hodlbod

Christian Bitcoiner and developer of coracle.social. Learn more at info.coracle.social.

Merch is available at https://geyser.fund/project/coracle/rewards

Public Key

npub1jlrs53pkdfjnts29kveljul2sm0actt6n8dxrrzqcersttvcuv3qdjynqn

NIP-05 Address

[email protected]

Profile Code

nprofile1qqsf03c2gsmx5ef4c9zmxvlew04gdh7u94afnknp33qvv3c94kvwxgspr3mhxue69uhhyetvv9ujumt4w35ku7thv9kxcet59e3k7mgpp4mhxue69uhkummn9ekx7mqcldwjk

Publishing to

relay.mutinywallet.com

Author Public Key

npub1jlrs53pkdfjnts29kveljul2sm0actt6n8dxrrzqcersttvcuv3qdjynqn

Show more details

Last Notes

2025-05-12 16:26:52 GMT

- reply

Just... use bitcoin

2025-05-07 01:36:36 GMT

New release of Flotilla is out, my first 1.0 release on nostr! Even Coracle never felt quite good enough to call it "ready".
This latest release includes an integration with anchor.coracle.social, which gives me a story for keeping communities engaged with the platform via email notifications. This was the last major piece missing before I felt ready to start onboarding communities.
Obviously, lots of work is still to be done, so give Flotilla a try and let me know if you run into any issues or find glaring omissions!
PS, the new version is coming soon (tm) to App Stores near you, but you can get it today on web or via ZapStore.

2025-05-06 12:55:38 GMT

- reply

I don't see the problem?

2025-05-05 16:02:04 GMT

- reply

This was good, I wish these fediverse people would add nostr as a source, it would fit perfectly. Bluesky is mainstream, mastodon is fiefdoms, nostr is censorship resistant. They could exist in synergy, but everyone ignores nostr.

2025-04-22 18:21:59 GMT

"Here I suggest, therefore, the content of four important acts of remembrance that, in concert, re-narrate the world to the Christian mind in such a way as to call into question the otherwise tacit response to the world in the context of modern alienation. First, such a person must recollect the arguments for and the nature of God's being. Second, one must recollect the primal created freedom of humanity and its relationship to history. Third, the Christian must have an honest assessment of humanity's misuse of its freedom. And fourth, per Christian acceptance of divine revelation and promise, such a person must recall that the deepest human quandary is ultimately not resolved by human attunement to the divine but via divine attunement to the human."
Joseph Minich, Bulwarks of Unbelief

2025-04-21 03:58:43 GMT

- reply

Same, but for BJJ

2025-04-21 03:33:39 GMT

- reply

From what I understand it's because convention was to count inclusively. So he spent time in the tomb on three separate days. But one is enough for me anyhow

2025-04-20 19:40:05 GMT

He is risen

2025-04-19 23:28:21 GMT

I saw my first mushroom cloud when I was 9 years old

2025-04-17 23:11:38 GMT

Just recorded a podcast with C. R. Wiley, and I'm stoked. He is thinking about technology in a deeper and more constructive way than most Christians. Go listen to his podcast: https://fountain.fm/show/arKWodxWbGwOVzgw3ciI

2025-04-17 23:03:23 GMT

Building a micro-app with mithril.js and I'm happy again. No major changes since 2019, no preprocessor needed (it's vanillajs friendly), no magic syntax, javascript only, manage state however you want, it's beautiful.

2025-04-17 00:54:12 GMT

- reply

I only recently discovered your articles, and they're quite good and very applucable to what we're doing here. Welcome to nostr!

2025-04-16 15:04:32 GMT

- reply

I think about this a lot. No one wants to be a plumber, why would they want to go into manufactuting? But I'm still in favor of bringing manufacturing back, if only because it will force us to take a step away from the abstract worlds of financialized or digitized work.

