Nostr?

2025-10-03 15:59:09 UTC

- reply

If its like mine then you aren't getting Google results. The fact that I didn't even realise that SearchXNG wasn't there is a good sign that maybe I don't need Google search anymore. I switched Google off for now, lets see how it goes.

2025-09-29 14:52:20 UTC

- reply

On a scale from Canadian/UK “spicy” Indian/Mexican/Thai/Sichuan food to the real thing, where do you fall?
The problem is, it’s hard to please everybody. Fake Brit “spicy” food is already too much for my wife… and I can barely feel any heat. Meanwhile, anything close to real levels of proper Mexican (not Tex-Mex) spiciness means I’ll be spending a couple of days at the loo, spitting fire out of my behind. Apologies for the graphic scene 🤣🤣🤣

2025-09-26 02:34:20 UTC

- reply

#nevent1q…yhkq

2025-09-26 00:32:28 UTC

- reply

Ok, this one I can get behind.
https://media.tenor.com/miUhtY-7N3MAAAAM/engage-startrek.gif

2025-09-23 08:53:57 UTC

GM Nostr! In the unlikely event there’s another Nostriche at DevOpsDays London 2025, give me a shout.
https://haven.accioly.social/720dacc47926625bf8941936571b1352f40f31b64c473f8666676370807f2526.jpg
#GM #DevOps #TechConference #London

2025-09-23 06:27:05 UTC

- reply

I get the feeling. And thanks for your "Monday Adventure" on top of everything you already do for Nostr.
GitHub’s algorithms are quite messy and often go against their Terms of Service. I ran into issues myself after creating a machine account, which is actually allowed under their ToS. Luckily, after going through the appeal process a few times, a human eventually reviewed it and resolved the problem. But GitHub is broken in several other ways; death by a thousand papercuts.
Dan Conway is doing great work for sure. Honestly, some of the coolest tech on Nostr.

2025-09-23 00:39:49 UTC

- reply

GitHub is quickly becoming unusable, just like LinkedIn and everything else Microsoft has acquired. I really hope Nostr's adoption of GRASP / NIP-34 gains momentum. My Forgejo instance has everything I need really. I'm slowly but surely moving away from GitHub.

2025-09-17 16:49:04 UTC

- reply

Oi Hertu, olhando seu perfil eu acho que você é brasileiro, confere? Fique a vontade para escrever em português se quiser. A gente se entende ( translate.kagi.com/ salva :)).

2025-09-10 14:29:33 UTC

Who is the developer (or developers) behind @nprofile…8zqj? Is it open source? If so, where can I find the code?
#asknostr

2025-09-10 09:41:26 UTC

- reply

GM Logen. Wise words!

2025-09-08 17:17:29 UTC

- reply

Interesting thread. Thank you.

2025-09-06 21:19:23 UTC

- reply

Run your own relay, your own Blossom server (or NIP-96, or whatever). Run your client locally. Block all Cloudflare IP ranges across the board and be happy.
Yes, you will break several clients. Yes, folks can still broadcast or mirror your stuff to one of the gazillion Cloudflare-enabled relays and media servers out there. But overall, there’s not much anyone can do to fully "opt out" of Cloudflare on the public internet.

2025-09-06 01:25:57 UTC

GN Nostr. Today (well, yesterday) was a good day 🛌😴🌙

2025-09-03 18:15:13 UTC

- reply

I honestly don’t know. User activity levels vary substantially, there are different kinds of notes beyond just short and long text, and a single black swan event such as “Reply Guy” or even a misbehaving client can drastically change the outcome. Maybe folks running WoT relays, such as Utxo and @nprofile…4t0k can give you a better idea.

2025-09-01 23:37:32 UTC

- reply

Obrigado pela lembrança, Girino! @nprofile…266u I'm happy to have a chat, feel free to DM me. Even if this one isn’t for me, I can probably at least point you in the right direction.

2025-08-28 22:12:01 UTC

This is so cool!
https://dev.nosotros.app/feed?relay=wss%3A%2F%2Frelays.land%2Fspatianostra
This is exactly the kind of experience I wish Nostr users had right after onboarding.
Actual human curation >>>> algorithms, trending posts, most zapped, etc.
#nevent1q…s4gr

2025-08-28 21:48:45 UTC

- reply

This is actually a great idea! A curated, "democratic" relay feed that clients can default to when onboarding new users without followers. What’s the criteria for an account to be able to vote?

