what's the best way to secure your devices in your opinion? for instance, if a user is new to a hardware device like Trezor or even a software wallet for that matter like Electrum how can one be abolutely sure that the laptop is not already compromised before doing the installation process?
Have a couple questions, also thanks for doing this AMA Jameson!
With the ever-changing landscape of technology, where do you see the next generation of Bitcoin builders emerging from, and what factors do you believe will motivate them to engage with the space?
Also keen to hear your thoughts on the El Tor project. Not sure if you know but it's described as a high-bandwidth Tor network fork, incentivized by the Bitcoin Lightning Network, where relays can earn satoshis through BOLT 12 and blinded paths.
What potential impacts do you see this project having on privacy and network performance? We have someone in PlebLab working on this, interested to hear your thoughts.
First I've heard of el-tor, will definitely have to investigate.
My initial concerns would that high bandwidth" makes me worry that it may be more centralized or susceptible to timing attacks.
Given that it's such a new project, will want to see a lot of eyes on it before anyone considers it secure.
At a high level I do like the idea of incentivizing people financially to help improve privacy. Hopefully the characteristics of the network aren't limited by the level of Lightning adoption.
In regards to timing attacks, it's a pretty well studied topic and a lot of good research is documented here - Thirteen years of Tor Attacks - https://github.com/Attacks-on-Tor/Attacks-on-Tor Much of the timing attacks can be mitigated by incentivizing more independent individuals to run relays. The "High bandwidth" tagline is just simply the idea that more incentivised relay operators can lead to more overall bandwidth for the network. Currently the total aggregate bandwidth of the traditional Tor network is about 800 Gbit/s. My home network has gigabit fiber, so this number can probably be greatly increased and decentralized with a bitcoin incentive.
The Classic Relay Conundrum:
To go into more detail about timing attacks, one of the sneakiest attacks happens when the same person or entity controls both the entry and exit nodes of your Tor circuit. Imagine this: you're sending your data through a tunnel, but both ends of the tunnel are under the same roof. This setup allows for statistical correlation and timing attacks, where attackers can match traffic entering and exiting to de-anonymize users.
Current Mitigation Tactics:
Tor's defense? "Entry Guards." You pick a trusted entry node and stick with it for days, even months, betting on the hope that your entry and exit aren't controlled by the same malicious entity. But here's the catch - with only about 8,000 relays, many managed by NGOs/charities/maybe bad actors, the diversity isn't as robust as we'd like.
A Solution? More Individual Relays, Bigger Anonymity Set:
To beef up security, we could incentivize a broader range of individuals to operate relays. Imagine if running a relay got you paid. Suddenly, you've got a larger, more diverse set of nodes, making attacks harder to pull off. Plus, with more users jumping on Tor due to faster speeds, the anonymity set grows even larger. Adding techniques like data padding and latency padding, as suggested in various research papers, could further fortify Tor against these threats.
Lightning Network: Privacy in the Age of Speed
Switching gears to the Lightning Network, privacy is getting a boost with innovations like BOLT 12 and BOLT 11's blinded paths. These features blind the receiver's identity/path, making it tougher to track transactions. The Lightning Network's adoption is on an upward trajectory, but it's the unannounced channels and blinded paths that keep the growth under wraps, making monitoring and usage stats a challenge (which is good).
User-Friendly Innovations:
Tools like Alby Hub and Phoenixd are making Lightning more accessible. As these platforms evolve, they could significantly expand the user base, enhancing both privacy and usability. Plus users already running Umbrel and Start9 make them a great candidate for earning sats hosting a relay.
Tools like Alby Hub and Phoenixd are making Lightning more accessible. As these platforms evolve, they could significantly expand the user base, enhancing both privacy and usability.
Keep in mind that phoenixd is not a solution if privacy of the payments are important, see #866536
All assets held with Casa vaults are eligible for Casa Inheritance. Casa members can store bitcoin (BTC), ether (ETH), USD Coin (USDC) and Tether (USDT) in their vaults. As we continue to monitor broader innovation in the crypto space, we will prioritize assets and protocols that meet our security standards.
Pretty important - I think this is a middle ground with regard to trust.
For example, you can fully trust a third party to hold your keys, but then we're just recreating tradfi.
Or you can go fully trustless and do everything yourself.
But with Bitcoin's programmability there's room for a middle ground. You can create hybrid distributed setups where you're not fully trusting any given third party, but are rather only semi trusting one or more third parties to help you secure your assets.
I was about ready to give up on friends and family orange pilling until I found Sats VS Fiat. Joe Bryan put together a really good video here: https://www.satsvsfiat.com/
I showed it to my mom and after watching it I think she got it. She said "On the Brady Bunch they had 6 kids and a maid on 1 salary. Can't do that anymore with Fiatellos""
There's more to it, from a privacy and opsec point of view, the more you talk about something it becomes more likely that will be heard or shared with others outside the group and potentially lead to physical vulnerability.
