melvincarvalho
Melvin Carvalho

In Bitcoin and Nostr, a false dilemma (ake either/or fallacy) often divides people: either adoption must reach millions, or it’s considered a failure. This mindset is like saying a small plant must become a massive tree overnight, or it’s not worth growing at all.
Cypherpunks like Satoshi ignored these extremes. They focused on writing code and solving problems, knowing growth happens step by step. Polarization halts progress—pragmatic solutions nurture it.
https://en.wikipedia.org/wiki/False_dilemma

2024-09-15 23:05:43

Thinking about a game on testnet. Every nostr account can roll one character, and it gets generated in the next block. The character has to die, before you can reroll a new one. Characters can fight each other for XP.
https://m.primal.net/KrKJ.png

2024-09-12 10:57:18

2024-09-12 06:26:11

#mindstr explain Bitcoin Economics
https://m.primal.net/KnZW.png

2024-09-04 16:46:11

Talking about NIP-98 with the inventor of the Web :)
https://image.nostr.build/d4dc887aa49f438774df24092a29d768330b60f8fd7dcdcbe8a3ad9eb51c6677.jpg

2024-09-04 08:42:58

I have liked this interview, for a long time, and often think back to it.
"I usually don’t speak on behalf of myself. I don’t fight for myself. That was the one time that I did. I don’t know if it worked. No one really cared." -- Jack
I think some people care. I think some people will fight for Jack, or with Jack.
https://www.rollingstone.com/culture/culture-features/twitter-ceo-jack-dorsey-rolling-stone-interview-782298/

2024-09-04 08:37:50

100M Token Context Windows
https://magic.dev/blog/100m-token-context-windows

2024-09-03 13:11:17

- reply

Thanks for the clarification. Its a worry that the privacy oriented OSS project says nothing about the privacy invasive company and closed source project. I've seen this movie too many times.
I remember now I went to a fedimint workshop. The speaker asked how to send keys to one another, and I said, from the audience "Nostr".
Fedimint has too many flags for me to blindly trust it.
I hope it works though, and the docs are interesting, and could easily be done with the *existing* relay infrastructure with a 2/3rds honest majority. Rather than having to run new software, learn it, audit it etc.
"One such protocol is AlephBFT. We will mainly use it as a reference for BFT consensus properties but note that similar but more efficient ones exist (most notably Dumbo and hybrids built on top of it)."
https://fedimint.org/docs/MiniMintDetails/Consensus

2024-09-03 11:32:23

- reply

Before nostr, I founded and was first chair of the W3C Solid Community Group. I always wanted to build Solid and personal storage to work together with nostr. In we made pub/sub websockets in solid before Nostr did it. I spoke to Alex and I am hopeful we can get some Solid ideas to work with ditto. Then we have Nostr + Ditto + Solid to be better than ActivityPods, and a whole scalable financial layer can be built on top of that.

2024-09-03 10:21:11

If you think nostr is good, #ditto is going to blow your mind.
What has been achieved, with simple text strings, and small relays, is incredible.
When devs get all of that, PLUS the full power of structured JSON, a massively scalable architecture, and permissionles dev, literally anything will become possible!

2024-09-03 09:50:43

- reply

Good catch! Fixed!

2024-09-03 09:45:29

- reply

Great!

2024-09-03 09:41:37

- reply

https://play-grounds.github.io/ditto/?user=stu@poster.place

2024-09-03 09:36:57

- reply

I have tried adding:
object-fit: cover;
"CSS property ensures that an image fills its container while maintaining its aspect ratio, cropping any overflow to prevent distortion."
Maybe it helps a tiny bit? I cant really see much difference.

2024-09-03 09:15:14

- reply

Ditto is a new scalable architecture built on nostr. It has both relay string API and an interoperable JSON API. In this case, I am building new small (micro) apps, which are self contained, on the ditto API. I love it!

