Last Notes
Yes, bad outcomes definitely possible: they've happened with previous technologies, too. But extrapolating to super-intelligent machines has a long history: the real damage was elsewhere.
I've disagreed with Bryan about several things, but this I agree with 100%.
https://bcantrill.dtrace.org/2026/09/13/the-contagion-of-fear/
Hi from https://cashu.centurymetadata.org
I take birthdays seriously!
https://npub179e9tp4yqtqx4myp35283fz64gxuzmr6n3yxnktux5pnd5t03eps0elz4s.blossom.band/dd4c1dec153d7b2619e2c115b344d4d250ad89832f94f25a7bfd1d38b0dc127a.jpgbirthday
Time for a Rocky Horror / Muppet Treasure Island double feature!
Don't confuse a 2 week embargo with closed source.
They're in a hard place.
Manager of the Lightning Team, she signed the last 5 releases?
You can also restart with
--developer --dev-no-version-checks --subdaemon=openingd:/DNE --subdaemon=dualopend:/DNE
This prevents new channels from opening (by telling it to use subdaemons which don't exist: the dev option stops it checking at startup).
If you think your node is unlikely to be messed up by existing peers, and you're just worried about new ones.
#nevent1q…uk05
Calle misread. It says upgrade when the release comes, or go --offline.
It absolutely did not recommend "shut down NOW" 😕
Very few builders in this space are actually malicious. But many need to get out and touch grass.
Myself not excepted!
Was he actually malicious? I disagreed with him, but there is a difference.
Yes, classical C has the problem that operations are low-level, so if you overrun the end of a buffer, instead of a crash report the attacker can make arbitrary commands run. Also, manually deallocating memory can be omitted (oops, program grows over time, crashes) or done twice (also classically tricked into running arbitrary code).
Also, the standard routines are minimal, forcing programmers generally to deal with these issues everywhere.
CLN has a *lot* of internal infrastructure to avoid code having to do this. Firstly, the tal hierarchical allocator removes most manual deallocating, and all the infrastructure relies on it. Second, all the wire demarshaling routines are generated from spec files, and all come down to two hard-to-misuse routines (note: the bolt11 parsing was hand-coded by me, didn't use this, and had serious bugs!). Finally, we use patterns designed to highlight issues early, including running test cases under Valgrind (a memory debugger) and internal memory leak detection.
But as one developer said "It's not coding in C, it's coding in Rusty's C"!
Oh, winning! Pretty good stats on a really cheap die.
Yet last I heard, libsecp256k1 held up pretty well?
On a less vague - posting note, I had Kimi work through CCAN for bugs, resulting in about 100 commits. I'll go through this morning and see how many are C mistakes, but I don't recall many (and yes, I reviewed each one carefully).
But it's mostly my code, and I've got a fairly defensive style built over a long career of mistakes. I should probably post about that...
Here's something I've been working on along similar lines: https://github.com/rustyrussell/bip39dice
Haven't published it yet, since it needs a helper video and some serious analysis (i.e. how biased are normal dice, and how biased is hand rolling) and of course concrete recommendations for the final word fixup.
But being able to *verify* that your phrase is random is far better than other approaches, if you're going to roll...
5 minutes to our chat!
https://centurymetadata.org/chats/
Day 3 of grinding kimi to do a complete CCAN audit (CCAN is the repo of C snippets I maintain, used by many projects including Core Lightning).
So far, some minor issues: mainly for weird corner cases not used by any codebase I'm aware of. But many of them, and this is battle-tested code.
It's impressive how cheap this is: I'm still under $50 spend, and less than 8 hours of my own time.
No: if the grid is the entropy it's just a memorized wallet. The point is that the grid is low enough entropy that you can remember it.
I have not found a correlation between developer attitude and brilliance, regretfully.
The rise of Free & Open Source Software added a bias, though: developers who played well with others made more progress in the long term. That and the internet made software a social activity.
https://centurymetadata.org/chats/
Online in 10 for all your questions!
No, you got a convincing-looking letter in the mail and upgraded your firmware, coins lost. Oops.
I don't think burying Coinkite is the right answer, BTW. But reality rarely listens to my opinions!
But it won't stop someone like this who doesn't use it. Perhaps a secure_rng() API which refused to fall back? But that would have to be plumbed though all the Python libs too. And maybe they would have hacked it out anyway, in the hurry to remove GPL code?
Because it was a fake letter sent to their home address based on the ledger leak and in fact was instructing them to install malware that stole all their coins :(
Today's side quest: Great Speculations (greatspectations.com).
