waxwing @waxwing@x0f.org

Almost certainly the best mathematics education vlogger on the remarkable proof that an arbitrary closed curve always contains 4 points of a single rectangle, using extremely intuitive visual reasoning (and surprisingly easy to understand topology!)

https://www.youtube.com/watch?v=AmgkSdhK4K8&feature=emb_logo

#mathematics

Joinmarket 0.6.1 bugfix release out

(bug was basically deposits not showing, without a rescan, if there are a lot of them because of a failure of the address import mechanism)

https://github.com/JoinMarket-Org/joinmarket-clientserver/releases/tag/v0.6.1

#bitcoin #joinmarket

Reviewing libsodium release notes, got an unexpected gift lol :

https://user-images.githubusercontent.com/124872/58669328-1bd91a00-833c-11e9-8d6b-9e51e9e3be4e.jpg

Scammers are ever inventive; get you to login and claim your mined bitcoins, only you have to top up to meet withdrawal minimum 😂 :

https://www.reddit.com/r/Bitcoin/comments/e6qu33/i_find_this_well_architected_scam_mind_if_i_show/

#bitcoin

Joinmarket showerthought:

long term yield generator can now (as of 0.6.0) deposit to yg and have those coins joined without restart - but what about getting money out? How about they start up with say 30 destination addresses and 1 of 10 joins goes to a destination (could be in your "other", spending wallet). It's actually good for joinmarket's privacy model, too.

This way you could have it running for months on an isolated machine and still move your money through it.

#joinmarket #bitcoin

Comes out of taproot review but this summary of how difficult it is to get threshold multisig working (by Poelstra) just makes me think we're going to have to stick with the combinatorial n of n musigs (in taproot branches) approach, at least until someone figures this all out!

https://github.com/apoelstra/secp256k1-mw/blob/2019-01-threshold/src/modules/thresholdsig/design.md

#bitcoin #taproot #musig #crypto

Oh; Jonas' write up on multi hop locks has moved on quite a bit, looks like a very useful resource now:

https://github.com/ElementsProject/scriptless-scripts/blob/master/md/multi-hop-locks.md

(wish we had TeX in github markdown btw, although i think there are some half-assed workarounds)

#bitcoin

Here's the full article which I admit I haven't read yet:

https://medium.com/digitalassetresearch/a-look-at-innovation-in-bitcoins-technology-stack-7edf877eab14

#bitcoin

Whilst I really do appreciate the effort, this is a little bit buzzword bingo 😂 ... a lot of things are kind of mixed up and not correctly placed (not that it would be easy to do it right!).

https://miro.medium.com/max/12360/1*iAiXJb0MzHMOhJxsElrUUQ.png

A direct quote from Yi Gang, central bank head for the CCP:

"We should not let the money held by the Chinese people become worthless .."

(this is in reference to limiting stimulus measures)

He's been known to talk pretty straight in the past, but it's rare to hear a central bank governor talk like that!

(Quoted translation is from Bill Bishop's Sinocism newsletter)

#china

They solved 795 bit DLP (not ECDLP btw) using a mere 3100 CPU core-years :)

https://lists.gforge.inria.fr/pipermail/cado-nfs-discuss/2019-December/001139.html

#crypto

A good example of where it doesn't really work was mentioned by Bonneau in the podcast: timestamping; no good saying "this VDF proves I came up with this document 1 year ago" if there's a +/- of like 1 month.

In contrast with PoW used for mining, the latter doesn't attempt quite as much, since it's just going for a statistical truth globally, i.e. that the whole world can't compute target values faster than X in aggregate (and then adjust per two weeks or whatever).

VDFs are more ambitious.

Joseph Bonneau on Verifiable Delay Functions; pretty interesting episode imo!

https://www.zeroknowledge.fm/103

While I'm more interested in the concept after having this quality explanation, I'm still a bit unconvinced about the general idea because it's a sort of "linear" usage of a difficulty - 'we expect N iterations to take 10 minutes, give or take' rather than 'we expect this to take a billion years so no one can do it'.
I'm sure there are applications where it can be workable, though.

#crypto