2025-04-15 22:23:44 GMT

Drafted a quick NIP, I think a separate "aliases" kind might be a better way to handle this, since overloading profiles would result in overwriting all aliases with your global profile. Aliases being separate also allows us to avoid having to synchronize the entire profile across all versions.
https://github.com/nostr-protocol/nips/pull/1880
#nevent1q…v3h0

2025-04-15 21:52:57 GMT

I am attempting to add "protected" profiles to flotilla, so that people can create an account, but only share their name with the groups they're a member of.
I implemented this today by adding a`-` tag to profiles and sending them to group relays (as well as fetching profiles from group relays and keeping track of which relay a given profile event came from). This allows for different profiles on different relays, but it breaks the "profile" page because without the context of which relay you're on, you don't know which profile to show if they diverge.
I suppose you could show your "global" profile, with some "aliases" below it. But then you're in the business of managing multiple profiles, which seems really annoying. Maybe an "alias" event kind would be an improvement? You could specify just a subset of profile data, and merge it into your global profile in particular contexts. This could be good for privacy, but also would just be a fun feature to have in groups.
I'm curious if anyone else has worked on this problem and come up with a good solution.

2025-04-15 21:37:34 GMT

- reply

Web of trust, out of band key exchange, lots of ways. The objection you're making doesn't really make sense.

2025-04-15 18:49:00 GMT

- reply

If you sign your data it becomes self-authenticating. True, you'll never have 100% certainty that your data won't be removed unless you self-host, but you can replicate your content across multiple more or less trusted hosts, who would all have to deplatform you at once for it to be effective. You can also keep a back up and re-upload it to a new host at any time. This is good enough for most people, and the option of self-hosting is always open to people who feel they have a higher risk of deplatforming.

2025-04-15 17:37:27 GMT

- reply

It could be a lot of things, likely the algorithm for relay selections deciding not to select your haven relay. How many relays do you have and what's your relay limit (app settings > Max relays per request)?

2025-04-14 23:33:57 GMT

I don't know who puts the Runner's World marathon plans together, but 2 miles + 24 minutes of hills does not equal 5 miles

2025-04-14 17:05:35 GMT

Some interesting design patterns that might be applicable to nostr apps in some cases: https://www.prosocialdesign.org/

2025-04-13 22:24:52 GMT

When asked to explain his choice to major in physics, Wolfgang Smith said, "I want to study physics because I believe that physics is the key to the understanding of the universe." He amends, "Needless to say, I have changed my mind in the interim."

2025-04-11 20:53:07 GMT

“mom, we didn’t use our bows, just our arrows, and we hunted down a rat and killed it with our arrows in the forest and now I’m getting gloves to go bring it to show you”

2025-04-11 15:01:33 GMT

Let's give the people a faster horse
https://rakhim.exotext.com/but-what-if-i-really-want-a-faster-horse

2025-04-08 19:33:24 GMT

- reply

Does this do the trick?
https://coracle-media.us-southeast-1.linodeobjects.com/flotilla-tour.mov
I also added it to the website.

2025-04-08 18:30:23 GMT

New episode on FROST is out:
#nevent1q…llpm

2025-04-08 17:53:49 GMT

- reply

Sure, I'll put something together

2025-04-07 17:05:42 GMT

Hey @nprofile…fksp I'm getting an error when trying to zap @nprofile…mqyq: "Could not generate a lightning invoice for this project". Can you tell me what's going on there?

2025-04-07 16:09:02 GMT

- reply

Sorry, I am mid refactor, I should have released an alpha version. Pin welshman to the previous version or use what's in the lockfile. I'll hopefully be able to update flotilla this week

2025-04-07 03:15:25 GMT

- reply

NIP 29 support is not needed for Flotilla or Chachi

2025-04-06 15:21:15 GMT

I'm learning go, rate my game
https://cdn.nostrcheck.me/97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322/71b036982007a3f0523e388092d09c30021bca8338dc2706df76cfc2f8bf2252.webp

