you guys give a warm welcome to Giacomo Zucco he's going to be moderating our next panel let's
welcome him up to the stage he is going to help kill some time and do whatever you want to do
okay the fact that my wife got the watch is nothing prepared so there is everything is fair
I promise there was no coordination she's just too good so we are going to introduce a panel about
covenants we will go straight into the discussion with James Lopp and Paolo Arduino and Adam Beck that
I will call the stage in a couple of minutes but before I just wanted to give a very brief introduction
who knows what Bitcoin covenants are raise your hands so we understand if we have to start from well
okay who will be very very bored if I explain again what Bitcoin covenants are one two okay sorry for
your loss guys so in Bitcoin you can spend some sets and you send them to what you call a Bitcoin address
right when you spend on chain this Bitcoin address encodes a series of conditions to spend again
trivially the most simple condition is you have to have the private keys corresponding to some public
keys so very simple you have the address like an IBAN address and you have the key like a password this is a
very simple kind of address and then Bitcoin also has a smart contract system who in the audience were
thinking that only shit coins have smart contracts raise your hand good good you're scared good Bitcoin actually
Bitcoin does have a very capable smart contract system it's based on very simple primitives
multi-signature hash pre-image and lock times relative lock times absolute lock times so it's basically
you like in a bank account you can do one signature over two or two signature over trees or you can delay
the payment of the payment of three months three year block some money for some times and that's a contract
a smart contract but what you cannot do is to encumber these coins after the next spending so you encumber
these coins with this contract but once the contract is satisfied the coin the coins are free to go wherever
they want the idea of covenants is to give additional condition after that in which you can say okay if you have this private key you can
spend but you the people you spend it to must also satisfy this additional condition this seems like
something let's imagine this you have a pirate's treasure chest you have the key you open there is
the treasure now it's yours covenants are like this there is a there's a treasure chest you open it and inside
there is another treasure chest with a different key and now you have to satisfy the second condition
this seems useless complication but it's not because it's very important to for example create second
layers like lightning that work better because they don't need synchronous communication right now in
lightning you have to coordinate with the other party to open a channel to close a channel with
this kind of covenant smart contracts we can actually make better uh smart contract the point is that we have to
change the rules in order to do that and we will debate about when how to change these rules so i call
on stage our panelists the first one is jameson lop and he works at casa and he's also
very vocal very vocal about uh bitcoin innovation you are at the center and then i call up paolo arduino
you know him already it was already i think it's up there okay i mean no introduction needed right
i guess he's also on the top list of favorite italians right and then the party animal uh the cypherpunk
legend adam back that is with now with us right now
nice t-shirt nice t-shirt yeah just another clarification before we start i have this t-shirt of the lugano
soccer team i never play i'm the only italian who never played soccer in his life but they they work
with plan b lugano so i am proud to to wear this so guys covenants uh there are at least three different
debates we can have here the first debate is about changing bitcoin rules again at the fundamental level
i know that jameson already had a debate about that in lugano plan b forum jameson is an is advocating for
changing the rules again continuously because innovation is necessary for bitcoin and especially
right now while other people notoriously uh i don't know michael saylor but even even real bitcoiners
like uh like jimmy song was debating uh against uh jameson about that they tend to favor the idea of
eventual ossification of the fundamental rules the second debate is if we want to change the protocol rules
uh is introduced in the introduction of covenants the actual priority right now and we may have some
discussion about that but i assume that for example adam beck was uh promoting the idea of this simplicity
soft fork that could basically encompass covenants and many many many other things in order to be basically
what he calls the last soft fork and then the third question is if we assume that we need to change the
rule and the covenants are a priority which one because there are several incompatible proposals
competing right now for mind share and approval among bitcoin developers and my opinion is that we are
very very far away from from reaching a consensus so let's start i will go first with the first question
to jameson but of course paolo and adam feel free to uh to wait in uh do we really need to change bitcoin
right now taproot i remember i was i was uh celebrating taproot activation in el zonte in 2021 sure it's i mean
it's four years but it doesn't seem like uh uh taproot is finished to be implemented used explored so for some
people were for some people maybe it's too soon to to fight about the next activation where we still have
to harvest the power of the last one uh very few uh we know about taproot assets uh there are other
applications in security but my is casa fully supporting taproot scripts for example so that's the the
