Hi, I’m Murch. I’m a computer scientist currently hailing from NYC, originally from Germany. I have been contributing to Bitcoin Stack Exchange for about a decade, wrote my master thesis about Coin Selection in 2016, and worked for over three years at BitGo on the Bitcoin backend. I now work as a research engineer at Chaincode Labs contributing to research, Bitcoin Core, Optech, and the Chaincode Podcast. I’m currently interested in mempool improvements, block building, and wallet development. I have been involved in organizing Bitcoin meetups for many years, recently co-hosting NY BitDevs. You may have seen me on Twitter as @murchandamus.
pull down to refresh
Hey Murch. Any concern over a number of core devs/maintainers leaving, or announcing intent to leave over the past year or so?
A little, but not debilitating concern. Most of these people are still around, just not active. On the other side, there are also new contributors coming in, making great contributions, some of whom hopefully grow into long-term contributors. We do have to keep working at on-boarding people, though!
Is reviewing code and writing tests still the best way to get into core development? I read that the last time I looked this up
Generally, yes.
https://jonatack.github.io/articles
Thank you
I’m a big fan of LN-Symmetry (aka Eltoo) and the derived variants. I’m looking forward to it getting much easier to have channels with more than two participants and the removal of the toxic outdated channel states. I guess that’s already on the way and just a matter of time, though. The biggest hurdle is presumably the consensus change to add
SIGHASH_ANYPREVOUT.I don’t think it’s possible to force a channel to remain open, in the sense that it must be operational until a specific date. Ultimately, the channel co-owner can simply not cooperate to create new updates, so even if they cannot close the channel, they can force inactivity.
So how can be improved the protocol to have less force closures. Yes, I know that most the fault is from each user, in which conditions they run their nodes. But something maybe could be improved in the protocol, so we can have less force closed channels.
Soon will be a fortune to pay for opening/closing channels.
I guess some force closures are due to subtle differences in how implementations handle disagreements on channel parameters, while others are due to node operators’ expectations not being aligned. I would expect both to naturally recede over time, as node operators homogenize expectations or learn to better check in advance whether they are a good fit for a shared channel, and implementations improve interoperability. I don’t perceive the protocol itself as a major source of force closures.
I guess also, there were a lot of people that were just experimenting and dabbling with Lightning Nodes so that there was a large population of nodes that were just not well maintained. As implementations mature, less effort will need to be put into maintaining a node’s health, but also, the cost of operating a node in conjunction with the risk of having funds in a hot wallet will probably serve to dissuade ill-maintained nodes over time.
Yes, I have the same "feeling" and from what I see/listen from many noobs talking with them, this is the main issue for these FC. Thanks for your insight.
As we discussed in yesterday's Optech Recap, package relay, TRUC transactions and Pay-to-Anchor well hopefully what lightning channels with zero fee in the commitment transaction in the next year or two. This would remove a huge source of channel closures.
Yes, I was listening that episode! Really good stuff.
Hey Murch! What's the funniest person you ever worked with and how big part of it is his Polish accent and bald head?
Haha, good times. Hope I’ll see you at a Bitcoin conference this year?
When do you anticipate Bolt12 will be implemented widely enough to begin replacing LNURL in applications where it will be the better option?
(Not sure how relevant that question is to your exact background)
I guess the big question is when support for Offers is implemented in LND. As far as I can tell, all the other implementations have already started work on Offers, or even support it, but the majority of the network still consists of LND nodes, and from what I understand, the maintainers of LND have so far signaled that they prioritize other projects over BOLT12 support, even to the extent that external developers offering to work on it got dissuaded.
So, I guess it’ll be until either the users that are really excited to use BOLT12 migrate to other LN implementations to the point that the node population significantly changes, or LND gets support.
It is being pointed out to me that LND meanwhile aims to add support for blinded paths in Spring 2023: https://github.com/lightningnetwork/lnd/pull/7195#issuecomment-1352444897
deleted by author
What are some of your biggest tips / tricks for bitcoin meetup organizers? I have been hosting a monthly one here in Canada since last June, I can use all the help I can get :)
I find that the location is an important aspect for social meetups. It should be comfy and facilitate a conversation with multiple people. Additionally, you just need a seed of people that can’t stop talking about Bitcoin of which at least a couple actually follow recent development.
For the technical meetups, it helps to have a good mix of novice friendly topics, bike-shedding topics that everyone can weigh in on, and a few topics to pique the interest of experts. Also, have a guest or two to build some momentum! We have a few suggestions on running a Great Socratic Seminar on the BitDevs blog. You may find it interesting.
appreciate it Murch!!
can we expect to see Lightning payments integrated into Bitcoin Stack Exchange soon?
would be a nice way to reward people with great answers to questions.
Unfortunately, I am not aware of any efforts to facilitate that so far!
One problem the Bitcoin Core Team spent energy solving for is:
A bunch of volunteers and Bitcoin Core contributors submit their own attestations to the Bitcoin Core repository. These attestations are tied to a specific commit and thus you can check that at least these attestations agree on the version they commit to. If you want to be thorough, you can set up your own Guix build and check that you arrive at the same binary yourself. This commit should correspond to the release tag signed by one of he maintainers. The PGP keys of the attesters are also held in the repository. You should import their keys and check that other keys you have encountered before certify the attester’s keys.