What's your best take on the problem of inheritance these days? (And also saving in BTC for kids etc).
Trade-off between security and privacy, building up of too many UTXOs, reliance on legal system to transfer ownership (i.e., hold a key with a lawyer).
With inheritance you're trying to balance very different sets of security. On one hand you probably want to be the only one controlling funds while you're alive, but if you die you want a switch to magically flip so that some other set of people can access the funds.
This is where multisig really shines in my opinion. You can distribute keys in such a way that executors don't have control of your funds, but can collaborate to recover funds in the event of your passing. We developed a pretty nifty user friendly inheritance solution at Casa that only takes a few minutes to onboard executors. The neat thing is that it doesn't rely upon the legal system. https://casa.io/inheritance
With UTXOs the main thing I caution against is withdrawing to self custody too frequently. Probably shouldn't do that more than monthly or you'll regret it many years down the road.
I'm a big fan of a suppressed Kriss Vector chambered in .45 with a 25 round magazine. Also think the Kel-Tec KSG is pretty reliable. I wrote about my thoughts here https://blog.lopp.net/firearms-for-home-defense/
One of our greatest mistakes / failed experiments was the Casa Node for bitcoin & lightning. It was a great concept and possibly before its time. Unfortunately the market was too small in 2019 / 2020 and the support costs for dealing with all the complexities of the hardware / software / networking were far too high.
Probably that we will need to eventually revisit the block size debate and that failing to allow the base blockchain to scale along with technological growth will have its own undesirable ramifications.
Though I could be wrong... if everyone just decides to "adopt" bitcoin via ETFs and nobody actually wants to use bitcoin in a permissionless manner!
I saw your Lugano presentation and agree absolutely that it makes sense to scale with technological scaling laws
Do you think this is a debate that will have to be revisited periodically, or do you think bitcoin devs will be able to agree on some algorithmic scaling solution?
I think block size / weight shouldn't be "pick a number" political games every so often, but rather should be thought of similar to the difficulty target adjustment, but with economic factors. I'm hopeful that an algorithm can be developed that takes everyone's concerns and incentives into account.
Of all the dynamic block sizes algorithms I've looked at, I like Monero's the best. I think some of its ideas could be worth incorporating. In particular, the economic aspect that larger blocks also requires higher transaction fees - any block size proposal should take long term thermodynamic sustainability into account.
How out of date a lot of people's software / operating systems tends to be. A lot of issues we have helping people getting their hardware signing devices working is due to them running very old unsupported desktop or mobile software.
Generally not a fan of any such theories because I dislike speculation, but it seems pretty clear to me that there was a lot of funny business around JFK... looking forward to seeing the files released soon.
I think pretty much all of the alien related conspiracies are just BS from folks who prefer that we not be alone in the universe. Though to be clear, I doubt we're actually alone in the universe, I just think we're too distant from other civilizations and thus we're practically alone.
Or the pussies like you who target others for jealousy.
If this is related to my project or plans ahead, you'll be proven wrong you know. So, you're not coming out with your nym for embarrassment.
If you think this will break me and I'm gonna leave SN, you're wrong here as well.
We're very opinionated about what features and processes we build into our software since our primary motivation is to eliminate points of failure and prevent pitfalls / footguns.
On the service side, we're expanding beyond merely providing consultation services for managing your keys, but also now for our Private Clients we're helping harden other aspects of their lives such as privacy, cybersecurity of computing devices, operational security, etc. Because all of those things can ultimate affect the security of your assets.
The most common one is the long-term issue of not having a well vetted inheritance plan. Those who do have "a plan" usually have some convoluted "treasure map" type of system that hasn't actually be tested and usually has potential points of failure.
The biggest problem that folks run into when using Casa is not heeding our advice, ignoring our reminders, etc. For example, we'll remind you to perform a health check if you haven't used a given key in 6 months. This is just to ensure that your key is still working as expected and hasn't been lost or compromised. But we can't force anyone to maintain their key health, and so if a client ignores our messages for many years they may find themselves in for an unpleasant surprise when they finally go try to use a given key.
the world could really use a comprehensive user-friendly security plan creation solution imo - determining & mitigating attack vectors in a "normie-friendly" way
while everyone's situation is a bit unique, it seems like like are commonalities or principles that could be abstracted up and offloaded from manual human research and effort. perhaps like a private local llm that figures out what attack vectors your likely exposed to through a chat interface and then recommending tangible links and steps to address
My biggest concern about using a company for custody is that they might disappear, forcing me to find a new provider or handle it myself anyway. What would you say to someone with this fear?
