<unnamed>

Public Key

npub176p7sup477k5738qhxx0hk2n0cty2k5je5uvalzvkvwmw4tltmeqw7vgup

Profile Code

nprofile1qqs0dqlgwq6l0t20gnstnr8mm9fhu9j9t2fv6wxwl3xtx8dh24l4auspz3mhxue69uhhyetvv9ujuerpd46hxtnfduqsqqgkwaehxw309aex2mrp0yh8qunfd4skctnwv46qzrthwden5te0dehhxtnvdakqtlfz5u

Publishing to

relay.damus.io

relay.primal.net

nos.lol

Last Notes

2026-03-13 18:16:22 UTC

- reply

they can check it already on npub.world

2026-03-13 18:11:26 UTC

- reply

benvenuti!

2026-03-12 22:33:04 UTC

- reply

Imma gonna zap myself to be top of the feed.
jokes aside, dealing with impostors, fake zaps ect. is why WoT matters

2026-03-12 18:29:21 UTC

- reply

In go I would definitely use a constant. With JS I am much less proficient

2026-03-12 16:39:31 UTC

- reply

so you want const to be used only for global constant, for example const SecondsInDay. Correct?

2026-03-12 16:38:07 UTC

- reply

you mean like private follow-lists? Yeah I think they exist.

2026-03-12 09:47:53 UTC

- reply

why?

2026-03-11 21:01:18 UTC

- reply

check out blossy, a framework for building custom blossom servers that I think will suit you well.
https://github.com/pippellia-btc/blossy

2026-03-11 12:01:26 UTC

- reply

So many things are so broken... Yet it's amazing!

2026-03-11 11:11:48 UTC

- reply

If you want I can give the REQ to reproduce it

2026-03-11 10:55:33 UTC

- reply

Same bug in wss://satsage.xyz and wss://nostrelites.org (which also is not responding to new REQs at all).
Is nostrelites yours @nprofile…0hy6 ?
#nevent1q…p0ql

2026-03-11 10:44:15 UTC

hey @nprofile…j5ga, I found a bug in your relay.
When sending REQs where there are many authors (like 100), it returns events that don't match the filters.
https://blossom.primal.net/8917f446835d2eba7dfcc98ede58a78444782e00a8552ee2713b49274f202b7e.png
For some reasons it seems to return jack dorsey's follow list even though one doesn't ask for it.

2026-03-10 10:06:29 UTC

- reply

Thanks for sharing, this is really valuable

2026-03-07 23:05:57 UTC

- reply

Thanks you mate🙏

2026-03-07 13:39:59 UTC

- reply

This is cool! GG

2026-03-06 22:41:50 UTC

- reply

man this is really good! I would have multi input forms for multi codes, but other than that it's 🧑‍🍳

2026-03-05 14:42:02 UTC

- reply

Since social graph computation is off by default, you could give the option to use a WoT provider like @nprofile…y38p 👀

2026-03-05 14:39:05 UTC

- reply

I've just downloaded it, and MAAAAN the UX is impeccable.
Big kudos.
One small note about mints management. I think there should be a ⭐ rating next to each mint, to help users search and choose their mints.
These ratings should use WoT (default is global, but can configured with a target npub) to filter out scams, impersonators and so on.
I've been thinking about ecash mint reputation in @nprofile…y38p for a while, maybe I should give it a go what do you think?

2026-03-04 14:30:35 UTC

- reply

How true!
Websockets are great at performance if usage is high. However they add state and thus are more complex than HTTP.
#nevent1q…yr3g

2026-03-03 10:39:38 UTC

- reply

just wow

2026-02-27 15:48:33 UTC

- reply

cc @nprofile…3xzs

2026-02-23 08:45:20 UTC

- reply

It's an interesting concept, but imo there are several issues:
- First of all, I don't understand why calling PUT /mirror client side "doesn't scale". It actually scale very well because there is no body to stream, except for a tiny URL the server will use to download the blob.
- the /list/<pubkey> endpoint is unrecommended, and this spec uses it
- server talking through nostr relays makes blossom server require websocket support, which makes compliant implementation heavier
The normal case for having duplication of your blob is:
- calling PUT /media (or PUT /upload) for the blob you want to upload
- you receive back the blob descriptor. From there you extract the URL
- use the URL is all PUT /mirror requests across o all servers you want to duplicate
Or, there are distributed servers implementation, where you upload on one and all receive a copy. It can be used as a single server, but internally it's multiple servers.