2025-04-05 23:14:14 GMT

- reply

💯

2025-04-05 13:21:38 GMT

- reply

Check out bitvora for some recent development in the personal relay space. I don't know if they have tutorials, you're right that the user documentation is awful in general.
https://github.com/bitvora

2025-04-04 23:18:25 GMT

- reply

A lot of lists are also out of date. The first wave of projects is probably 90% dead, which can give the illusion that the whole space is too

2025-04-04 17:59:29 GMT

- reply

Thanks, it's a big day for me

2025-04-04 15:58:16 GMT

- reply

We shouldn't merge the PR, relays should continue to support multiple filters. But, having implemented this in coracle recently, it's actually super nice because it encourages callers to split requests by filter, which results in slightly faster response times for each result set.

2025-04-04 01:38:09 GMT

- reply

To do it for real, yes, I completely agree

2025-04-04 00:52:35 GMT

- reply

An attempt at building go on nostr: https://github.com/coracle-social/nogo/blob/master/index.html

2025-04-04 00:38:32 GMT

- reply

Keep recording it, thinking about it, even prototyping. Nostr has a lot of momentum right now, so I don't think a toy protocol will divert substantial effort away from nostr. But keep it visible enough so that if/when a successor is obviously needed we have a good starting point. It's a good project, and I think (at least for now) it "counts" as nostr

2025-04-04 00:35:45 GMT

Feeling pretty good about myself, just vibe coded 700 lines of utter garbage in only 30 minutes

2025-04-04 00:20:05 GMT

- reply

I am 35 and I still don't know how to use commas

2025-04-03 21:01:59 GMT

- reply

I mean that John may have been a nazirite

2025-04-03 17:38:01 GMT

- reply

👀 I will take a look

2025-04-03 17:36:31 GMT

- reply

Oh yeah, I forgot, that's Gigi 😂

2025-04-03 15:22:13 GMT

- reply

If not, click "skip this step" and go to the relays tab and choose some relays

2025-04-03 15:21:54 GMT

- reply

Have you published relay selections? I can't find them on your profile. If you haven't created a 10002 event, coracle will fail this way to avoid accidentally overwriting important metadata events.

2025-04-03 13:45:10 GMT

- reply

Have you tried coracle.social?

2025-04-03 00:27:09 GMT

- reply

Block their IP? Ignoring closed is definitely bad behavior. EOSE + CLOSE is forcing me to do weird hacks in coracle

2025-04-02 21:36:03 GMT

@nprofile…dwak auth.nostr1.com is sending an EOSE along with a close, which doesn't seem right to me. Is this deliberate, or a bug?
https://cdn.nostrcheck.me/97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322/d6111b09e4df10ac4dfedc089afd6a7e13850bfd9dc9de5f344612e582e1278f.webp

2025-04-02 17:51:20 GMT

- reply

I can't imagine being someone with more time than ideas, but somehow that seems to be extremely common with developers

2025-04-02 16:14:21 GMT

- reply

Slow, I'm currently shaving yaks in the himalayas

2025-04-02 16:13:48 GMT

- reply

Yeah, there were some other red flags in the docs, like:
https://cdn.nostrcheck.me/97c70a44366a6535c145b333f973ea86dfdc2d7a99da618c40c64705ad98e322/9e764a50d7d2361693dd671ae0c14cd5e876f6183e74b6a239dffb7df00980a1.webp
So we'll see if they deliver anything worth using

2025-04-02 15:20:28 GMT

- reply

Big if true, I love the idea of a lower-level backend + event sourcing. This might check the box for @nprofile…20a8 as well, he and @nprofile…jyqh were talking about decoupling application logic and rendering recently.

2025-04-01 19:19:01 GMT

- reply

https://tenor.com/i6ZfyIZDWKl.gif

2025-04-01 17:17:27 GMT

- reply

Ah, nice, that's the opposite approach to what I'm doing but it makes plenty of sense

2025-04-01 12:12:02 GMT

- reply

I made it opt in because it requires additional trust and additional processing. I think it was a good call for now at least

2025-03-31 22:58:17 GMT

- reply

There is no authority, just like nostr. Embrace the chaos

2025-03-31 18:09:19 GMT

- reply

Thank you!

