Râu Cao ⚡

Redecentralizing the Web w/ remoteStorage (npub1v2p…tuxy) and Kosmos Cooperative (npub1286…6nf2). Evading winters w/ Hacker Beach (npub1rnm…7590). Traveling full-time since 2010. INTP.

Move slow and fix things.

:xmpp: / ✉️ / :nostr:: [email protected]
Cycling, coffee, beer, food: Yes, please!

Public Key

npub1z6gtynkn398jcf42xclzjnnw2jm7qglga0m5c9kwpg39ep70cs6s363yt9

NIP-05 Address

[email protected]

Profile Code

nprofile1qqspdy9jfmfcjnevy64rv03ffeh9fdlqy05wha6vzm8q5gjusl8ugdgpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqg5waehxw309aex2mrp0yhxgctdw4eju6t03gm294

Publishing to

relay.primal.net

relay.damus.io

nos.lol

Author Public Key

npub1z6gtynkn398jcf42xclzjnnw2jm7qglga0m5c9kwpg39ep70cs6s363yt9

Show more details

Last Notes

2026-07-09 10:38:22 UTC

- reply

What is the protection?

2026-06-22 07:32:27 UTC

> Dialog, a private invitation-only organization cofounded in 2006 by billionaire tech investor Peter Thiel, has spent two decades refusing to disclose its membership.
> The records sat in Airtable, a commercial database, and included for each participant their membership status, every retreat attended, biography, home city, and a private access token functioning as a login credential.
Should've invited an actual software engineer to help with the organization I guess.
https://securityaffairs.com/193880/intelligence/peter-thiel-secret-society-leak-creates-a-perfect-target-list-for-espionage-influence-operations-and-blackmail.html

2026-06-01 15:13:12 UTC

- reply

Why not just use a proper OS?
https://s3.kosmos.social/media_attachments/files/116/675/495/114/161/610/original/b47e76517481b767.png

2026-03-24 12:24:58 UTC

- reply

That page is 403 for me.

2026-03-24 12:15:01 UTC

- reply

"across Germany"?

2026-03-10 08:27:49 UTC

- reply

The other reason is less CO2 from combustion, and cleaner than coal. But yeah, shipping it in liquid form eradicates the emissions savings.

2025-11-04 12:42:46 UTC

- reply

It should be called Lightning Address, not lnaddress.

2025-10-27 16:03:40 UTC

- reply

In short: here entire point is based on the fact that they want to be one centralized provider servicing people everywhere on the globe. Federation with many local servers would obviously scale better for the vast majority of their users.

2025-10-03 07:31:08 UTC

- reply

Oof, I hope this was the right one!

2025-09-19 14:32:35 UTC

- reply

Yes, that's likely. Also, it's not deprecated in favor of Lightning Address, because LN Address is outside of the Lightning Protocol and requires HTTP servers. The actual/native replacement would be Offers I think: https://github.com/lightning/bolts/pull/798
The problem with keysend is that, as Oscar mentioned, nobody will implement this in new software, and there's already a lot of software that doesn't support it. Meanwhile, Lightning Address has gained enough popularity.

2025-09-19 14:24:44 UTC

- reply

I'm not sure who "they" is supposed to be.
In LND, it's described here in the sample config for 4 years now: https://github.com/lightningnetwork/lnd/blob/master/sample-lnd.conf#L520-L523
And the default has been set to not accept keysend payments over 2 years ago: https://github.com/lightningnetwork/lnd/commit/178f646692e7911d635af7d478cac4d8dae0c60e

2025-09-19 12:32:11 UTC

- reply

That's definitely a 0% chance, since it had been deprecated years ago, not even mentioning ease of use/understanding for customers.

2025-06-23 18:05:12 UTC

If you want something with the format and price tag of a Mac Mini, but for running #Linux, these look really good:
https://www.bee-link.com/products/beelink-ser8-8845hs

2025-04-25 08:53:51 UTC

- reply

Honestly, I'm not sure what you're talking about. I travel around the globe *a lot*, and I have more issues with centralized services being blocked than I ever have with standard XMPP ports. I can't even remember the last time I wasn't able to use XMPP on standard ports without a VPN.