Use Casa only as a backup solution but still have own plans how to recover even without Casa?
It's very important to us that we put our clients into an architecture that eliminates single points of failure, and that includes Casa. The first thing that happens when you complete onboarding with us is that you receive your Sovereign Recovery instructions which explains how to recover your funds without using any Casa software or infrastructure. We encourage clients to do a test run for themselves!
Yeah it's a pretty simple trade-off: with ecash you get really strong privacy at the expense of rug pull risk. Fediment seeks to reduce that rug pull risk by distributing it, but it will always come down to the reputation of the federation members.
With a password manager? Nothing fancy. I have a ton of different phone numbers, email addresses, physical mailboxes, etc. Each identity has its own encrypted note with all of the relevant contact details.
No KYC at Casa, so that's a big distinction between us and Unchained & Swan.
Biggest differences between Casa and Nunchuk & Keeper is that we A) support other assets and B) have higher levels of consultation that expand beyond just the key management / asset security.
bigger market share, more revenue ... there is nothing else there :) My guess is that if there was enough clients for bitcoin, they would not support eth
I refuse to do business with a company that supports alt-coins, especially ones as bad as ETH. For me, it's an ethical issue, but it also says a lot about a company's values. Supporting ETH is an instant no-go for me.
I don't bring this up to be unnecessarily confrontational. I seriously want to hear what the rationale is behind such a move.
P.S. I absolutely loved your take last week on WBD about Bitcoin ossification! Really one of the best podcast episodes I've heard in a while.
Because people have security needs beyond Bitcoin.
Over the long run we expect cryptographic keys to become important in a variety of aspects of our lives, not just finance / assets.
We recognize that some people disagree with this perspective and are willing to cede that market segment to Bitcoin-only competitors.
I suspect a lot of folks aren't aware of my BitGo history where I went through the same process of transitioning the company from Bitcoin only to multi-asset. I wrote about some of my issues doing so at length, and I still stand by my position that Bitcoin is better designed. https://blog.lopp.net/the-challenges-of-building-ethereum-infrastructure/
But at the end of the day we're running a business, not a church.
20 years ago I let a friend hold one of my handguns and I had removed the magazine but failed to clear the chambered round. He immediately aimed it at a wall and pulled the trigger, resulting in a negligent discharge. I was deathly afraid for a few minutes that he might have shot someone on the other side of the wall. Thankfully no one was hurt!
I'm sure it was a scary moment. Now, its a good story to share.
When I was about 8 years old, my 10 years old brother and two cousins found my grandfather's revolver. Imagine, 8 arms trying to grab the gun...then, I very loud sound when off....no one was hurt, the bullet when through the wood furniture and the door.
Next day, with my dad we started shooting training.
I think it will mostly come down to the team and level of service. Any software we write can be copied; for example we were the first to build key rotation functionality and created the concept of health checks. Several others have since followed suit.
I saw your website. Basically, you guys secure our Bitcoins and other assets, but I don't know if I would purchase your plans because I can secure them on my own, like seed phrase. What do you do that's special? Is it a totally unhackable vault?
Lifestyle changes - basically, getting comfortable with lying to people and service providers about your true identity. It's almost always legal to lie about your identity if you're not using the deception to cover up committing a crime or entering into a legal contract.
So the origin of Casa was actually from its original name, Bedkin. Before I joined the team and we pivoted it to "Casa" the company's goal was to build a decentralized AirBNB on the original Blockstack protocol. It seemed fitting to go with Casa for the vault product because we're trying to help people build a comfortable home for their bitcoin.
That's tough. I've had the privilege of working with so many incredibly smart people.
I also run the Mensa Bitcoin Interest group which technically has a lot of smart people, though in my experience high IQ tends to come with trade-offs like low EQ.
Pieter Wuille and Olaoluwa Osuntokun are near the top of the list.
Personally I'm a big fan of reverse searing cowboy steaks so that I can get them the perfect temperature all the way through. Extra points for dry brining a few days beforehand.
I got similar answer second time in a row #833958
It's a bitter truth, if your order medium you'll get well done, that's why I have to cook myself or order rare
"meta question":
Have you considered setting up your lightning address here on SN, so you can receive sats as sats and not as Cowboy Credits?
For what it's worth there are folks that can exchange CCs for sats if you want, or if you actively participate on SN for a while you will actually get real withdrawable sats via rewards.
Because people have security needs beyond Bitcoin.
Over the long run we expect cryptographic keys to become important in a variety of aspects of our lives, not just finance / assets.
Casa is designed so that we're not a single point of failure. The first thing that happens after you initialize a vault is you receive your instructions for how to recover funds without use of Casa software or infrastructure. https://blog.casa.io/how-casas-sovereign-recovery-works/
phoenixdis not a solution if privacy of the payments are important, see #866536