2026-02-22 16:59:59 UTC

- reply

What do you mean? Let's brainstorm together. I think we can make something awesome out of it

2026-02-22 15:42:18 UTC

- reply

@nprofile…auzy bro, why haven't you asked me? ;)
No seriously I built a very similar producer-consumer architecture for Vertex. Also, Vertex search ranks search similarity results using pagerank or personalized Pagerank.
@nprofile…tckv @nprofile…u8g0 how can I make Vertex so private that it can be considered for integration in WN?

2026-02-20 17:56:50 UTC

This new wave of spam is inflating all @nprofile…kclk followers count.
Mine increased by 500 in the last week, while the real followers reported by @nprofile…jku2 remained almost flat.
https://blossom.primal.net/6c00fb4e6c865040dae13b3ecc4a9d851f2e19e60ef73e809174be95651b1023.jpg
https://blossom.primal.net/5346c11679205561b221410011707c5e948bd1a763980d7493293df03886a52d.jpg

2026-02-18 15:03:05 UTC

- reply

IMO we'll reach $100T in 2037. Total gut feeling, no math done, just pure vibes of the collapse

2026-02-14 20:27:55 UTC

- reply

holy shit bro your memes are 🔥🤣

2026-02-13 16:04:00 UTC

@nprofile…wszm yesterday you asked me if I saw much spam in @nprofile…kclk.
Notifications are a bit spammy.
https://blossom.primal.net/96b242e248ad5288a96875b0181fe6b8037f354ee3265ba1de70a7c7c0f75c8e.jpg

2026-02-12 16:19:12 UTC

- reply

Thanks! I really appreciate

2026-02-11 00:21:22 UTC

- reply

O don't think it will be that hard to write a bridge

2026-02-10 17:25:21 UTC

- reply

don't know

2026-02-06 16:42:07 UTC

- reply

It's actually very hard for a bad bot to get more than 10 idiots to follow it. So for most practical scenarios, it does and will work.
Graperank, in the CURRENT form, is not contextual and so it would not be a meaningful improvement (if an improvement it is).
If Graperank has evolved from the last time I checked, send me a link to the code.

2026-02-06 14:28:29 UTC

- reply

They are below the real one, and even below other (relevant) profiles.

2026-02-06 13:31:13 UTC

- reply

You keep forgetting that Vertex deals with link farms.
1. Vertex does not dumbly crawl all events, builds a graph and computes Pagerank. It crawls based on hard to fake internal metrics. A person can include only a few bots, and those bots can't include others. It takes a week to even being able to add a few bots from the moment one got included.
2. Vertex computed acyclic Pagerank, so the classical "I follow you, you follow me" strategy is nullified. Walks can't be acyclic.
3. Vertex supports personalized algorithms as well.
I think this is probably the third time I tell you this. If you want more details about how it works you can check the code.

2026-02-05 13:08:33 UTC

- reply

There is a list for that.
It's not a mega vulnerability, but it's a vulnerability caused by the fact that blossom auth doesn't allow to scope the token to a particular audience.
Nostr Web Tokens allow that, and are indeed very simple, as they are inspired by JSON Web Token which are commonly used for these type of things.
The new blossom Auth spec (BUD-11) will just add the server tag, which is fine and fixes it. In my own framework I'll just support both.

2026-02-05 13:01:07 UTC

- reply

not at all, it's an attack a malicious blossom server (not relay) can perform against users

2026-02-05 10:29:41 UTC

- reply

There is a potential of a replay attack with DELETEs, because tokens are not scoped to any particular server.
Check blossom PRs

2026-02-05 08:58:19 UTC

- reply

You would be able to use regular Auth events or NWTs for every request. Different validations based on the event kind.

2026-02-04 22:10:07 UTC

- reply

My new framework called blossy is perfect for building that.
Auth is not well implemented because blossom Auth is insecure and currently being rebuilt.
I'll add support for Nostr Web Tokens tomorrow.
https://github.com/pippellia-btc/blossy

2026-01-30 22:01:34 UTC

- reply

Ferrari always wins tests. Mercedes is totally the favourite this year.