2025-08-28 21:43:15 UTC

- reply

Apologies, my server went down and I didn’t receive a notification for your question.
I have… a lot to write about this, and it’s probably a better subject for a long-form article or another thread. But the gist of it is that Nostr is currently in its exploration phase, with an exploration culture. Will it, at some stage, find the right combination of opportunity and maturity to surf the next big wave and reach the expansion phase, where scalability, reliability and maintainability gain more space? I don’t know. We’ve already had two or three possible opportunities thanks to Musk, Zuck and the gang messing up traditional social media for normies. We were unprepared to capitalise on those opportunities and grow Nostr back then (and if you ask me, we’re still unprepared now).
That said, we have very smart people here thinking about all of this, so who knows? Maybe someone will build a very popular client that drives folks to Nostr on its own merits. Nostr’s culture, financial incentives (or lack thereof), architecture and social contracts aren’t set in stone. They can adapt. I dare to say, given a huge influx of users, clients will either adapt or become incompatible forks like the "defederated" stuff on ActivityPub based social media.
Maybe the next time big tech screws up, Nostr will be ready to capitalise and skyrocket to ~40m total and ~4m daily active users like Bluesky did. Or maybe it will stay small and eventually servr as a stepping stone or at least as inspiration for an even better protocol, like SSB that you mentioned above. I really don’t know. One way or another, it’s good to be here for this “experiment” we call Nostr.
#nevent1q…7chs

2025-08-25 10:24:46 UTC

- reply

I don’t think our opinions will converge on this one, unfortunately. You can still do all of this on public relays and with DMs. Plus, IMO, spamming new users with ads about Lightning wallets, even with the intent of “helping them set up a wallet” and sending them some sats, is a good way to shorten the 5-minute experience to 3.
The number of people I’ve seen installing something like Alby (partly thanks to welcome bots) and then realising they either need to buy a cloud subscription or set up a node, pay to open channels, manage liquidity, and only then actually receive their welcome sats is not insignificant. Well-intentioned or not, the most likely outcome is that the user not only leaves Nostr but also tells their friends the whole network is a giant crypto scam.
I’m not sure what Ritzful’s model is here, but I’ve quickly grown allergic to welcome bots on Nostr because of this sort of thing. Lightning requires users to at least have some basic understanding of what they’re doing (and self-custodial Lightning specifically requires some upfront investment and a lot of knowledge).
Also, it’s not like your bot is pointing new users to a curated list of wallets. These are ads for your wallet and Lightning products. And that’s fine, but again, it can be done via DMs on public relays.
The wallet tester service is useful for existing non-functional Lightning addresses, but again, I think DMs are a better place for reports. If the intentions here go beyond advertising, I’d build in some optional user initiated invocation by mentioning the bot in a comment tagging the user. I’d even pay some sats myself for invoking this service (e.g. I pay you folks 50 sats, you zap the user 25), since when zaps fail on my end I often don't know whether it’s a liquidity/connectivity issue between or if the other user’s Lightning infrastructure is actually down. But, IMO, the two keywords for making this truly useful and good for Nostr are consent and user-initiated flows.

2025-08-24 19:45:56 UTC

- reply

Got you. I would still prefer Rizful to DM users and stick to public relays. You could also offer an opt-in zap test service that users can invoke by mentioning the bot, Reddit/Discord style. What I find annoying are the unsolicited comments.
At the very least, let users give consent and show interest in Lightning before the bot reaches out to them. IMO, this would still give Rizful a fair amount of brand visibility and reach the vast majority of npubs.

2025-08-24 19:26:08 UTC

- reply

I couldn't see messages from the other npub notes as well before you and Carlos replked (this one is clearly human, so the company behind the bots is at least monitoring mentions which is already better than a lot of the other "welcome to Nostr" stuff out there). Ultimately its up to each relay operator to decide what is useful or not. The only reason I even saw the bot messages is due to Amethyst new Outbox model. I personally find all and any of those welcome bots, wallet detection bots, auto zapping promo bots and variations of the above a nuisance. I try my best to curate my own Nostr feed for human content. But I'm not trying to tell any of you how to run your relays other than commutating that to me they are a nuisance.

2025-08-24 16:35:40 UTC

- reply

@nprofile…zq3c - already reported it.

2025-08-24 14:38:26 UTC

- reply

@nprofile…7c63, @nprofile…4zgq, @nprofile…kejf - just FYI, annoying bot made it to your respective WoT relays.

