At least once a week, somebody on the street tries to sell me cannabis. I'm sure it's because of my long hair (and maybe my beard and frequent sandal wearing). What's interesting to me is how many unique people have made me this offer. I'm wondering if there's an algorithm I can use to estimate the population of local people with a pot-selling side hustle based on the frequency of my encounters, the number of hours I spend in public, and the local population density.

$1000 worth of Coke is worse for you than $1000 worth of coke.
-stactup, Sep 2014

Nice discussion here on privacy challenges in Bitcoin by @pwuille

Pieter's answer is well worth a read, and thanks to @bitcoinoptech for surfacing this in their newsletter.

bitcoin.stackexchange.com/ques

@giacomozucco @stevenroose @udevnull I dunno, it's all easy if someone else implements the code for you, whether that's shaolinfry with BIP148 or someone flipping LOT=false to LOT=true and recompiling.

@stevenroose @udevnull @giacomozucco I don't understand what you're saying. Do you know that BIP8 LOT=true users require a mandatory signaling period? (The specifics of that were changed a few months ago, separating it into a distinct period.) github.com/bitcoin/bips/blob/m

Bitcoin Optech newsletter #137 is here:

- describes the results of discussion about choosing activation parameters for a taproot - includes selected questions and answers from the Bitcoin StackExchange
- notes changes to popular Bitcoin infrastructure software

bitcoinops.org/en/newsletters/

@jb55 I think @pete had a tool that would spend your dust to a particular output signed with SIGHASH_NONE (or something) so that multiple dust spends by different people could be aggregated together, splitting the cost of the output. For a 100,000 vbyte tx assuming all inputs are P2WPKH, you can split the 31 vbytes of output 1,475 ways for a relative cost of 0.02 vbytes per input---which is pretty close to zero.

Still, I agree with your point in principle.

@waxwing @pete If applied to all transactions, it'd add roughly a block per day of extra capacity to the network for an entropy reduction of 1 bit per sig. That seems like a good tradeoff to me.

You mention the fingerprinting risk, but at the time it was implemented, Bitcoin Core was also one of the only wallets using anti-fee-sniping, which is an even stronger fingerprint. Now there are a few other wallets doing that (including C-Lightning, which also low-R grinds).

@stevenroose @michaelfolkson no insult taken. The computer and API projects were basically dead by the time I left, which was months before the Coinbase acquisition.

21 was the first company to do anything serious with micropayment channels before LN, and I'm proud of some of the work I did on that. But, yeah, the company never had much of a public impact.

@stevenroose @michaelfolkson the "21 Bitcoin Computer" was their first public product. Then they released the 21.co marketplace for bitcoin-payable APIs. Then they released the "21 Earn" paid email service, eventually renaming themselves to Earn.com and being acquired by Coinbase to become Coinbase Earn, a platform for paying people to learn about scamcoins.

Source: I worked there from 3 days before the launch of the computer to a couple months before renaming to Earn.com. :-)

Bitcoin Consensus and Solidarity:

rusty.ozlabs.org/?p=620

I deleted a much more technical post, because it derives from this base idea: that consensus is vital but unverifiable. Our only defense is to have our signalling be open, broadly-based and sustained.

the mars landing video was recorded on the rover by a computer running #linux

stock off the shelf camera and computer hardware on that system, per press conference

youtube.com/watch?v=4czjS9h4Fp

Bitcoin devs Antoine Riard and Gleb Naumenko announce The Label

thelab31.xyz/blog/announcement

Also on birdsite Antoine Riard and Gleb Naumenko announce The Label, a new Bitcoin R&D org. thelab31.xyz/blog/announcement

@raucao yeah, the August 2019 relicense from GPLv3 to "Business Software License" that forbids you from using their software to compete with them.

I didn't know that happened; I've been using ZT since 2015 and never rechecked the license on subsequent downloads.

@raucao right, one engine should be fine per the FAA's ETOPS policy: en.wikipedia.org/wiki/ETOPS

That doesn't make me happy about it, though. If there were more four-engine planes flying to and from HNL, I'd probably prefer to book them, but they're rare.

12,000 meters is only ~33% higher than Mt. Everest where people are on the edge of breathing, so I assume jet fuel and lots of other things can still burn.

Show more

David A. Harding's choices:

Mastodon

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!