2024-09-03 09:06:45

#ditto micro app can now pull in multiple profiles
https://play-grounds.github.io/ditto/?user=alex@gleasonator.dev
https://image.nostr.build/92aab4a8ea68a4e65e813385c2c26b630c1e430967a8a2a6c5fe49c262d4a70d.jpg

2024-09-03 08:37:41

- reply

Not sure I trust fedimint to be honest. They claim to be a privacy solution yet have spyware on their high profile new launched site. Nevertheless, I hope they are successful. I may use it one day. The principles behind can be replicated in any number of ways. Relays already exist, so they can do something useful, without needing a 3rd party spec or system, that I would have to learn and audit. End result is more choice for the end user.

2024-09-03 08:32:01

- reply

Keep going. And JavaScript is your friend ...

2024-09-03 08:25:16

My first #ditto micro app. Not much, but it's a start!
https://play-grounds.github.io/ditto/
https://image.nostr.build/31648dbeb32c81cae3505612b51edb1b6e42edee8f20d0ea97da56c54370d9a7.jpg

2024-09-03 07:55:28

- reply

Testnet coins have value. The mythology of it is to say that they dont. You cant actually wipe a testnet, any reorg will make the chain of accumulated work longer. All you can do is stop linking to it.

2024-09-03 07:51:16

- reply

Seems very similar to fedimint. But using relays, which already exist.

2024-09-03 07:46:56

- reply

Have you tried ditto yet?
https://docs.soapbox.pub/ditto/install

2024-09-03 07:43:23

- reply

🍿

2024-09-03 07:42:34

- reply

ditto
lightning
qtorrent (search)

2024-09-03 07:21:04

- reply

Running it now at https://nostr.me ... loving it so far!

2024-09-03 06:52:46

If you could create a dynamic federation of relays, with 2/3rds honest majority, it would be superior to cusodial services or mints. This should be tried on bitcoin testnet, to proof it works. And then offered as a service on nostr.

2024-09-02 19:43:22

Finally got new pc with ubuntu 24.04 working. Time for a new subkey... and to write a subkey linked data spec...

2024-09-02 18:52:12

- reply

Good talk—it really opens people’s eyes to possibilities they might not have considered. I’ve been pushing this message for a decade, about what apps, socials, and discovery can bring to Bitcoin. But it’s only now, with Jack on board, that more folks are listening.
That said, Nostr isn’t the only way to tie metadata to keypairs. It’s one of the clunkier, less extensible options, and changing anything in the NIPs takes forever. You mentioned the existing network has utility, but I doubt there are thousands of relays—it’s probably smaller, more centralized, and less reliable.
It’s a solid message, but I don’t think Nostr is the only path forward. The censorship resistance might be overhyped too—it could just be that Nostr’s smaller than Twitter, which might also be why people still use Twitter.
The micropayment angle is strong, maybe even a holy grail use case. But Bitcoin itself isn’t used much, which is a shame since npubs are taproot addresses. We could’ve built a whole free market on a non-rigged base layer by now, but there’s nothing happening at that intersection—a missed multi-billion dollar opportunity. And honestly, there’s no real appetite for Bitcoin base layer integration with Nostr. Hope you don’t mind me pushing back a bit here—I figure you might not get much of that in the comments

2024-09-02 16:21:13

- reply

Good point. We’re still aiming at a moving target. In the next four years (the next epoch), we could see AGI emerge, which could change everything—how chains are used, how development happens, how we discuss things, and the path to ossification. Once AGI arrives, we’ll likely get used to virtual assistants that communicate on new layers, invent tech we haven’t seen yet, and do all this without needing forks.
We’ll be able to evaluate things like the mining fee auction more efficiently, figure out things like the nash equilibrium and iterative prisoners dilemma e.g. how much will a high value tx pay to get priority. All our research and explanations will get a big helping hand.
The app landscape for consensus and polling might shift too. We’ll be able to replicate the best parts of DeFi on Bitcoin, Nostr, Ditto (maybe Pear too), and we’ll be better at identifying and countering social attacks. The future could be pretty interesting, and we're not in a rush for a killer feature or osification I dont think immediately. We do want to avoid catastrophic failure vectors, and guide the project carefully through the next phase.