2025-08-22 17:11:41 UTC

- reply

It is, for sure. I can’t force myself to send a PR detailing my “cool, PGP-inspired local ownertrust + TOFU WoT model” just to have it ignored in the official NIPs repo. But I might consider sending a patch that links to my own repo’s documentation/whitepaper and proof of concept.
Nostr is a fun and very unique experiment, and I think folks have built something special. But… governance is still there. It can be implicit and chaotic, but ultimately someone (or some group) will always have to decide if a patch gets merged. And in the absence of a more formal process, things tend to work through informal networks of trust and mutual interest.
This isn’t wrong by any means, and it works especially well in the exploration phase. But it has limitations when moving into the expansion and extraction phases: https://medium.com/@kentbeck_7670/fast-slow-in-3x-explore-expand-extract-6d4c94a7539
I’m highlighting communication because, in my opinion, Nostr is struggling a bit with switching gears. It’s still early, and exploration is fun. Personally, I’d like it to stay this way for a while. But we’ve missed a few opportunities where Nostr could have really taken off if the right people and resources had been directed to the right things. For example, about a year ago, when a huge wave of people joined after X screwed up. I don’t know when or if we’ll see the next big influx of users, but I hope we do, and that Nostr is a bit more mature by then, so we can actually work on retaining those users instead of telling them to touch grass like we did back then 🤣.
(To be fair, the caveat is that I sorta agree with you, even the most wrll structured, truly open project can’t steer the ship as fast as a startup like Bluesky. Plus, I think Nostriches have ethical priorities that trump growth. Still, we could try to be a bit more organised, responsive, and user focused if another big wave ever hits Nostr shores)

2025-08-21 20:01:56 UTC

- reply

Thanks again. So, if it’s all yes, then in practice Gossip is also trying to connect to broken relays, right?
What sort of timeout settings and retry logic does Gossip currently use? @nprofile…dvuv also mentioned that broken relays aren’t as much of a problem in Amethyst as they are in Haven. I get the feeling that each of us may be handling this differently.
Haven tests relays during startup. It also uses a Penalty Box: if a relay fails to connect, it retries after 30 seconds, and then exponential backoff kicks in. Out of curiosity I put a counter on it, and with my current implementation this meant ~110k failed connection attempts in the first 24 hours after restarting the relay (though as exponential backoff kicks in the number of connections reduce substantially). Of course that I could make the "broken relay detection" converge faster by tweaking exponential backoff, but this also means that the next time that one of the popular arrays go offline Haven will quickly give up on them.
This is where a good blacklist could be useful. You can do this without fully trusting the list as well. I.e., if I can use Nosotros list above, still try to connect to all relays during initialisation, but if the connection fails and the relay is in the "black list", then I set a much higher time until reconnect for the exponential backoff algorithm (e.g. 1 hour or even 24 hours). So if the list is wrong or one of the dead relays gets "ressurected" Haven will eventually connect to it, but it makes the whole process much cheaper.

2025-08-20 16:36:47 UTC

- reply

Tell me you’ve got someone to sponsor your and Tony’s work, please. I know I’m not a great example myself; if I were charging for my hours of free Nostr development, you could easily add another zero to the sum you mentioned and it still wouldn’t cover it. But I’m genuinely concerned about how much Nostr is costing my fellow OSS devs.

2025-08-20 08:13:32 UTC

- reply

Apologies again, but just to remove any ambiguity: when you say "for picking your own relay", what exactly does that mean? Is this something only for you (Mike) to filter writable relays? How is Gossip actually using the resulting list of vetted relays for its users, if at all?
For example, if I have a brand new, non-
NIP-11 compliant relay with a non-standard path tagged in my Kind 10002:
```
["r", "wss://brando-relay.com/inbox", "read"],
["r", "wss://brando-relay.com", "write"]
```
This is the first time that Gossip encounters brando relay. Will it:
1. Write my own kind 1 notes to wss://brando-relay.com/?
2. Write other users’ notes tagging me to wss://brando-relay.com/inbox?

2025-08-19 22:04:41 UTC

- reply

Hey Mike,
Thanks for coming back on this one. Just so I’m clear on what we’re discussing, given the sophisticated heuristics here: are you talking about choosing “default” Outbox relays for new keys? Or is Gossip ignoring write relays from users that don’t match the above criteria? For example, if I manually set a write relay that has a broken NIP-11, will Gossip respect my choice, or skip the relay I configured myself when I publish my own events?
Either way, despite your comment below about not wanting shared blacklists, it feels like every client going through this process to compute lists of dead relays is a lot of duplicated work and wasted effort. I understand the value of having stricter criteria for suggesting default relays to new npubs and potential fallbacks, but I wouldn’t expect every client (or relays needing to calculate WoT and other advanced functionality) to handle all of this aggregate data collection and processing.
IMO a shared blacklist (optional for clients to use, and that they are of course free tk verify) seems like a more pragmatic solution.