2025-04-24 11:51:03 UTC

- reply

https://s3.kosmos.social/media_attachments/files/114/392/790/463/527/375/original/39d977ca48e7760d.mp4

2025-04-24 09:15:19 UTC

- reply

Attachments are usually not sent directly in XMPP messages, but uploaded to an HTTP endpoint. You might want to find out why that is failing.

2025-03-19 10:57:47 UTC

- reply

https://s3.kosmos.social/media_attachments/files/114/188/736/561/317/817/original/ea47415cd4638687.png

2025-02-27 10:41:39 UTC

Hey #fediverse devs, is there a comprehensive resource for deployed, real-world ActivityStreams vocabularies?
We'd like to make Sockethub somewhat interoperable with whatever is used on the rest of the Web (eventually).
Example: Object for a Web page. There's a simple Page object described in the Activity Vocabulary spec, but it's nowhere near complete in terms of metadata properties. How can I find out what others are doing, short of reading every possible implementation's source code?

2025-01-29 15:20:35 UTC

Oops.
https://predictors.fail

2025-01-19 16:00:54 UTC

- reply

They wasted my time reporting 3 of your posts, claiming that you're "simping for a fascist". The only posts that violate either of our servers' rules were their own. 🤡

2025-01-19 14:50:15 UTC

- reply

That person consistently cops out of discussions at stage 2 of this pyramid:
https://s3.kosmos.social/media_attachments/files/113/855/571/575/723/199/original/d67e6c310001d8e7.png

2025-01-15 23:10:53 UTC

The Indian e-visa application is killing me.
https://s3.kosmos.social/media_attachments/files/113/834/881/592/055/655/original/e0c3bcc09c4ee442.png

2025-01-09 15:45:51 UTC

- reply

In theory, yes. Not sure what happens if they need to reconnect and yours doesn't initiate the connection. It's obviously better if your node is reachable in some way.

2025-01-09 15:43:42 UTC

- reply

A connection is not the same as a channel. You have to first establish a p2p connection to a node, and then you can open a channel.
Yes, the other option is to just send sats out to gain incoming liquidity. You could even swap out the entire channel's balance to an on-chain address using something like https://boltz.exchange/swap for example.

2025-01-09 14:16:12 UTC

- reply

Without either a public IP or an onion address, nobody will be able to establish connections to your node from theirs, only the other way around.

2025-01-09 14:15:12 UTC

- reply

Looks like it neither has a public IP, nor is it advertising an onion address for Tor. It is wrongly advertising its listening address, which is 0.0.0.0 (which just means all network interfaces on your own machine).
https://1ml.com/node/035ad2c954e264004986da2d9499e1732e5175e1dcef2453c921c6cdcc3536e9d8

2025-01-03 16:21:18 UTC

- reply

https://s3.kosmos.social/media_attachments/files/113/765/336/529/254/808/original/17052a3d616be893.mp4

2024-12-23 07:47:44 UTC

Woke up to snow outside! With a little bit of luck, this is going to be the first White Christmas in decades for us.
https://s3.kosmos.social/media_attachments/files/113/701/031/029/422/183/original/b308df434d9523d5.png

2024-11-23 20:42:52 UTC

- reply

I don't get it. Now it's less flexible, because everyone has to implement that endpoint, whereas the user address itself (user@host) can be resolved to whatever anyone wants in the future. You can still self-host it and point to all the options, and then point it to a different provider, too.
Just as an example, here's how simple it is with Nginx: https://gitea.kosmos.org/kosmos/chef/src/branch/master/site-cookbooks/kosmos_website/templates/nginx_conf_website.erb#L33-L37
(Has to be slightly adjusted depending on the situation, but not by much.)

2024-11-23 17:19:26 UTC

- reply

The Lightning Address is the abstraction already. What is the benefit of making another request to an options endpoint instead of directly to the various options, in order of priority/preference?