2024-09-02 15:31:15

- reply

Thanks, that’s helpful. My take is that we should really push single-use seals to their limits first. It’s a Layer 2 solution based on (u)txos, and like with Nostr, once people see what’s possible, they’ll get it. Single-use seals might not be perfect, but they’re likely better than Tether, and people use Tether. It’d be great if folks used and tested the tech we have before asking for more features or forks.
We’re also assuming sovereign identity is just for humans, but maybe machines will use Bitcoin more than humans. We need to see how things play out to make the right trade-offs and keep in mind the risk of catastrophic failure—like a chain split. Remember, there are vested interests that might actually want a split. A fork needs both social and technical consensus and has to be something the user base, many of whom have entrusted their life savings to this, actually wants.

2024-09-02 14:12:05

- reply

Glad you brought this up. We should definitely go through these points and possible solutions, though it will take some time—probably weeks or months—to cover everything in detail.
On risk, I think the biggest threat to Bitcoin is a fork and chain split, even more than a 51% attack. We've never had a 51% attack, but we've seen multiple forks and splits. Any new feature has to be worth that risk, and it should be explained clearly enough that at least 10% of users can understand it from first principles. Let's keep in mind the real dangers of a chain split.
Right now, I don’t think we’re fully using the tools we already have. I monitor testnet4 closely, and it’s clear that Bitcoin + Taproot (which ties into Nostr) isn’t being used much. Sometimes my transactions are the majority in a block.
What do you mean by critical primitives for scaling sovereignty? One trade-off could be that Nostr npubs and Taproot addresses offer sovereign ownership and property rights, whether on the base layer or something like Liquid. We’ve barely scratched the surface of using the (u)txo model for a chain of custody, like with single-use seals. A decentralized free market could be built on Bitcoin with this, but I’m not sure there’s much appetite for it right now—especially for niche use cases like vaults that no one’s really asking for.

2024-09-02 13:21:32

- reply

https://image.nostr.build/5929730a702f5bcd594ad73800aad5faca817faacf161ab80ad6551afeaf0b20.jpg

2024-09-02 12:06:56

- reply

We dont need a soft fork. The features we have have barely been used. If they want OP_CTV try it out on elements and or liquid, and show there is some demand. There is of course no demand for it. The upside of a soft fork is very little for the bitcoin community. The downside is a chain split, which of course, all the enemies of bitcoin want, and will fund.

2024-09-02 05:27:32

The initial version of caniuse for nostr from Marco.
https://marcodpt.github.io/nostracker/
https://image.nostr.build/42c2e3ff354fd4518173895e5140a26874ad8a75de7097b18dec7dab261b57ff.jpg

2024-09-01 23:01:40

I've been thinking a lot about what Nostr 2.0 should look like:
1. Bottom-up, permissionless dev instead of top-down control through NIPs.
2. A scalable architecture that’s interoperable, extensible, and standardized, including relays.
3. Cutting-edge features like smart widgets to compete with the latest tech trends.
4. Full npub-taproot infrastructure with financial interop, smart contracts, DAOs, markets, and naming.
5. AI integration where AI agents are first-class citizens on the network, leveraging all of the above.
#Ditto seems like the first step toward Nostr 2.0 (with shout out to nostrocket). Nostr 1.0 apps and the relay network are still going to be key, but I see the next generation starting to take shape.

2024-09-01 21:13:26

- reply

Which clients?

2024-09-01 21:11:04

- reply

The neglect of the nostr relay network, has been quite remarkable. Everything else was supported, but the relay network was thrown under the bus often is a quite unkind way. It is a shame because without the relay network, there would be no nostr.
Ultimately I think it's a case of "tragedy of the commons". It's never too late to help the relay network. But 80% have left and most wont come back. Let's try and do better with #ditto !

2024-09-01 20:57:57

- reply

Sad :(
Relayable.org has been downsized from its network of worldwide relays to just one in Los Angeles. This is due to a lack of financial support and large monthly hosting bills. We tried numerous times to get funding from the usual suspects but were not even sent back a denial email. So due to the ongoing costs and slowing of direct support with the price of BTC going up, the hard choice was made to downsize. Thanks to everyone who has supported and helped over the last 2 years!
- Jascha
https://relayable.org/

2024-09-01 19:53:53

#ditto allowed me to create a new nostr client, relay and website, in a weekend. Still some work to be done, but happy with the results so far.
https://nostr.me/

2024-09-01 11:52:33

- reply

It is a relay server. But it is also an extensible scaling architecture with mobile and web front ends.