2025-08-19 12:41:58 UTC

- reply

IPv6-only land is a good place to be :)

2025-08-19 12:40:42 UTC

- reply

Pesky persistent bug. Go to settings and click "Delete local cache". v3.4.6 should show up.

2025-08-18 23:42:53 UTC

- reply

I didn't know that sandwich was behind nostr.watch :).

2025-08-18 23:37:27 UTC

- reply

I didn't know that sandwich has behind nostr.watch :).

2025-08-18 23:36:06 UTC

- reply

Yeah, same problem here (I'm reading Kind 10002 and NIP-05 relays for experimental WoT stuff). I bet we're not the only ones running into this. Calling ensureRelay on each of the formerly popular relays is just a huge waste of time. I bet a lot of Outbox enabled clients are facing the same issue. @nprofile…nq40, @nprofile…dvuv, correct me if I'm wrong here. Also, do you have a better solution to this issue than "big shared broken relay list"?

2025-08-18 21:27:45 UTC

- reply

CC: @nprofile…6u4e, @nprofile…vhu6, @nprofile…xuyp

2025-08-18 21:26:57 UTC

Hey Nostr, I hear you like vlogs now… so I’m completely ignoring the trend and releasing **Haven v1.1.0** instead. 😂
https://github.com/bitvora/haven/releases/tag/v1.1.0
https://haven.accioly.social/a6b87db33cd059d61b2b95bd7e02de6ab725d29949cf245e21aaa7b653088a78.png
This release is packed with lots of bug fixes and plenty of background work paving the way for v2.0.0. It also has a few new features. Here are the main highlights:
1. Signed binaries: no need to install Go, build tools, or compile Haven yourself anymore (at least if you’re on Linux, macOS, or Windows with a x86-64 or arm64 CPU).
2. Improved import logic: Haven should crash less and no longer get stuck during imports. My own instance has been running crash free for 26 days. Logs also provide clearer info when Haven can’t connect to relays, so if you’ve unreachable relays in your `relays_import.json`, now’s a good time to fix this.
3. Blossom media: Support for deleting media, plus a bug fix that lets the Primal Android client mirror media to and from Haven on mobile.
4. Blossom Blob Descriptors: migrated to a separate database and no longer queryable from the Outbox relay.
⚠️ Breaking change: If you’re upgrading from v1.0.5 or earlier, you may hit a BadgerDB migration issue. Please back up your database before upgrading. If you see
```
error running migrations: failed to delete index key 07935dc9483d36f24464b36e6c000001b9: Txn is too big to fit into one request
```
Delete the `db` folder and reimport your notes.
Huge thanks to @nprofile…4t0k, @nprofile…k6x3, @nprofile…h5pr, and @nprofile…9vt4 for alpha testing, troubleshooting, and helping me improve the docs.
Enjoy!
#Haven #newRelease #Nostr #growNostr #relay #personalrelay

2025-08-16 23:49:15 UTC

Good night, Nostr. I hope you’re having a great weekend.
Here’s an interesting discussion about binding arbitration and the degree of freedom we enjoy here on Nostr compared to other social media networks:
@[email protected] 🔗 https://mastodon.online/users/mastodonmigration/statuses/115040517122921946
- Bluesky now has a binding arbitration clause as part of their ToS. Users hate it. There’s nothing they can do about it. There was massive user backlash. Complaining won’t solve the problem. Bluesky belongs to a company. Protecting investors comes first. Users are the product.
- Mastodon attempted to do something similar. Got a ton of backlash from users and paused the initiative. Unlike Bluesky, Fedi users are way more likely to move to alternative tech (be it based on ActivityPub or not)
- Nostr: I can’t think of anyone even trying to do this. I mean, who are you going to sue to begin with?
TL;DR: Use a social media network where you aren’t even sure who you could sue or persecute if you disagreed with something. Use a social media platform you can build yourself. Use social media where you own the content. Make sure you (not anybody else) decide what to boost and who to mute.
#SocialMedia #FreedomOfExpression #Nostr #GrowNostr #Bluesky #Mastodon #GoodLuckWithYourArbitrationClause