2025-03-31 15:44:32 GMT

- reply

Because I want a better internet. I have values other than efficiency and profit. Idealistic, I know, but why work on any of this stuff otherwise?

2025-03-31 15:43:11 GMT

- reply

That's a slick interface, I like it a lot. The main reason I wouldn't use this implementation is that a lot of the policy is baked in, which might need to be configurable by the application. I have a bunch of policy stuff here: https://github.com/coracle-social/welshman/blob/net2/packages/net2/src/policy.ts which is mostly reasonable defaults, but if users wanted to configure auth policy per relay I don't see how that could be done in applesauce. Also, why no`subscribe` ?

2025-03-31 15:38:00 GMT

- reply

You're welcome! Notifications being missed depends a lot on which relays they're published to, if they're not sent to your read relays they won't get picked up. Lots of clients aren't good about this, but if you can narrow down which client sent them I can look into it. And of course coracle has bugs too, some of which I'm attempting to refactor away this week.

2025-03-30 14:08:57 GMT

- reply

HIFO FTW

2025-03-30 03:47:04 GMT

- reply

Hmm, sorry I missed it, mind trying again?

2025-03-29 23:25:25 GMT

- reply

😂

2025-03-29 14:16:34 GMT

I'm honored
#nevent1q…ca5a

2025-03-27 21:25:08 GMT

- reply

You really are from the future

2025-03-27 18:26:24 GMT

- reply

Thanks, I might just do that. For now, I'm probably going to go with emitter version since I really need to get this refactor done. But let me know when you publish the rxjs relay stuff, I'd love to learn/contribute there.

2025-03-27 18:23:30 GMT

- reply

This is the most nuanced take I've seen on this subject yet.
First of all, yes, I agree github is not ideal and we should move NIPs elsewhere. But so far no one has stepped up to build a viable replacement, so for now we have to deal with the two-tier system through culture, by encouraging devs not to worry about things not getting merged. I personally have 10+ unmerged open PRs which are in varying levels of implementation in my stuff. Having them unmerged is 1. an invitation to collaborate, and 2. a signal that they're not adopted widely.
As far as lexicons go, when I was designing my own protocol that was the direction I was taking things. I'm still sympathetic to the idea, except for what you pointed out about control over the lexicon. Numeric identifiers are much better, because they're neutral — *if* we can get into a good habit of publicly signaling their use and adoption. That's not trivial of course, and points back to my first paragraph.
As far as specificity, the terseness and ambiguity is (I think) one of nostr's strengths. I could be wrong about this, because it basically goes against all the conventional wisdom of specification writing, so we'll see if it works out in practice. But specifying the minimum interoperability surface area allows there to be more flexibility in how a given data format can be used. This can lead to inconsistent UX, which is a problem but something devs are motivated to work through because it aligns with their interests, or it can lead to innovative new ideas. In either case though, it forces devs to be more conservative about how they treat events, encouraging defensive coding (in case of deliberately malformed events as an attack) and wider interoperability. This is an adaptive approach rather than a validate-and-throw-away approach.

2025-03-27 17:47:00 GMT

- reply

Yeah, I think I understand all that in the abstract, I'm just finding it quite hard to actually execute, partly because of the duplex nature of websockets, which is what I've been focusing on far, and partly due to the forking nature of streams, which makes me unsure what subscriptions need to be managed and which don't.

2025-03-27 17:43:41 GMT

- reply

Me too, and getting deeper. I hate falling down these holes 😂

2025-03-27 17:02:05 GMT

- reply

I do too. @nprofile…rxj5 is there a repository with wip code anywhere?

2025-03-27 16:04:16 GMT

- reply

Hey, did you guys get my DM? I'd still love to get something set up.