2024-11-23 17:10:00 UTC

- reply

I don't understand what this is for. We already have two different endpoints for the invoice and keysend methods:
https://kosmos.social/.well-known/lnurlp/raucao
https://kosmos.social/.well-known/keysend/raucao
Bolt 12 should be yet another one, and clients can implement fallbacks if their preferred method isn't available.

2024-11-11 10:11:05 UTC

It's amazing how every new #bitcoin all-time high reliably brings out the people that don't understand it even in the slightest, and who somehow have the urge to re-broadcast every single misunderstanding of the last 15 years to the world, instead of finally looking into properly.

2024-11-07 10:08:31 UTC

- reply

I think you need to use the beta?
"2024-09-16: GPG Suite beta includes a GPG Mail version that supports the just released macOS Sequoia."

2024-11-06 12:55:52 UTC

- reply

Thunderbird does. It's not the most macOS thing, but it's a decent email client with native PGP support.

2024-09-26 16:57:56 UTC

- reply

That would be the ideal scenario, because it would allow private DMs across protocols.
Nobody has it implemented for any decentralized protocol yet, so if we don't all ignore each other's work, we might actually be able to make that happen somehow. However, I'm still not very hopeful that everyone can agree on the details. But surely we can try at least.

2024-09-26 16:24:37 UTC

- reply

Ultimately, it would be amazing if they could all be made interoperable, of course, but I doubt that everyone will ever agree on the same algos and other details. But still, everyone would need the same MLS primitive functions for their Web clients, so that's why the library/API question is interesting to me.

2024-09-26 16:23:05 UTC

- reply

Yes and no. The mid- to long-term plan is to add private group chats to a Web chat client of ours. So I'm a bit more interested in MLS over XMPP, and I'm following work for MLS over Nostr, which are both good bases for private group messaging, because users of those protocols already have key pairs and encryption for 1:1 chats in place. Also, the client currently only does XMPP and IRC.
But I want to follow the AP work on it, too, because it could be added to the same chat client as well.

2024-09-26 15:48:51 UTC

- reply

Thanks! Yeah, it seems to me like a high-quality library would solve the same problem, and we'd have to polyfill one anyway, for browsers that don't support MLS primitives natively yet. In the long run, it could increase performance by not having to download that code, of course.
I asked, because I had just looked into the library situation a couple of days ago, and it seems like there's nothing available in TS/JS at the moment:
https://kosmos.social/@raucao/113192277991674267

2024-09-26 14:57:20 UTC

@npub1764…nuj0 I just read your blog post, where it says "One thing I learned is that Mozilla is working on an MLS implementation in the browser". Did they/you mean that as they're working on a Web API for MLS, which they want to implement in Firefox first? 🤔

2024-09-12 13:52:23 UTC

- reply

That can't be the end goal of the protocol design. Are you saying their goal is to never be able to self-host the full stack of services? Since you can use your own domain, how can you not just point to whatever you want for all of them?

2024-09-12 13:40:20 UTC

- reply

How could they enforce that for your own instance? Wouldn't that just be a restriction for their own web UI and official client builds?

2024-09-09 15:47:34 UTC

TIL: You can comment Regular Expressions in #ruby:
https://thoughtbot.com/blog/comment-your-regular-expressions

2024-08-28 05:57:17 UTC

- reply

https://s3.kosmos.social/media_attachments/files/113/038/106/700/180/668/original/3a77ff3c7903767a.mp4

2024-08-22 08:42:14 UTC

> Kopra Bio is curing cancer using viral immunotherapy technology we developed at UCSF. We’re making the next Keytruda, a $25B/yr cancer drug blockbuster, starting with the most aggressive form of brain cancer, glioblastoma. In preclinical models, we improve survival rates from 0% with standard-of-care treatment to greater than 90% survival with our treatment.
👀
https://www.ycombinator.com/launches/LeF-kopra-bio-making-cancer-cures-using-viral-immunotherapy

2024-08-14 07:11:06 UTC

- reply

Which features would that be? 🤔