2025-08-11 00:39:11 UTC

- reply

I’m not British myself (well, technically I am; I do have a British passport - but I was not born in Britain, and can’t take credit for whatever they did for good or bad in the past). Also, Canadians came to the rescue of this island many times, so the appreciation for your people (even the unapologetic separatists) is mutual.
I hope you get what you are looking for, and that all your certainties and convictions about what the world is or isn’t, and what your friendly southern neighbours will or won’t do when world leaders go insane and war-time military conscriptions begin, turn out to be true. The world would certainly be a much better place if rich “farms of the world”, be it in Canada, Australia, Brazil, Europe, the Middle East, etc. couldn’t be bullied in times of need (or at any time, really).
I’m nowhere near as certain of my own convictions or my ability to predict how people will act when push comes to shove. As far as I’m concerned, if I could bring my family members out of Canada at the moment, I would (not that the UK is doing any better at the moment). Unfortunately, I can’t.

2025-08-10 18:52:05 UTC

- reply

Entendi. Ainda assim é estranho... Tipo, subir o kind 3 de outras pessoas para o seu relay não afetaria o conteúdo do seu próprio kinf 3. Para rolar essa salada você teria que assinar um novo evento que adiciona a pubkey desses estranhos ao seu kind 3. Enfim, eu já vi clientes fazendo coisas bizarras com essas listas (já vi até XML dentro dos eventos lol). Mas enfim, pelo menos a sua chave parece não ter sido comprometida. É isso que importa! E obrigado movemente pelos testes

2025-08-10 17:55:52 UTC

- reply

Faz sentido. Usando /haven --import o haven carrega o kind 3 mais recente dos relays que você configurar no arquivo de import. Mas se você assinou um kind 3 mais novo é isso que ele vai importar. De qualquer forma, se você achar um kind 3 antigo no Citrine ou qualquer outro relay pode assinar novamente e o Haven vai importar sem problemas. Se você não tiver o Citrine, adicione o relay do Semisol abaixo que ele prove a mesma funcionalidade. No futuro você pode restaurar sua lista de seguidirrs em um passo ao invés de sair caçando kind 3 antigos em um bando de relays e ter que assinar uma nova versão manualmente.
Realmente gente que você não conhece na follow list é estranho. Por outro lado acho que se sua chave privada tivesse sido comprometida mesmo eles fariam mais do que bagunçar os followers. Muito provavelmente algum se confundiu com a sua restauração de lista e fez alguma coisa doida.

2025-08-08 20:25:10 UTC

- reply

Got it. Sounds awesome! I sorta miss the days of mailing lists and patches. I’ll definitely play around with it. Since you won’t take zaps, here’s my kudos and appreciation from one “Indie Nostr dev” to another.

2025-08-08 17:05:16 UTC

- reply

Added to my "play with" list. Without trying to pitch x vs y (the more Git over Nostr options we have, the better), but how would you compare n34 to, say, ngit CLI? Any differences in approach, are both tools compatible? (I.e., can I init a repo with ngit-cli, push it somewhere and then use n34 to submit a patch?)

2025-08-08 14:10:33 UTC

- reply

I'm still stuck with NIP-46 when logging in to ChaChi. Any known workarounds other than installing a browser extension or pasting my nsec directly?

2025-08-04 01:56:48 UTC

- reply

Hey Girino, yes. Unfortunately, this is the same error that @nprofile…k6x3 reported here: https://github.com/bitvora/haven/issues/85 — which I also faced myself.
I don't have a better workaround than copying the current database somewhere, starting fresh, and running /haven --import again. (You can even start two instances of Haven and use the second one to import from the first if you don't have a copy of your notes on other relays.)
I looked into the possibility of breaking down Khatru migrations into smaller transactions, but it's not a trivial thing to do. After speaking with Fiatjaf, it seems there isn't much appetite to tackle this, as it would substantially complicate the migration logic.
In the future, I hope to provide more robust Haven backups along with proper "restore from backup" functionality as an alternative to the current import mechanism. I'm also planning additional support for relational databases such as SQLite and PostgreSQL, where migrations are a much easier problem to solve (and easier to work around on Haven's side if anything breaks in EventStore).
Apologies that I don't have a better "solution" for the issue, but hopefully the re-importing gymnastics won't be too bad.