2025-03-27 13:07:53 GMT

- reply

Most apps do follow this. It's a dafinition based on that "plain text" recommendation, @nprofile…ucgf just didn't want us to use the term plain text, so we made up a new name.

2025-03-26 23:30:00 GMT

- reply

Dang, sorry to hear that! I am re-writing some core networking code, which should help with general stability, but it sounds like the system might just be getting clogged up with those hellthreads. I'll have to do some load testing at some point.

2025-03-26 21:01:13 GMT

- reply

Yeah, the tricky part seems to be intercepting send messages so that you can buffer them and re-send after authentication (or remove reqs that have been closed before they were sent while you were waiting for auth). Here's my (half-baked) policy for that:
```
export const socketPolicyDeferOnAuth = (socket: Socket) => {
const send$ = socket.send$
const buffer: ClientMessage[] = []
const authState = new AuthState(socket)
const okStatuses = [AuthStatus.None, AuthStatus.Ok]
// Defer sending certain messages when we're not authenticated
socket.send$ = send$.pipe(
filter(message => {
// Always allow sending auth
if (isClientAuth(message)) return true
// Always allow sending join requests
if (isClientEvent(message) && message[1].kind === AUTH_JOIN) return true
// If we're not ok, remove the message and save it for later
if (!okStatuses.includes(authState.status)) {
buffer.push(message)
return false
}
return true
}),
)
// Send buffered messages when we get successful auth
return authState.subscribe((status: AuthStatus) => {
if (okStatuses.includes(status)) {
const reqs = new Set(buffer.filter(isClientReq).map(nth(1)))
const closed = new Set(buffer.filter(isClientClose).map(nth(1)))
for (const message of buffer.splice(0)) {
// Skip requests that were closed before they were sent
if (isClientReq(message) && closed.has(message[1])) continue
// Skip closes for requests that were never sent
if (isClientClose(message) && reqs.has(message[1])) continue
socket.send(message)
}
}
})
}
```

2025-03-26 20:08:52 GMT

- reply

Here's a question: I feel like I started wading into rxjs with a very hard problem, websocket connections. Websockets are shared resources, and so need to be modeled using a subject somewhere along the line. They're also bi-directional, and so I'm not sure how to pass them around in an idiomatic way. I'd like to define a function like`implementAuthPolicy(tx, rx)` that can intercept and buffer both sent and received messages. Maybe that's less of a question than me just thinking out loud.

2025-03-26 19:54:42 GMT

- reply

You're missing out

2025-03-26 19:54:15 GMT

- reply

Thanks, that rfc looks really nice. I also looked through nosotros/core and the code is super clean and includes a number of patterns that took a while for me to identify (like the relayFilters stuff). Have you looked into https://github.com/penpenpng/rx-nostr at all? My impression was that everything was pretty tightly coupled, making it hard to implement custom policy. Speaking of which, how do you handle relay AUTH when using websocketSubject? I am using policy functions that patch my socket adapter: https://github.com/coracle-social/welshman/blob/net2-rxjs/packages/net2/src/policy.ts

2025-03-26 18:28:33 GMT

- reply

Sounds like fun

2025-03-26 18:27:35 GMT

- reply

Because I'm writing a web client

2025-03-26 18:12:21 GMT

- reply

Cool, that's good to hear. I was like "cool, a way to avoid memory leaks". Then, 5 hours in realizing I had to deal with memory leaks.

2025-03-26 18:05:39 GMT

- reply

As in, you tried it, and went back to imperative event listeners?