Useful tool for checking source code quotes against a document: I used this in CLN and it improved compliance *and* the spec itself, as well as making it easy to follow spec changes.
This is the polished and published version which handles RFCs, BIPS, and any other markdown or other docs.
What a horrible day.
I am haunted by the fear that a normal Bitcoin use can do everything right, only to lose everything because you didn't hand-roll your entropy, or upgraded your firmware because "Ledger" sent you a warning email.
😢
Euro chat in 15!
https://centurymetadata.org/chats/
Wow! Congratulations, both of you! ❤️
That's great! I hope some of them work out.
I imagine there are numerous companies who decided to keep treasury in Bitcoin and are now seeking both capital, and frankly, reassurance? If so, your timing here is excellent*
*Unless it's not. We will know in three years.
The problem with this is that well-run, boring, positive cash flow companies are really hard to find. All the obvious ones were snarfed up by previous variants of this play.
It *sounds* great, until you get to the messy reality that the firms you want don't need you. So you have to find and pitch them. And why would they say yes? Because they're looking for an exit, which is the opposite of what you said you wanted.
This is not to say it's impossible, but there's a reasonable case that it's going to be far harder than VCs think, who are used to people approaching them.
#nevent1q…77pv
For some reason, the calendar invite for the meeting now says "Invalid meeting".
Try https://meet.proton.me/join/id-NPG552FNQC#pwd-PDKhBp9miOUt for now...
For some reason, the calendar invite for the meeting now says "Invalid meeting".
Try https://meet.proton.me/join/id-NPG552FNQC#pwd-PDKhBp9miOUt for now...
Europe-friendly timezone chat in 10! https://centurymetadata.org/chats/
https://media.someecards.com/someecards/usercards/MjAxMi03YWRiNGEzM2U2NWExNjkz.png
Fifteen minutes to the CM open call (North America-friendly time)!
https://centurymetadata.org/chats/
Had a delightful chat last week with Craig Raw. He is, obviously, a proponent of open protocols. It's something I assumed, but I also assumed nobody else would want to run this service! Still, building independence in from the start is important.
The second key takeaway is that this should be an Electrum protocol extension! That makes life easier for users, (no extra dialog to choose a Century Metadata server) and for wallet devs, who already use the Electrum protocol.
So I read the Electrum protocol spec. It's restricted JSON-RPC (no \n, except at end). Fairly simple, easy to extend. Looks looks a good fit...
Wild huh? Let's be friends.
Apparently the idea that we should be our best online, not our worst, is radically outside the mainstream?
Three hours to the CM open call (Europe-friendly time)!
https://centurymetadata.org/chats/
As a Free Software developer, I've followed expansion of IP law quite closely. "Home taping is killing music", et. al.
The Internet turned a monopoly on duplication into a monopoly on distribution. DMCA created a ban on white-market copying assistance. The trend was all one-way: copying was harder, monopolies ossified and were further normalized.
Then the AI train drove through and ignored copyright. And, so far, it's ok. Nobody is going to shut them down, and they're going to get better at copycatting. So now there's actual competition to the monopolies.
Should be fun!
This was fun! Good chat about finer grained control and incentives to mirror....
#nevent1q…ewts
Century Metadata is a product for Bitcoiners, because they're the only ones who have motivation to maintain their secret key. That makes this whole "public service" data model feasible: you encrypt, so we don't have to keep any secrets. We just have to make your data available.
Precision isn't appreciated any more! Brink's form was harder than I thought...
https://npub1jw5z46xmasx06rephwy60rdk5myn4cvnngc29cejmc7vgk7el0xsw7csux.blossom.band/198f01fa5094cf152559ba4dc916a3d7d3ba1251033581196f6a35978079ccfc.jpg
Not kernel, but he was Debian Project Leader at one time!
Sometimes I forget there are old pictures of me on the internet...
https://lwn.net/Articles/66669/
4 hours before you can hear this live :)
#nevent1q…pw5d
More like "Bitcoin ALREADY HAS cured cancer, you just don't know about it yet!!!!"
At #OsloFF several people asked when Shit Bitcoiners Say would be back, so let me put it on the record: it's hibernating during the bear market.
The account serves me as a balance, to keep things grounded. It's not there to kick people when they're down, it's to serve as a reality check. When Bitcoin is a grind it should be offering free hugs or something.
When people are extrapolating to infinity and you feel the urge to research low-gravity espresso machines for your inevitable
Bitcoin moon-citadel: I'll be back!
Hey @nprofile…6z2t should I be pitching @nprofile…g954 to OpenSats?
I'm back wearing a Pebble watch! I loved the original, and wanted to support the reboot. Let's see how we go after a month...