2025-08-01 15:55:53 UTC

- reply

Yeah. That’s where I want to go with Haven as well. Having a nice, private UI where you can say:
* This person is trustworthy; let their followers write to my relay.
* This person? I don’t know. If three other folks follow them, then let them write to my relay (default).
* This other person? I love them, but they follow bots and weird people, so their follows don’t count when deciding if someone can write to my relay.
This is all "private" and local.
I’m currently running a private fork of Haven with a barebones implementation of this. No UI, just local files with npub / trust-level pairs, and so far it’s sorta working for me.
It’s still essentially a dumbed-down version of PGP ownertrust, but (hopefully) something that can be done in a way that’s more intuitive and attractive to normies than the whole PGP “signing key party” stuff.

2025-08-01 15:52:51 UTC

- reply

Yeah. That’s where I want to go with Haven as well. Having a nice, private UI where you can say:
* This person is trustworthy; let their followers write to my relay.
* This person? I don’t know. If three other folks follow them, then let them write to my relay (default).
* This other person? I love them, but they follow bots and weird people, so their follows don’t get write access to my relay.
This is all private and livsl.
I’m running a private fork of Haven with a barebones implementation of this. No UI, just local files with npub / trust-level pairs, and so far it’s sorta working for me.
It’s still essentially a dumbed-down version of PGP ownertrust, but (hopefully) something that can br done in a way that’s more intuitive and attractive to normies than the whole PGP “signing key party” stuff.

2025-08-01 15:15:52 UTC

- reply

I noticed. Endorsing content you like and people you trust is a powerful tool. Your followers doing this (interacting with your main profile) is the reason your content reaches me, and why we have all sorts of interactions despite not following each other directly.
They stop following you, your content stops reaching my feed (and vice-versa unless you are using one of the indirect ways to keep up with me mentioned above).
There has to be a way to solve Nostr’s cultural issue of “degrees of separation and endorsement from Jack / Odell / Derek / Lyn / Will, etc.” as a measure visibity without burning down the social graph and scorching the earth behind us.
My hypothesis is that we need a culture of more intentional following like what you're advocating for, but with proper ownertrust-style tools. I've asked around and a lot of people want to “follow that one person who produces crappy content we’re all embarrassed by, but we still love them regardless”. So kind 3 is better than nothing but not sufficient for WoT.
My main problem is how to bake in that intentionality in a way that’s accessible and fun instead of recreating a dumbed-down version of the PGP Web of Trust experience. It’s definitely an interesting research problem, one I don't have an immediate answer for, unfortunately.

2025-08-01 14:15:51 UTC

- reply

I'm really not interested in your data specifically as what happens if everyone follows you down this path (we’ve had this conversation before and agreed to disagree on this inr). But IMO, just to entertain this argument, we can take this line of thought to the extreme.
I have no illusion that AES-256 encryption is truly private against folks with the right resources. But going down that path, are you 100% sure that a relay running on your mobile is private?
Take this to the extreme and we’ll end up with non-social, 100% offline, share-nothing "social media".
At some point, you have to decide on features vs threat model. I’m happy to expose my social graph if it helps other folks discover good content on Nostr.

2025-08-01 13:44:13 UTC

- reply

Encrypted p tags on Kind 30000 are a good alternative, if more client devs decide to support it. IMO, coupled with the Outbox model, it’s even more powerful, as you can have Mastodon-style lists without having to cache stuff in a local rral.
https://github.com/nostr-protocol/nips/blob/master/51.md
The trade-off, of course, is that if everybody decides to do this, a whole aspect of social media (“I wonder who anon is following?”) goes away. It also kills one aspect of Nostr that I really enjoy " logging in" with someone’s npub.
Privacy++, Discoverability--.

2025-07-30 10:30:37 UTC

- reply

Thanks! Intention is what matters. Currently my lightining address is redirecting to Wallet Of Satoshi, it seems to be online and working well...
429 usually means you / your client is being rate limited (too many requests). If you're still having issues, could you let me know which Nostr client you're using?