2026-01-26 16:47:31 UTC

✨ Meet blossy
The easiest way to make custom blossom servers come to life.
Think blossom server that support ecash, WoT gating and more.
All so easy to use that your LLM is going to one-shot it (probably 😝).
https://blossom.primal.net/a0ac480a1ab2987ac7ce37c9a09720a013ef9c27c6328e3b4a9ba6ecbdf97f52.mp4

2026-01-24 17:05:23 UTC

- reply

@nprofile…7ea3 if you need help in fighting bots and spam, I'll gladly help with @nprofile…jku2

2026-01-23 07:39:30 UTC

- reply

This is phenomenal UX
#nevent1q…ga45

2026-01-23 07:33:13 UTC

- reply

WTF I'm exited
#nevent1q…9qjc

2026-01-17 23:56:59 UTC

- reply

I'll definitely try it. Thanks for making it and sharing shawn

2026-01-17 15:20:50 UTC

- reply

Yeah 😂
I was just curious because the thumbnail it's close but not quite the real Jeff

2026-01-17 02:04:48 UTC

- reply

Thumbnail doesn't even look like Jeff

2026-01-15 14:13:17 UTC

- reply

Yo Ver. In the Vertex relay, COUNTs only count tags that have been indexed. For the sake of efficiency I am not indexing "p" tags.

2026-01-13 17:51:21 UTC

- reply

Thanks mate, I appreciate.
I think changing kind is the best way to deprecate the old auth, otherwise there will be confusion and people will try to support both the old and the new spec.

2026-01-13 17:47:45 UTC

- reply

No encryption specified in the spec, but it can use NIP-44, or being sent using White Noise.
https://github.com/nostr-protocol/nips/blob/master/44.md

2026-01-13 15:12:52 UTC

- reply

Link to the NIP PR
https://github.com/nostr-protocol/nips/pull/2187

2026-01-13 15:08:13 UTC

- reply

@nprofile…tfpf we can improve blossom auth

2026-01-13 15:04:57 UTC

Improving signing UX while fixing a replay attack in Blossom.
👋 Say hello to Nostr Web Tokens.
https://blossom.primal.net/0c8193e58b189d6bf16d937f6957c95a4c8a4564d76638d15397108856c529ad.mp4

2026-01-13 13:21:17 UTC

- reply

That's a good take. Yeah the best as you said would be to categorize these users, giving the option to remove from the stats the non-reputsble according to some threshold and ranking algorithm.
Let's continue this convo on Signal

2026-01-13 10:41:08 UTC

- reply

This is cool, but I feel there is a TON of bots and fake profiles causing inflated metrics.
You show the new users of yesterday are 35k, but according to my metrics only ~300 are reputable (followed by at least one reputable).
This would also explain why the retention is basically at the bottom.
I am willing to give lifetime access to @nprofile…y38p so you can filter out some of the crap. WDYT?

2026-01-12 23:03:48 UTC

- reply

Re read my previous message. I am saying default, not always

2026-01-12 16:57:39 UTC

- reply

Me: I think it's bad to display the scores in general. The default IMO should be not to display them
You (quote): You want to use these scores in the background but pretend like you’re not?

2026-01-12 10:42:32 UTC

- reply

Nutmix is becoming one the best cashu mint implementation.
Kudos to Leito for the great work!
#nevent1q…2ua8

2026-01-12 10:40:09 UTC

what is the best NIP-07 extention signer with multi key support?
#asknostr
I have been using alby but now it cleared all my pubkeys and I don't know how to re-import them from their nsecs

2026-01-12 08:22:38 UTC

- reply

Mate, you are misreading what I am saying, again. I haven't even mentioned Vertex, because it's not relevant in any way to this conversation. The provider is totally and absolutely irrelevant.
I am talking about HOW to display these ranks. I don't like the idea of displaying them, whether they come from Vertex, or Brainstorm, or whatever.
Design discussion.

2026-01-11 22:34:24 UTC

- reply

The DM example IS filtering.
If you do it manually every time, based on the score you see on the screen, one could send you 1M DMs...
What I am saying is that we can hide the numbers from the head of people, and place a threshold selection in settings and teach how to use it.

2026-01-11 11:50:06 UTC

- reply

What's your Elo? I am curious because these types of mistakes are made I would guess by sub 1200.

2026-01-10 09:22:24 UTC

- reply

I am talking design and UX.

2026-01-09 23:38:05 UTC

- reply

sorting and filtering, not ranking

2026-01-09 10:07:49 UTC

- reply

exactly

2026-01-09 08:52:16 UTC

- reply

I don't think scores on the forehead are a good idea. It's objectifying people, as if any algorithm is going to be 100% precise in capturing the very contextual notion of trust.
But they don't have to be shown, they can simply be used to increase signal and decrease noise, magically, and in the background.
#nevent1q…d5xa

2026-01-08 13:43:02 UTC

- reply

I've built rely, which is a framework for building very custom relays.
It's not plug and play but you can do anything with it.
https://github.com/pippellia-btc/rely
If you choose to try it and find some issues or the docs aren't clear, let me know, I'll gladly help!