angle i want to give to the first question right now i think need is probably the wrong question because
bitcoin doesn't need anything like bitcoin will be fine i think the real question is how does bitcoin
evolve how does the rest of the world evolve if we do or do not add certain functionality and so i think
the the tap script stuff is a good example where casa has not implemented that we would like to uh
get to that point and especially be able to do schnor signatures but we have so many dependencies on a
lot of the hardware manufacturers being able to support stuff that we haven't been able to do that but it's
it's still fine like it's great that some other people are able to take advantage of functionality
even though we aren't and i think that should generally be the case whenever we're talking about
changing bitcoin it's not so much do i personally need or want it but do other people need or want
it and can they use it without harming me although uh even if i promise to skip to the next question i
think you also have a take about changing bitcoin in general i mean the etfs are here uh michael salari
is talking with the etf guy do you think we still can change bitcoin rules in general besides covenant
yeah to me it's not about um what is the best outcome for for bitcoin i believe that i'm um i'm
a simple man i see what is going around i see that there is a lower and lower incentive uh from the
biggest holders of um bitcoin that you still could argue that they are not the decision makers but i think
that there is a lot of pressure for not changing bitcoin coming from people that uh and institutions that
are investing a lot of money in bitcoin so i think any change that should have happened in bitcoin should
have been done before 2024 uh now i feel like maybe it still can happen but very pragmatically i think
that the chances the probability of any change to happen uh to bitcoin are becoming lower and lower just
because there is a lot of uh pressure um to and then i heard that there is a lot of pressure to become
bitcoin core developers not focus on um on structural changes um to to bitcoin and i think honestly i'm
happy either way either it's ossified or not i think that the the the community shall decide i think that
is kind of sad to see that uh eventually uh now bitcoin became so institutionalized that uh that uh people
with a lot of money will do huge campaigns to trying to block any further change uh to bitcoin that could
impair and cause uh potential financial losses uh you know if you are a bitcoin etf and you are custodying
uh you know 800 000 bitcoin why you should take the risk of uh of uh something goes wrong even if there
is a 0.001 percent of something goes wrong going wrong why you should take that risk the same can go to the
uh you know the newly acquired biggest holders of bitcoin so to me it's kind of um um an interesting
trade-off um also the other uh difficult incentive is that sticking your neck out uh in this moment is
is very challenging right so uh you might end up in doing uh for example becoming associated to what
what happened with with bitcoin cash bitcoin satoshi vision and so on so you won't on the wrong side
of a civil war exactly right so it even sticking your neck out to support in a strong way one of the
different covenant solutions might get you under a huge amount of attack and uh you know it's it's part
of life i think bitcoin became very much like you need to be a very good politician now to navigate all these
different situations and so you know it's uh it will be very entertaining to see what will happen in the
future i'm just content either way
good so pragmatic and descriptive more than prescriptive adam uh if we assume we can coordinate to change
bitcoin once again or a few times again or forever uh what are your thoughts about competing ideas not just
within the covenant proposals but in general like you uh can you explain what simplicity is but also
what the great consensus cleanup is if there are other ideas that maybe are circulating from many years but
are now forgotten like cross signature cross input signature aggregation just give us please a a panoramic on
different proposals that and and your and your opinion about the priority among these proposals yeah i mean i
think um it takes a long time to get changes into bitcoin start starts with a problem statement that people
think that they have a use case that's difficult to do or hard to do securely and then different candidate
solutions and of course if there are sort of it's kind of becomes like a language design debate which
is about aesthetics and arguments so there are other languages like the c language had some big debates
about the pre-processor and so it becomes a more fierce debate when it's about some small choice
about uh which which option is used because the option has to be supported forever into the future
um and so from that point of view um simplicity in a way is a simpler debate because it's very very low level
there are nine operators and you can do everything with them so the question is not you know do you
want this covenant or that covenant the question is do you want extensibility or not and people generally
want that right now simplicity is probably a bit early to be talking about that for bitcoin so we're hoping
to get it into liquid we have a release candidate to go live in liquid so we're hoping that you know this
quarter we should have that live in liquid so that's probably in bitcoin time looking at previous
sort of timeline between uh schnor's signatures in elements and schnor's signatures in bitcoin i think
it was eight years so you know you look you're looking a long time but it it's there as a potential
candidate last soft fork because it's low enough level that you can implement other opcode like
functionality without needing a soft fork and without having a design debate about which