2025-07-29 22:18:06 UTC

- reply

Yes, it can. The short-term "right" fix here is to start small and grow the MapSize on demand, which, again, may or may not preallocate space fot the DB file depending on the OS, filesystem, and architecture. For example, on Windows and macOS ARM, the initial MapSize has to be smaller than the available disk space, but once it has started (despite file managers displaying a 273GB file) it’s not actually using that much disk space.
I considered implementing dynamic growth in Khatru’s EventStore, but it’s a reasonably complex, non backwards-compatible change. And when I discussed it with Fiatjaf, it seemed there wasn’t much appetite for it.
There are other LMDB forks/wrappers that handle this more gracefully, so Fiatjaf is probably right. The best long-term approach is to migrate db libraries or db technology altogether. That would likely also help with some other Khatru issues around migration, transaction size, etc. But that’s a separate conversation.
The static MapSize parameter is good enough for Haven’s needs, and I assume for most other Khatru relays as well. Relays with the pretension (or real need) to scale on demand, potentially to petabytes in an enterprise DB cluster won’t be well served by EventStore, and to be honest, not even by Khatru in its current form.
Still, that would be a good problem to have, and right in my area of expertise (Distributed Data-Intensive Systems) when we get there. Semisol is already building something along those lines, so he’ll likely be the one pioneering this kind of thing on Nostr.

2025-07-29 22:16:31 UTC

- reply

Yes, it can. The short-term "right" fix here is to start small and grow the MapSize on demand, which, again, may or may not preallocate space got the DB file depending on the OS, filesystem, and architecture. For example, on Windows and macOS ARM, the initial MapSize has to be smaller than the available disk space, but once it has started (despite file managers displaying a 273GB file) it’s not actually using that much disk space.
I considered implementing dynamic growth in Khatru’s EventStore, but it’s a reasonably complex, non backwards-compatible change. And when I discussed it with Fiatjaf, it seemed there wasn’t much appetite for it.
There are other LMDB forks/wrappers that handle this more gracefully, so Fiatjaf is probably right. The best long-term approach is to migrate db libraries or db technology altogether. That would likely also help with some other Khatru issues around migration, transaction size, etc. But that’s a separate conversation.
The static MapSize parameter is good enough for Haven’s needs, and I assume for most other Khatru relays as well. Relays with the pretension (or real need) to scale on demand, potentially to petabytes in an enterprise DB cluster won’t be well served by EventStore, and to be honest, not even by Khatru in its current form.
Still, that would be a good problem to have, and right in my area of expertise (Distributed Data-Intensive Systems) when we get there. Semisol is already building something along those lines, so he’ll likely be the one pioneering this kind of thing on Nostr.

2025-07-29 12:20:53 UTC

- reply

Fingers crossed yes. But when politicians, lawyers, etc are involved… Who knows? I think that public relay operators like @nprofile…dx2g will test the waters first. But I really wonder what's everyone plab will be (if any).

2025-07-25 14:17:59 UTC

- reply

Not sure if you've already fixed Alby, but it looks like you've set your cash address as your NIP-05 (which is a separate thing meant to be a sort of user handle). Your Lightning address is still pointing to Alby.
If you don't mind telling. How many sats have you lost so far?

2025-07-25 11:44:57 UTC

- reply

VC/PE in a nutshell.
Full disclosure: I used to work for investment banks (as a techie; not front office, M&A, or anything fancy but close enough to interact with VC and PE folks). I still get a personal laugh out of all the finance bro memes here on nostr, especially when folks end up parodying the very finance culture they mock.
Add some laser eyes and a (a conveniently curated to allow him to profit from fiat) cypherpunk persona to the finance bro in a jumper meme, and you get the stereotypical Nostr VC-backed crypto bro.
The difference is mainly that some people in traditional finance are actually making money out of it.

2025-07-25 10:21:59 UTC

- reply

Agreed to an extent. Crucifying devs for something that’s mostly out of their control (it would be a nice world where devs and product folks could actually predict how people use their software) isn’t productive. So, to be clear, I don't think mobbing devs is the way to fix this! Especially since I’m not a believer in algorithmic solutions to complex social problems.
However, I disagree with the "bad vibes bring bad vibes" argument, especially since the toxic Nostr monoculture block isn’t exactly passive either. They don’t just exclude you for saying something that goes against the grain. They ruthlessly sabotage people with different ideas, all so that the "GM Pura vida 🤙🤙🤙" bubble can keep monopolising the Nostr trending lists and pushing the same rehashed marketing over and over and over again, scamming users out of their first 10k sats or so to buy services they don't need.
Refusing to talk about it or call it out just because "every dev already knows" reinforces the idea that everyone on Nostr is fine with it… or worse, that everyone’s part of it. New users don’t know what we know.
Voicing user concerns, or at least showing some agreement, helps avoid churn. It shows there are real humans who care behind the machine.
I could ask AI to create a more explicit illustration of what I think about the whole "let’s just not talk about it" situation, but hopefully this family-friendly version gets the idea across.
https://haven.accioly.social/a380aa7abd7bed9dec72dc1b80774311312609d489324050c19735a0d6920426.png