2026-01-02 13:22:56 UTC

- reply

Another great article by @nprofile…u8g0, this time on the Web of Trust.
Nostr is accomplishing what PGP could never do: establishing and maintaining a network of cryptographic identities that vouch for each other, thereby enabling the development of a reputation system.
This is huge
#naddr1qq…5t4t

2026-01-01 03:22:22 UTC

GM good year

2025-12-30 13:45:39 UTC

- reply

Solving equations is much more important than learning how to lay taxes

2025-12-18 16:57:59 UTC

- reply

you can use this endpoint called RankProfile, either by DVM (websocket) or API (http).
https://vertexlab.io/docs/endpoints/rank-profiles/
you just choose the sort algorithm to be "followerCount".
And in the content of the response you get everything you want.

2025-12-09 10:36:40 UTC

Hey @nprofile…waas, I can't access my profile ranking at /profiles.html
https://blossom.primal.net/5ea3f7e8c3e5d814a6344384c1fb58c29b6df0503a60d3c1da3e302baacf6b50.png

2025-12-03 16:28:15 UTC

Hey @nprofile…a2cl the web version has bugs.
https://blossom.primal.net/36597955ae9a4b0dc1e5a423f9ad4f0750edbef3e9a332b3e1064ac8e7ce42b9.png
I got this status http 500 when trying to upload a video. (this image was uploaded fine idk).
I had to post this from coracle because I couldn't even post from Primal

2025-12-03 08:24:06 UTC

- reply

> That’s not how it works. We’re focused on Alice’s Grapevine, not Eric’s.
I must have misread the article, sorry.
> If the users we discussed become a problem, and people start issuing NIP-32 labels to flag the problem users, then GrapeRank is ready to ingest and interpret those labels a turn them into a metric that can be useful.
What if the "problematic" user starts issuing NIP-32 labels back to all that labelled him? Have you considered stability / convergence in these scenarios?

2025-12-02 10:52:05 UTC

- reply

> If she chooses to interpret the data in a manner so that Chuck’s opinion overwrites Eric’s, then her algo should reflect that choice.
Considering the graph you have posted in which there are only follows and 1 report, by the grapevine formula, Chuck would not in the Grapewine of Eric, so the report won't be considered.
> We can see that at npub.world. Search for a popular account and you will typically see multiple impersonator accounts
Yes, because people using inadequate clients have been following these impersonators, so they show up.
> I fully expect some subsets of users to do the YT thing and try to report each other. When that happens, it will be a simple matter to create a flag for users who utilize “weaponized reports” and diminish the weight of reports authored by those users
Where in the formula is this "flag"?

2025-11-29 11:28:41 UTC

- reply

Link farms are a problem for sure, but the way you are presenting it I think is erroneous.
First of all, that targeted report should have never been used.
Eric only follows the bot, and is not connected in any way to chuck, so why should Chuck's opinion overwrite his own? That's the opposite of personalized.
Also, reports can be and have been weaponized: many YT communities have been "at war" against each other, meaning one community members would report the opposite YT channel in order to get it banned.
It's tradeoffs all the way down. I agree that integrating other signals have the potential to improve the ranking, but at the same time it shouldn't create loopholes like the weaponization of reports.
Can you please show the scores that Graperank (default config) would give in both scenarios? That way we can have a deeper and more technical conversation.
The way I am dealing with this kind of link farms is at a earlier step, meanin when I am crawling. Eric, being reputable, by following the first bit would make me include in the database, but then this other bot won't include the others unless it receives many more follows.

2025-11-20 14:29:10 UTC

- reply

I just tried and it seems to be working fine @nprofile…x895. When I have credits it produces a response, when it doesn't it returns kind 7000.
https://blossom.primal.net/cb8379eb0a48e0c7f0050e62f4d5073bdf16e8a97b76bdc4135d5a1d30be5df4.png
npub.world and zapstore are both working as well.
Please share the event request so I'll try to reproduce the issue.

2025-11-20 08:32:21 UTC

- reply

yeah, that's an interesting idea. But what do you mean by personal? A relay that accepts only your notes? Maybe your notes and the ones of your follows? A relay that proactively fetches your follows posts? A relay that does DMs as well?