it's it's roughly for programmers like the debate between risk processes
and cisk processes so bitcoin script is a bit like cisk so you've got these lumpy
operators that do weird things like floating point or matrix multiply or sha256 you know opcode right
whereas the simplicity is more like the risk approach it's a bit very low level you can do everything
with it you don't need as many operators right um so i think that's in the future but i think it's
important to keep progressing the expressivity of bitcoin scripts um one one sort of important benchmark is can
you transfer security from the base chain to layer twos and side chains and things like that of course
that's a bit of an r d area in itself but it's important i think it's important to have expressivity to
enable those things because what if we achieve that benchmark robustly then we can have faster innovation
and permissionless way in a safe way in the layer two so if you get to that world i think you still want
some expressivity improvements anything focused on cold storage bolts and so i think covenants are in
play for that and for improving the efficiency of lightning thank you adam jameson yeah we can we can clap
jameson simplicity is one option you can do everything with it eventually with some with some
with some caveats and then on the other extreme we have a check template verify very simple very clear
super studied super tested then we have allen hands which is a check template verify plus checks in the tool
from from stack plus internal key in order to improve lightning then we have op vault or or secure the bag
which are uh security specific of codes then we have cat this uh like this uh this hack in order to do
different things in a very inefficient way then we have uh met merkelize all things this proposal to
basically extend merkelization introspection so uh can you give us a little a bit of like assuming we
want to change bitcoin and we can assuming covenants are the first priority uh is there any clear consensus
emerging and what's your opinion about the status of consensus formation about that yeah i mean
there's like eight or nine different proposals out there in various stages um over the past few months
there has been a concerted effort on the bitcoin wiki to try to actually visualize you know what the
rough consensus looks like so there is a covenant support page with a matrix of all of these different
proposals and then anyone who wants to who has an opinion especially if you have an informed opinion
where you're willing to write up your own evaluation of the different proposals you can just go on there
and submit your uh you know approve disapprove uh and so on of what you like and dislike and uh very roughly
because this is roughly because this is rough consensus right this is not a democracy uh it seems like uh ctv
cat and op check sig from stack have the most rough consensus with the fewest uh well-formed
technical objections there are of course plenty of personality objections uh to to some of these proposals but
you know i think the question should be like what do we get from them and so you know paulo mentioned that there's a lot of big
institutional holders whales and probably a lot of people in this room myself included who have the
vast majority of your net worth if you know hundreds of millions or billions of dollars of your money or
other people's money that are in bitcoin now and so we're becoming very risk averse and i think we're at this
inflection point that's very similar to something that's called the innovators dilemma where we have to
decide like are we willing to continue taking risks in order to reap more rewards if we don't do that
what are the consequences so covenants can do so many different things whether it's better security
which i would think that the institutional people would want and i would think that the whales would
want and most importantly i think is scalability especially making layer twos better uh we've been
saying for many years now oh you don't need to change bitcoin because you can just do everything on a layer two
that's only half correct we could greatly boost our ability for people to do permissionless innovation
and make much better layer twos so that people can move their bitcoin back and forth between different
layers are you on the wiki expressing your preference no see that's the interesting thing no one on this
stage right is on that wiki so it's kind of funny that we're up here talking about it and it's also funny
we only have 20 or 30 minutes because this is an incredibly complicated topic yeah yeah right uh so
uh paolo last time we discussed this in a plan before lugano i was a panelist with you and so i i had
an opinion now i'm a moderator so i cannot express it so we'll try to trick you into expressing the same
opinion jameson was talking about personalities uh we remember debate about seguit during the seguit times
there was some technical merit in the seguit 2x proposal like yeah this is good this is too much
but eventually we ended up with basically a fourth a four times increase in the total capacity with the
discount which is even more than the 2x is basically 4x of course with a soft fork no not with the hard
fork so there was a technical debate but much of the debate was about intention politics game theory
precedence for example we say that this proposal is technical but there is a hidden reason to do covert
asic boost this proposal is a way to bribe miners to do something or to give a precedence of more control
to miners so do you think that this kind of argument can also have a weight uh in in the present discussion
including the uh the you know bitcoin phase two rhetorics of uh of some shit companies or do you
think that we should stick only with technical consideration ignoring the politics and the