2025-07-25 09:46:24 UTC

- reply

There are dozens of us. Thank you for your service, @nprofile…56gx. I appreciate your efforts as well. There's very little appreciation for independent devs on Nostr, and we should be more vocal about it.
Agreed, the business model for BTC/Lightning solutions marketing to new users will eventually eat itself. And I don’t mean in Ouroboros style (at current churn rates, there’s very little money to be made from naive new users anyway).
The bad news is that with its downfall, some of the funding keeping Nostr software alive will likely disappear. The good news is that… well, there are dozens of us trying to keep Nostr alive, regardless of financial incentives.

2025-07-25 08:53:30 UTC

- reply

GM verbiricha. Agreed. I have an interesting relationship with Nostr myself. It's been a while since I’ve spent this much time on any social media. The protocol itself is genuinely interesting, but…
Honestly, in a year, I’ve made very few meaningful connections. There are maybe a dozen people at most who even try to interact with me about anything (aside from technical support for HAVEN, I mean).
The fact that I’m not a Bitcoin maxi (and am vocally allergic to the scammy behaviour of some of the Nostr celebrity accounts pedalling paid services and "Bitcoiner busy boxes") doesn’t help either.
And yes, the culture of shutting down any "unapproved" ideas, only to later adopt them once one of the Nostr celebs "endorses" the criticism, is really annoying. It’s funny, for a bunch of folks who see themselves as free-thinking, sovereign, libertarian techno-enthusiasts, bla bla bla, most of the few humans still on Nostr act like a bunch of 🐑🐑🐑🐑🐑.
I hope that one day, when someone mentions Nostr’s toxicity and its 20-year-old-crypto-bro monoculture, I can reply with something other than “Yes, but…”.
The best I can do right now is:
“Yes, but… there’s also half a dozen artsy and techy folks mostly talking to themselves with hashtags that I really like. Plus Japanese Nostr.”

2025-07-21 12:04:33 UTC

GM folks. Just sharing this again. Now inviting macOS, Linux, and Windows users on Intel and AMD hardware as well:
https://github.com/aaccioly-open-source/haven-gha/releases/tag/v1.1.2
Feedback is very welcome! I'm especially looking forward to hearing from users who prefer binaries over compiling from source. Are the binaries working well? Is this helpful at all? Would installers and packages (.deb, .rpm, .msi, etc) be useful?
#haven #binaryPackages #devstr #asknostr
#nevent1q…hy9j

2025-07-19 17:05:29 UTC

- reply

Yes, unfortunately, there are both known and unknown issues with the import process (both the historical and "on-demand" imports), and it does get stuck. This is due to a combination of factors, ranging from misbehaving relays (e.g. relays purposefuly blocking Haven - nothing that I can do here -, dropping long subscriptions, getting stuck themselves, etc) to the fact that we're not leveraging the underlying libraries' mechanisms for retrying subscriptions, "grounding" flaky relays, and re-adding them using exponential backoff best practices to the best of its potential. On top of that, there are some hard to troubleshoot concurrency issues in Haven, Khatru, and the Event Stores that can lead to missing writes.
There’s a lot to fix here, as well as plenty of unknown unknowns. A proper fix for the entire import saga will likely only come after the Haven v2 release, as we’ll need to make several backwards-incompatible changes to address this properly. That said, if you collect any evidence that might help us troubleshoot specific cases where the import is getting stuck, please feel free to update the relevant GitHub issue.
For now, while not ideal, what I’ve been doing is:
- Keeping the import relay list to a minimum (just 3 trustworthy relays)
- Restarting Haven daily with a cron job, just in case the on-demand import gets stuck
- Running a full historical import once a week
To be fair, there have been weeks where the full import didn’t find any new notes, which is a good sign. It likely means that external relays are becoming more stable and Haven's on-demand import process is getting stuck less frequently. I may even consider disabling the daily cron job at some point. But overall, I think the weekly historical import is still a necessary fallback, given how Nostr works with relays going offline and coming back all the time.
I'm considering adding a feature to Haven so that it can run historical imports in the background, removing the need for users to do it manually (cc: @nprofile…6u4e). But to be honest, this is likely more than six months down the road as both Utxo and myself are a bit busy with other stuff, and I already have a giant backlog for Haven and Khatru. Unless, of course, some other contributor steps in, implements the feature and fixes the mentioned issues before I do 😅.