sedited

thecharlatan.ch
🇿🇦

Public Key

npub196cr58e3ds70njgqul6ndm3gu4yxxjgx00sp32t95lru5k600u7q2ftlhg

Profile Code

nprofile1qqszavp6ruckc08eeyqw0afkac5w2jrrfyr8hcqc49j60372td8h70qpp4mhxue69uhkummn9ekx7mqppamhxue69uhkummnw3ezumt0d5q3camnwvaz7tmwdaehgu3wvf5hgcm0d9hx2u3wwdhkx6tpds750jan

Publishing to

nos.lol

nostr.mom

nostr.bitcoiner.social

Author Public Key

npub196cr58e3ds70njgqul6ndm3gu4yxxjgx00sp32t95lru5k600u7q2ftlhg

Show more details

Last Notes

2026-02-27 18:27:13 UTC

- reply

Exactly.

2026-02-21 15:00:15 UTC

Cash on the internet.
No auto-updates.

2026-02-15 09:54:36 UTC

- reply

Or now again.

2026-02-06 18:08:22 UTC

- reply

> Its increasing
There's not been a significant increase in usage of large op returns over the past half year. The utxo set has been steadily shrinking in the same time frame.
The chart you showed seems to be measuring small op_returns too, which bip110 does not stop.

2026-02-06 13:00:48 UTC

- reply

Not against stopping spam, strongly disagree with the method.

2026-02-02 06:31:40 UTC

I wonder how many people still self host their own lightning node without the help of a third party.

2026-01-24 19:22:59 UTC

Sacha's fancy stuff not working today, but seems like stuff is not gelling in general.

2026-01-23 15:36:26 UTC

First proper work week in the new role.

2026-01-22 21:45:15 UTC

- reply

It is south :)
It's in my home town's area.

2026-01-22 16:59:39 UTC

Can't wait to be going back here again soon. https://image.nostr.build/15fe805ee35dc870e04b094a62dbaad4263187990a7ec1f150644da5cc022a0e.jpg

2026-01-20 23:08:18 UTC

This was solved by Marko Falke here https://gcc.gnu.org/pipermail/gcc-patches/2026-January/704817.html after days of sleuthing and bisecting. Looks like the underlying problem for this is good old c++ not having a canonical order for argument evaluation. Evalaution may work differently on different platforms, assigning a different name.
#nevent1q…jefe

2026-01-19 12:22:00 UTC

- reply

In short: The gpg version I had running back then still used sha1 has a default digest to generate the base signature attesting to the first subkey generated automatically during key generation. Like a good citizen, I changed the default digest algorithm to something strong, generated the remaining [a]uth and [e]ncrypt subkeys with this configuration, and changed the expiry on all of them. Turns out, gpg never refreshes the base signature even you tell it to update a subkey, or change its configuration. So when the Bitcoin Core commit signature verifier ran with a strict no weak algos policy it tripped over this weak base signature.

2025-11-02 11:18:43 UTC

- reply

Since v25:
Cutting initial blockchain download time in half, encrypting p2p communications, making lightning safer and cheaper to use, allow immediate bootstrap with assumeutxo, shipped better miniscript and descriptors support, easier compilation with cmake, fixed a bunch of CVEs that have been present since a long time, complete obfuscation of on-chain data, better coin selection, the mining interface for easy home mining with stratumv2, better and safer localnet hole punching with natpmp and pcp, and a whole bunch more.
You can read all about it at https://bitcoincore.org/en/releases/ .

2025-10-24 06:23:09 UTC

- reply

Two projects I know well, liana and the bitbox02 firmware, are written in rust and have reproducible builds setups. But maybe you mean something else?

2025-10-24 05:49:24 UTC

- reply

rustc does reproducible builds.

2025-10-23 21:15:12 UTC

- reply

Some actual movement happening here too!

2025-10-18 16:50:56 UTC

- reply

Yeah, shaky reasoning at best.

2025-10-18 16:49:02 UTC

- reply

Indeed :)

2025-10-10 16:21:46 UTC

- reply

What is the root cause that needs fixing here?

2025-10-10 16:21:00 UTC

- reply

These are not potential problems. We already saw this play out during a spam hype in 2015. Spammers moved to fake key outputs, which lead to unnecessary permanent bloat utxo set bloat. It is likely that such a cycle will occur again that it happened three times already now.

2025-09-30 21:00:05 UTC

- reply

Mmh, I'd actually be interested in giving you guys a visit some time. Would anybody there be interested in some of the work I've been doing with Core's consensus code? Will miss out on the conference again though, you always seem to be picking the dates around major events in my life :P.

2025-09-26 13:26:07 UTC

- reply

Actually I think that would be pretty cool. Probably also not too crazy of an effort, since it is well specced.

2025-09-26 11:30:50 UTC

- reply

Wow, great recruitment effort!

2025-09-26 06:00:16 UTC

- reply

> Conservative and well reviewed.
That is Core to be frank.
Should it ideally reimplement consensus, or use core's engine?

2025-09-21 21:23:56 UTC

- reply

There is a distinction and it indeed gets a bit fuzzy. I.e. outputs bigger than 64?bytes (iirc) require extra allocations when writing to disk, but obviously that's no argument for prohibiting other types. A weaker reason with stronger (lol!) assumptions about its usage is that it should be theoretically spendable unless marked as such. Some script pubkeys might not be, which is a reason to only allow "types" that at least formally are.

2025-09-19 15:26:58 UTC

- reply

Mmh, who might have written that? 🤔😄

2025-09-14 15:15:02 UTC

- reply

That is not what it does, it just tries to be diverse across different asans. It's not a shun list.

2025-09-12 20:58:22 UTC

- reply

I've seen this argument based on human action many times. I don't get it. A policy rule is only effective if everybody applies it. How does that account for human action? On the contrary it seems like a centrally planned and executed policing action. A keynesian argument in my book.

2025-04-05 14:11:30 UTC

Review progress on the kernel API has stalled over the past month. Not sure what to do about that. Maybe there are just not enough people that care?

2025-02-02 18:58:58 UTC

- reply

I love this clip so much.