2024-09-01 11:22:58

Still a long way to go, but in half a day, I have almost got my first #ditto instance running. Still in experimentation mode for now...
https://nostr.me/timeline/global

2024-09-01 10:02:14

- reply

Yes, indeed! Had a good chat with Alex about it, this week. A couple of other people are working on this too. #ditto has already done some work in this area, but we can extend it further.

2024-08-31 17:58:19

- reply

Not 100% sure you need a NIP-05, because I saw it does @npub... otherwise, but I need to check that more. I dont have any issue with NIP-05 myself, it's historically been quite good and useful. However it did get changed recently.
What I like about ditto is that Alex has ideas on how to scale it, and bring things to nostr not seen before. He has also scaled other projects (to millions), so we know it's possible.
What I like about ditto is that it has caching strategies, strategies for fast content delivery, a highly polished UX. It has high interop and an architecture that I love. It is extensible and modular. With ambitious plans. A slight personal bias that I have, is that MK and Alex are both very likable, and bring alot of energy to the project too. I have trust in the team, so I am willing to throw some time at it, and have confidence, it wont be wasted effort.
Ditto also has great professional, documentation. It ticks all the boxes for me, that I've wanted for a long time.

2024-08-31 17:32:59

- reply

Similarly concept, but not quite the same. I went through a lot of details with Alex and MK, when I met them. I'm convinced that #ditto is next nevel. I do hope relay.tools will work well too. Ultimately (friendly) competition is a win for the end user.

2024-08-31 14:54:21

#ditto is the best kept secret on nostr. If you want to try an amazing client, go to:
https://ditto.pub/
Or install your own version using:
https://docs.soapbox.pub/ditto/install
With an elegant, scalable architecture which can handle millions of users. This was the missing piece of the puzzle that nostr has needed.
https://m.primal.net/KXNT.png

2024-08-30 13:44:35

Few people know that the original web was designed a decentralized space where everyone could browse and edit thei own pages. The original browser, called worldwideweb, was a browser/editor, where anyone could make edits to the web as well as browse the web. What happened next was that VCs removed the editing functionality and created web based silos which locked in the users. We now have a chance to add back the decentralized functionality to the web, returning it to its decentralized roots.

2024-08-30 13:14:26

Ditto scales, so nostr doesnt need to. And through federation ditto has already scaled to millions of users and growing. You dont have to use ditto, just take the design patterns. Not only that, but we can add almost unlimited features to it.

2024-08-29 15:07:37

Great to mean Alex Gleason and MK! Bullish on ditto.

2024-08-29 10:39:42

- reply

It feels like it, but it isnt. Nostr is controlled by one entity. If it transitioned to community based, like bitcoin, it would be good. Might be a bit late now.

2024-08-29 10:34:28

- reply

yes

2024-08-29 08:37:42

- reply

It's the future

2024-08-29 08:35:38

- reply

Yes

2024-08-29 07:20:08

@npub16c0…6nvr could we fix this bug, seen at least 5 times. People coming to primal cant follow someone if they are on 0 followers. That's a v bad UX.
https://www.reddit.com/r/nostr/comments/1f3t262/cant_follow_anyone/

2024-08-29 06:58:31

Meanwhile, in nostr dev chat, censorship resistance is going swimmingly ...
https://m.primal.net/KTEl.png

2024-08-29 01:26:23

- reply

Nostr's really just a subset of HTTP—how you use it matters. The freedom angle feels a bit overhyped. People might think it protects them, but it could end up storing signed evidence forever, potentially more risky than regular socials. And let's not forget, some devs slip spyware into big sites. Personally, I see it more as a tool for social app portability, letting you move between sites easily. That's the real value.

2024-08-28 21:23:38

- reply

Why base32 encode the pubkey? Why not use the hex which is the canonical form?

2024-08-28 20:28:07

- reply

cool! yes im having the same trouble with dvms. this pattern with a node server sounds very good, exactly what i mean by using nostr as an API

2024-08-28 20:11:55

- reply

You mean existing end product using relays? Or something to connect a relay to an existing product, like a bridge? That could be huge yes.

2024-08-28 19:39:10

- reply

I didnt strictly define it. But something like a tool in your toolbox, rather than the toolbox itself.