2025-11-19 10:50:13 UTC

- reply

When I write by hand I find myself in a more creative mood, and it's also calming not to have a screen in front of you all the fucking time

2025-11-17 23:16:29 UTC

- reply

answer: @nprofile…ktem

2025-11-15 09:20:21 UTC

- reply

you didn't, contrary to others, I want my stuff to absolutely work all the time, and if it doesn't, I wanna know why and fix it

2025-11-15 09:16:33 UTC

- reply

damn, I'll take a look sir

2025-11-13 21:49:16 UTC

- reply

We agree with the fact that Nostr is too small right now. But email was small too, but the it grew and I imagine gmail and co. use some paid services or host then themselves internally.
It's so much easier for a user to relate with Amethyst, the client they use, rather than with a service provider they don't know it exists, don't know or understand what it does. Or rather, it does things that are so fundamental they never had to pay for them.

2025-11-13 19:51:53 UTC

- reply

Why do you think clients won't pay? Traditionally apps are made by companies and those companies pay for the services they use.
What makes you say nostr is different in this regard?

2025-11-13 18:50:00 UTC

- reply

It would be great.
I am skeptical whether the end user would actually pay for such low-level primitives that every social media app has built-in, but I'm open to discuss.

2025-11-12 22:56:41 UTC

- reply

Yes it's personalized, but still, searching for "nostr", no profile has "nostr" in the name.
I am pretty sure hazard follows Nostriga, or the Nostr report, or Nostr Band, or nostr build.

2025-11-12 19:31:38 UTC

- reply

Analysing kind:3 is more than enough. The problem is the ranking, where is gives a lot of weight even though these people have the search term in the nip-05 only. It should give more weight to people whose name contain the search term.

2025-11-10 09:45:03 UTC

- reply

are these lists public?

2025-11-06 19:29:15 UTC

- reply

These are fair takes.
No, I haven't considered badges yet in my WoT because few people use them (relatively to follow lists), so the "signal" I could extract would be little to none.
There must be a good UX and incentives for people to assign/check badges for this data to grow in importance. As an advice, I think you should focus on these aspects, meaning UX and incentives.
Why would someone issue a badge/certificate of skill on Nostr? How they have been doing it before? What problems do they have?

2025-11-06 16:21:06 UTC

- reply

It depends what you want to achieve in specific. The problem of categorisation of skills is itself non perfectly solvable, so you should pick heuristics.
These heuristics would inform your nostr design, which would inform your db design. It really is a problem that can have many many different working solutions, it all stems from the assumptions.
For example:
- who makes the categories? You or the users?
- who attributes skills?
- skills are always positive or can be negative (e.g. bad speaker)

2025-11-05 13:33:04 UTC

- reply

Daniel's post doesn't seem to be accusing anyone.
It says "take a closer look"

2025-11-04 22:16:26 UTC

- reply

How about a new library that has only data structures, no dependencies, minimal utilities like a relay pool, Query, Subscribe, and very performant JSON marshalling / unmarshalling?
Man I am so tempted to build this

2025-11-04 21:54:08 UTC

- reply

Yeah It would be nice.
Not to mention the dependencies, which is something I try to be aware.
go-nostr and nostrlib have 30+ dependencies, 2 different libs for websocket, and 3 for json! Like wtf

2025-11-04 21:35:30 UTC

- reply

First of all, thank you for this PR. As I've explained in my other message, I'm not sure if I'll accept it.
Questions for you. How bad would it be for you to migrate your own DB to using a new library? Do you use many of the feature of go-nostr / nostrlib provides other that structure methods (Sign, CheckID...)? If so, which ones?

2025-11-04 16:32:22 UTC

- reply

cc @nprofile…9qsr. I ping you since you previously shared appreciation for rely.
I think you'll love the updates

2025-11-04 11:52:25 UTC

✨ Rely v1 is here.
Build simpler, more performant relays with your own business logic.
You don't have to reinvent the wheel, you have to update your libs.
https://blossom.primal.net/f7ec87bb76426a063c0c361259ec01eaf23fd5cd716ee42d2dd124b9a060df98.mp4

2025-11-04 09:39:12 UTC

- reply

I think you tagged the wrong jb55

2025-11-03 20:01:41 UTC

- reply

or you can just drink and chat and enjoy the moment without bullshit

2025-11-03 19:59:24 UTC

- reply

This will take a while, but it's in the roadmap.
At the moment people can use Vertex for free, up to 100 requests per day.

2025-11-03 17:13:56 UTC

- reply

btw, what are you using Vertex for? Search?