2025-03-26 18:03:27 GMT

- reply

@nprofile…h7nv

2025-03-26 18:03:09 GMT

Attempting to learn RxJS because it seems like streams would be the correct way to model events flowing from a relay through various utilities for welshman. I'm currently learning about how to manage subscriptions using`takeUntil` , and it just seems like footguns and strange mental models abound.
I'm interested in others' experience with Rx*. Is it the amazing paradigm shift people say it is, or does it just make your code more obscure and come with its own problems? In anything less than something like Elm (which I do like), it seems like all you can do is move around the imperative parts of your code, not get rid of them entirely. I had my love affair with point free via ramda a number of years back, and eventually left it for non-curried utility functions (typescript especially made ramda painful). This feels very similar.
https://i.imgflip.com/9opw9g.jpg

2025-03-26 16:09:40 GMT

- reply

That vegan diet is really working for you

2025-03-26 13:03:17 GMT

- reply

👀👀👀

2025-03-26 13:02:57 GMT

- reply

There is always that danger

2025-03-26 03:23:57 GMT

- reply

Completely agree, this was never going to be anything other than a long, slow, enjoyable grind

2025-03-26 00:26:57 GMT

Hello from https://enterthenostr.com/

2025-03-26 00:24:57 GMT

- reply

Nice, that's a good one, I hadn't heard it before.
"content to fill a little space, if thou be glorified." is brutal 😂 but in line with John B's "he must increase, and I must decrease"

2025-03-26 00:22:48 GMT

- reply

Correct, it probably would have been better to use the bookmarks list nip (although that didn't exist at the time)

2025-03-25 21:19:20 GMT

- reply

No, I never got around to building that unfortunately. Collections is still in the "proof of concept" phase.

2025-03-25 21:18:28 GMT

- reply

I agree strongly with your point about "faithful presence", while at the same time mostly failing to content myself with it. The humility and diligence that being "ordinary" requires of us is way beyond what's required to have higher (and more abstract) aspirations.
I also agree with the idea that "we need to content ourselves with it being just plain _good and faithful human work_ that helps our families and benefits our neighbors", but I also see that often becoming a reason for complacency in practice. For example, Christians often just invest in index funds without thinking about what kind of activities they are becoming complicit in.
The balance between grand vision and being "faithful in little" is so difficult I feel like I can't fit both in my head (let alone heart) at once.

2025-03-25 21:12:24 GMT

- reply

😂💯

2025-03-25 16:16:00 GMT

- reply

I see you have only four people muted, I assume you had more before? Just to double check, are you using multiple clients/devices, and if so which ones?

2025-03-25 16:14:50 GMT

- reply

If you're right, I think I'm in danger of committing the sort of democratic "change the world" fallacy, where everything you do has to have global significance. I recognize that this isn't possible for everyone (and so can't be a prerequisite for a meaningful life), or even necessarily healthy for people who adopt it, but it's hard for me to let go of, and a big part of why I'm here on nostr. At the same time, I think we should have a cosmic view of the meaningfulness of our actions. How do you reconcile those two things? I'm probably just restating the question, but puzzling over this has been a challenge for me for years.

2025-03-25 15:36:22 GMT

- reply

This is the answer I was looking for, thanks!
I swim in mostly post-millennialist circles, and have lately started to feel the effects of over-realized eschatology you identified above. At the same time, I feel like God must have a place for our work, just because we spend so much time doing it. The amil answer isn't completely satisfying to me, because it seems to limit the scope of the effect our new identity as Christians has on our whole lives, but I think your reference to Jeremiah is also a good answer to that — we "mind our own business and work with our own hands" to build up the city of man not as such, but for the peace of the people within it (either evangelistically, or for the Church as a separate enclave).
Thanks for your thoughts, I'll check out those VanDrunen books.

2025-03-25 02:06:53 GMT

- reply

Tell me what you really think of me 😂

2025-03-24 22:10:59 GMT

- reply

No, haven't had a chance to look into it yet, this week

2025-03-24 20:19:32 GMT

- reply

Have you seen https://github.com/bitvora/wot-relay

2025-03-24 17:54:11 GMT

- reply

Yep, exactly. PZP solves this with "tangles" and I think p2panda has a similar data structure. There has also been talk on and off about adding synchronization or merkle tree stuff to nostr as well. I think it's a feature rather than a bug.