2024-08-28 19:37:59

- reply

The analogy is that you have a working system and add an API, ie add nostr. Rather than using nostr as a starting point. So, say, activitypub could use nostr for realtime updates. Git could use it for single sign on. Twitter could use it for micro payments. And so on.

2024-08-28 19:36:24

- reply

You are correct, I defined it loosely, or didnt define it. But it's more of an analogy. API being a tool that you use, one of many. Rather than the main system on which you build. If that makes any sense!

2024-08-28 19:33:16

Nostr is better as an API than as a protocol. It might end up as either. If it's an API it doesnt need to scale, because it inherits scale from http. So you use the API and are happy. I think that's where it will end up, but it could go either way. Both are good outcomes.

2024-08-28 18:22:04

rugged
#note1pks…8xhq

2024-08-28 16:29:38

- reply

The best website that nobody has heard of:
https://learnmeabitcoin.com/technical/keys/hd-wallets/extended-keys/

2024-08-28 14:04:06

- reply

Nostr isnt designed to scale at all. Connection oriented networks dont scale very well. But they can do a decent enough job.
Sufficiently decentralized is orthogonal to scaling and imho a red herring. We know what scales through observation and we know what doesnt scale through observation.
Nostr was under 100 users 2 years ago. It's done very well to get where it has, but it wont overtake the internet, and isnt designed to. It's a very useful tool, and unlocks a lot of developer innovation and creativity. Working as expected.

2024-08-28 13:58:46

- reply

Good. Now imagine if nostr was anchored in testnet4.

2024-08-28 13:52:44

- reply

You're responsible for your own actions. There can be any number of reactions. Where I see nostr's value is that people are able to speak the truth, and it's harder to silence that truth, at least, in theory. If the big relays ban you, you're going to have a bad time.

2024-08-28 12:12:13

Nostr - bring your own algo
https://uncensoredtech.substack.com/p/bring-your-own-algo

2024-08-28 08:43:39

- reply

I think it's different. Most people dont know this, but in the first few years of the web, there was actually a list of ALL the servers. They maintained it for a while. But it kept doubling every 3 months or so. In the end they could not keep track. Nostr is different. It was on a growth spike But then that growth stoped and reveresd. Now the relay is not growing exponentially, or even growing at all. It may be shrinking The two systems are quite different.

2024-08-28 08:08:37

Decent idea. I was looking at this yesterday. It is possible to run an idle relay in 20MB of memory. Hope it happens, although the relay network is currently shrinking and centralizing around a few reliable ones
#note1qat…we2z

2024-08-28 06:15:02

- reply

no no no ... nip 98 is hugely under rated

2024-08-27 20:28:19

Bitcoin 28.0 Release Notes Draft -- one of the best releases in recent years, finally w/ -testnet4
https://github.com/bitcoin-core/bitcoin-devwiki/wiki/28.0-Release-Notes-Draft

2024-08-27 20:14:30

- reply

Something more like NIP-98, but I would have to spend some time drilling into the details

2024-08-27 16:15:30

- reply

sorry new users on the right

2024-08-27 15:53:14

- reply

reddit spike too ... its_happening . gif
https://m.primal.net/KQAG.png

2024-08-27 14:16:56

- reply

I've been using about 6 clients, just to get the basics such as follow, reply, post etc. working. In the end they all have bugs, so I have decided to write my own client.

2024-08-27 13:43:00

- reply

Best hair in bitcoin

2024-08-27 13:41:18

- reply

worked

2024-08-27 13:19:08

- reply

Good, was worried! I do agree with snowden about privacy, but that is easier said than done.

2024-08-27 13:10:35

- reply

Not really true. fiajaf owns the protocol. He can change what he wants with or without community approval, as he did last week, when he ulilatearlly nixed parametrized replaceable events. Nostr may be open, but it certainly is proprietary.

2024-08-27 11:36:13

- reply

I like the direction in which you are thinking. Again, I think it's possible to be slightly dismissive of attack vectors we may not initially have thought of.
Given that the blob is signed. Could it be a problem that it could then be used as a ddos of all blossom servers. And the user gets the blame. What then? Block that user for uploading a long video.
I'm curious to see how this proceeds. Just playing a bit of devils advocate here, if it pleases the court.