So, you can either trust the “social proof” of a ton of people staring at the Bitcoin Core repository, hoping that someone would raise alarm bells if attestations disagree or not match the release commit, or you could use a web-of-trust per the PGP keys to assign a higher trust to some attestations.
You may also find the Release Process documentation interesting: https://github.com/bitcoin/bitcoin/blob/master/doc/release-process.md#building
Thanks for the link towards the PGP keys of the attesters. The link included in the Linux verification instructions on the Bitcoin Core website is erroneous.
If you would run a public routing LN node
I’m afraid that I am not knowledgeable enough on these aspects of Lightning Node operation to have a recommendation.
Lead, top rope or bouldering?
I’ve been almost exclusively bouldering in the past five years. I guess it might have been three times that I went top rope climbing in the same time frame.
Do you have any contrarian bitcoin takes? What's something you believe about bitcoin that most bitcoiners would disagree with you on?
I’ve found myself disagreeing with a lot of bitcoiners in the last three years. I find that there are many different reasons why people might like honest money which seems to surprise many novice bitcoiners that expect all bitcoiners to share their own motivation.
What are your thoughts on Ordinals and Inscriptions?
I’m not particularly concerned one way or the other about the additional blockspace demand and blockchain growth. Less blockchain growth would be better, more blockspace demand is also better…
I don’t find Inscriptions interesting, I neither understand why anyone would value them, nor consider it a sensible use of blockspace, but on the other hand, people can do what they pay for. I dislike Ordinals for reinforcing the notion that sats are non-fungible and traceable. I kinda like how it has piqued the interest of some people who had previously not looked into Bitcoin. Perhaps these new perspectives can inspire some interesting new applications and ideas, or at least challenge us to reevaluate our values.
What do you think of Spacechains?
I am vaguely familiar with how they work, and it seems like a reasonable construction, but I don’t have a strong opinion on Spacechains. Sidechains have not been much of a priority for me in the past few years. I’m waiting for a convincing motivation to bring one into existence.
What motivates you to work so thoroughly and consistently on Bitcoin Stack Exchange?
“Gardening” Bitcoin Stack Exchange feels like a great way to nurture one of the greatest resources in Bitcoin. I still learn new things every week.
I feel like Germany produces more Bitcoiners per capita than nearly anywhere. Any theories as to why?
Given the history in the previous century with two different flavors of authoritarian goverments, Germans tend to be more interested in privacy than in other countries. Cash is still a lot more prevalent in Germany than in many other Western countries. I guess Bitcoin just fits well with the predisposition of many Germans.
I can agree. Germany is also the only country in the EU where Google Street View is not prevalent due to privacy concerns:
https://i.postimg.cc/cCX0d4C4/2023-03-09-170356-1920x1080-scrot.png
Hi Murch! Thanks for all your work over on stack exchange over the years. I've always found the site to be a great resource; I've learned a lot there over the years.
Anyways, a non-bitcoin question: I think I remember you saying that you enjoy board games, do you have a favourite these days? And also, an all-time favourite?
In the past months Dominion has been my favorite, my all-time favorite might be Ricochet Robots.
Oh, Dominion is a lot of fun! I bought the base game probably a decade ago, and its been a long time favourite. These days I'm mostly playing online though.
You should check out the online Dominion league if you haven't yet. Its a good time, and a good way to test your skills.
I'll have to check out Ricochet Robots, haven't actually heard of that one before. Appreciate the recommendation.
What's the most surprising things you've learned working on Bitcoin for so long?
The power of a single individual or small group of people to will something into existence, e.g. BTCPayServer, libsecp, Nostr…
I guess in the end that’s the only way anything ever happens, but I don’t think I was as aware of that before I discovered Bitcoin.
We’re at the top of the hour. Thanks for swinging by, asking questions, and tipping. I’ll check in later to respond to stragglers. :)
Where do you see the trends of Bitcoin and renewable energy going?
I would forward this question to the people that seem the most insightful to me in this regard: Daniel Batten, Margot Paez, and Nathaniel Harmon.
Do you use any LN wallet? Which one is the best to you?
Sorry, I will leave this one to people that specialize in LN and have grazed more broadly than myself.
Does this mean that you don't use LN absolutely?
No, I do. I just mean that I have not broadly tested lightning apps, so I wouldn't say that I'm the right person to make recommendations.
What did you have for breakfast?
I made pancakes this morning
If you had no computer programming/coding experience whatsoever, and wanted to encourage your kids (no experience either) to become Bitcoin core developers/contributors some day, what is the very first thing you would recommend for them to learn, and what is the best place for them to begin learning it? Thank you Murch and hope you’re doing well,
That seems very dependent on the individuals involved. I guess I might try to expose them to opportunities to learn about programming and how Bitcoin works under the hood.
So, how safe would you say El Salvador's coins are right now? XD
I don’t know how El Salvador is managing their coins, so I have no idea
Where would be the best place to start bitcoin development journey if you not a coder?
I wrote a bit about that here: #141322
Do you feel like Bitcoin core is accessible enough for new developers looking to start contributing? Either way, do you have any ideas to increase developer accessibility? Thanks!!
It could definitely be better. My colleague Jonas has been doing a lot of work in this area. I think the Bitcoin Protocol Developer seminar is a big step, and there are a few other good initiatives trying to build paths to contributing to Bitcoin Core. It’s a work in progress.
Hey, based on your deep technical knowledge of Bitcoin, in your opinion:
A true master Bitcoiner Welcome to the Forum Murch
Thanks