A potential use of BTC is to incentive home lab utilization by third parties, but we have some obstacles: I don’t like your code running in my network; you don’t like your data going to my environment. AI can help on both sides: isolate external access carefully and care for data
A long while back there was a war in tech data storage philosophies: NOSQL vs SQL. This even spilled into Bitcoin via its adoption of LevelDB. The tide turned against the NO camp because of the difficulty of reimplementing Db functions, but with AI, maybe that's more reasonable?
It feels like the calm before the storm to me on BTC. The price has been relatively stable for a long while, the shared UTXO capabilities are just starting to get adoption, and AI is helping dev but being used for hardening first. Eliminate all bottlenecks, let 1000 flowers bloom
One reason I don’t like attacks between BTC projects is that it gives a false impression of quality of those that aren’t attacked. In fact money, influence, and tribal affiliations are way more at play in the discourse. Stay attuned to first principles; make your own assessments.
A bad take for me is that because AI will reign supreme it will suck all the capital out of BTC. I mean what’s AI investment doing? Building commodity electricity and data process where returns go to zero. But what is AI delivering? Very cheap digital everything, where BTC lives?
Bitcoin is kind of a paradox because you can get exactly what you want and you can live unto yourself, but if you do that to an extreme you will find that the utility of the system is limited. You get a free hand, but you'll find that you may want to bind yourself to compromise.
An annoying reshaping of the web is happening wherein all content sites will block open access and require login. But coding alternatives to these sites should hopefully get easier. This will help open models get data. Potentially BTC can be used to square the incentive circles.
The chain's like a court: you use it if there's a counterparty dispute. In the context of shared UTXOs, you can reason about the value you store on those with this court in mind and this actually incentivizes decentralization. Even with PoW, reasoning considers an honest majority
If you think AI BTC loss attacks are bad, think about the fact that those are very clearly incentivized, whereas someone hostile to Bitcoin systems at large doesn’t need to steal anything; all they need to do is denial of service. There are way more soft targets in that direction
I know shared UTXO systems as built today are needing to be complicated to deal with mass scale, but I still like a small scale shared UTXO that is simply a MuSig2. Let’s say you have a group of 10 and you transfer on-chain every week or so. You just have to be online and cosign.
Something I think we should reconsider in light of recent events is adding an ability for LN nodes to be able to do key rotation and even start a node from cold or a break but restore and continue peer connections. An easier life for operators means more network decentralization.
I’d like to see an alternative to Cloudflare that uses BTC UTXO sharing and LN to deal with DOS attacks. Instead of making people do the captcha, if inbound traffic becomes too high sell ephemeral user ids based on small payments. Each id then gets its own individual rate limit.
Now we are all alone, facing ghosts in the machine going after our BTC. All the code we run faces a new fitness check; a new napalm era and nothing is safe. But I'm hoping it will be a cleansing fire and we can put our insecure dark patterns in the past, and it's all up from here
If you want to reconsider your LND setup, like rotate your keys or switch to native SQL Postgres from the start, Lightning Loop can help make that seamless. Just setup a new node, open channels to your old peers, then send across to the old node and Loop Out to restore liquidity.
Maybe wallets are the wrong way to get shared UTXO adoption, instead we could build the Bitcoin into applications themselves. So a recipe publishing app would offer an “earn BTC” toggle to paywall recipes, behind the scenes it joins a shared UTXO and you don’t even realize that.
Security issues are tough for OSS BTC. I have no idea who actively uses what software. And for the users, I don’t know what version they use. And I don’t really know how they do use the software. So if there is a problem sometimes I know how maybe but not who when where and why.
People are just doing the very basic level of attacks on BTC now using AI, much more subtle and sophisticated ones will come as AI gets better and people learn how to use it better. Like imagine features are added by a new contributor but there's actually subtle problems with it.
When LN first launched I felt pretty nervous running my nodes. I recommend not allowing 3rd party connections or channels but I didn’t do that myself. At any moment only some code is protecting your funds. Over the years that trepidation has reduced, but recently it’s on the rise
Right now there’s a lot of effort being placed on finding and fixing bugs in BTC software. I think this could create a feedback loop from people losing funds to bad guys getting more interested. more fixes which reveal issues themselves. So now more than ever please stay updated.
A beautiful thing about Bitcoin is that you decide for yourself what it is. The forkers demonstrate this: 2X, bcash, SV, etc. Maybe their world is small but it’s still their world. And if they want to come back there is no one who can stop them. It’s free choice all the way down.