2024-08-27 11:19:54

- reply

sorry I should have written "replay" attacks
this is a security consideration, because you can impersonate another users or drain their credits, without their consent
actually this also is a security consideration for nostr, but folks havent realised it yet, but they will
so im glad blossom and nip-98 are different, and im very glad you are trying stuff, and have made a system that works, kudos
but careful not to be too dissmissive of security issues, as most in nostr are today
lets say i have a balance nostr.build and and on nostr.cat and i upload a large video to one for 100sats, that request could be replayed on my other account and cost me 100sats without my wishes

2024-08-27 10:05:37

- reply

Freedom is in decline Machines are on the rise.

2024-08-27 10:02:42

If npub is for display purposes only, then the checksum should not matter, only the prefix, and the key. A good way to test implementations.

2024-08-27 09:50:51

- reply

nostr. Though bitcoin needs always to be alert. There were a couple of pseudo attacks fended off in bitcoin in the last 1-2 years. Namely the incription trolls, and the soft for drivechain deragement syndrome. Before that there was the "block size" war, which really was a governance war. And before that the foundation and Gavin, who were slowed down by the BIPs and other things. I think bitcoin is in quite good shape, but complacency should not be a word in our lexicon.

2024-08-27 09:48:25

- reply

BTW why "bottom" quintile, rather than "top" quintile. I do like how you phrased it. It reminds me of Terence McKenna's "we are lead by the least among us".

2024-08-27 09:46:51

- reply

I wouldnt be so sure about that. It's not the same project it was 1-2 years ago. Lots of dogma now.

2024-08-27 09:40:34

- reply

Didnt fully realize this. Thanks for the info.
Things are likely gonna get worse.

2024-08-27 09:22:27

- reply

I get that. What im askign is the "checksum" itself actually ever used. If not, then it might not matter.

2024-08-27 09:15:24

- reply

shoulda been bech32m like taproot ... maybe we should just ignore the suffix ... is it used anywhere for anything ever? low-res qr codes?

2024-08-26 18:44:23

- reply

Agree with most of this. However, torrents and social posts are not quite the same. Torrents are files that are designed to be shared.
Posts could have some personal attachment. You could have a business nostr server. You could have a friends group. You could have special articles you want to share only in cretain places. You might share a party photo with some friends, that you regret.
There is absolutely no default social contract in nostr other than you are using a websocket API over HTTP.
If the system you use gets your consent, that's a different story. But if not, you can have the right to feel violated. Or not recommend nostr.
The social contract is largely inherited from the status quo of users owning their own data, imho. Unless you agree to infinite sharing, which is also potenitally forever and undeletable.

2024-08-26 16:47:14

- reply

Too big a topic, it was largely done in the 1990s through lots of debates. Here is one output document, which has largely stood the test of time (except for maybe the content negotiation part). An important aspect is that connection oriented networks (such as websockets) carry a few overheads wrt scalability, but there's much more to it.
https://www.w3.org/TR/webarch/

2024-08-26 15:47:22

- reply

That is very cool, thanks. It might be helpful to have some text to say what kind 5050 is. But I can look it up.

2024-08-26 12:47:55

relays are trusted third-parites

2024-08-26 12:03:03

- reply

Not on its own. On its own it is just noise. The internet are just 1's and 0's. Keypairs are just 1's and 0s'. They are neither free, nor unfree. We as humans, assign meaning to them over time. A new keypair, on it's own, always has to strart from zero.

2024-08-26 11:34:54

- reply

Thanks Marty. You opened my eyes when you said that more standardization might encourage companies to come in. That is something I am able to work on.

2024-08-26 10:57:40

- reply

Agents on nostr. I gave up on DVMs after about 10 attempts. They are too complex for me to understand. I'm going for something simpler.

2024-08-26 10:18:17

agentic nostr
https://m.primal.net/KNjG.png

2024-08-26 09:59:49

- reply

Essential might be over stating it. It's a nice to have. Facebook got to 100s of millions of users before addign websockets. There's ways round it, but websockets finally standarized it. It's the icing on the cake, rather than the cake, itself.