personalities and the intentions well definitely uh i would prefer to avoid the politics i think that
the beauty of bitcoin to me is that is uh money governed by math and there is no politics in math i mean
there is just uh right or wrong and so i you know i feel like uh the finalist of a beauty pageant where
you know i want uh the you know the the peace in the world and uh you know and and everyone to be happy
but actually i think that um tech arguments should always win i think that the the secure security is
the the most important uh argument followed by actual practicality of the of the implementation
and the fact that it provides the highest um range of of improvements again i i completely agree with
jameson in the sense that uh you know the there is so much opportunity to improve layer twos um in this
moment we have uh different solutions we have liquid uh we have um lightning we have few others
but um definitely the uh the the orchestration and there there is there are some hacks and some
incredible complexity that needs to happen in order to create these layer twos
um so if we could uh have a good solution one of the different uh covenants implemented or
will be the greatest outcome honestly uh eventually with the um you know you you can see that with the
with ethereum now everyone is launching layer twos it makes it so easy to launch layer twos on ethereum
and if that will become also reality on bitcoin i'm pretty sure that people will say oh well now now all
the scans are also on bitcoin but honestly i don't care i just care that bitcoin will become the biggest
and most used security layer in the world it is already but definitely with covenants uh with one
of the comments we will increase the use cases we will make it um so easy to use bitcoin for securing
real world assets real estate you know the trillion dollar economy in in the world but again
is unfortunately it's uh i will i think that it's not about tech it will be although i think that it
should be only about tech will be only about politics i i i'm disillusioned by the fact that at this
point we can easily change bitcoin without one of the big ego personalities going uh being actually in huge
support of this change so i failed tragically to plant my opinion into paulo but that's fair uh i will go back
back to jameson for a moment and then back to adam uh jameson you mentioned security and then we all
discussed improving layer twos uh can you explain to me like i'm five not just years old but also
q uh what is evolved because maybe uh raise your hands if you know what is a bitcoin vault
one oh three true okay this could be useful because the covenant discussion actually started out with
vaults but i the people always failed to convince me that you can do much more than you can already do
with the multi-seg with time lock while i started to be convinced about covenants when i heard about
non-interactive lighting channels and this kind of payment approach so can you give a very short
explanation of what you can do with the covenant from a security point of view so i just trust us without
verifying that covenants can improve lighting i think that's uncontroversial but how can covenants really
improve security can you give a very little explanation yeah so as you explained before we
started uh you know the way that a bitcoin address works is it's basically like a lock box you are
describing the conditions that are required to unlock that bitcoin and be able to move it
you know the covenants uh take a different approach where we're not only saying okay this is what's
required to move the bitcoin but we're looking forward in time and we're saying oh and also these are
additional restrictions on what the bitcoin can do where it can move uh what the uh the future spending
conditions are going to be so when i look at this from a security perspective i actually see vault type
technology as kind of a completely new side of security that we've never had before
basically up until this point the only real security you've had for bitcoin is proactive
security which means you have to set up some really really highly secure set of conditions
and you better hope that they never get compromised because if they get compromised it's game over
nobody can get your bitcoin back now with vaults we would be able to have a sort of reactive security
so you have your initial proactive spending conditions however if you're doing vaults and thinking ahead of
time if your initial locking security conditions get compromised somehow it's not game over you can
actually see if those funds get compromised they're only going to a place that you have already
predetermined where you will have other exit conditions you know escape hatches essentially ways
to get your bitcoin back before the attacker can then move on to an arbitrary address that cannot be
confiscated so casa hodl model will change dramatically with covenant you would use it for for security
actually uh we would love to use it at casa and i would think that like every major
institutional provider would love to have that additional layer of security thanks adam last
question is to you with blocks well sure adam last question is for you uh blocks didn't create a liquid
and liquid uh is not just a layer of uh fast settlement for exchanges or for uh assets now it's a layer to
open lighting channels swap into lightning or out of lighting like with aqua or bull bitcoin wallet
so liquid is very interesting for its functionalities but it's also interesting as a test bed for bitcoin
because it's a sidechain is another thing with another mechanism but it's the same code base of bitcoin
extended to some degree basically the claim of litecoin we are the test bed for bitcoin is actually
true for liquid without even having a shitcoin so liquid is the test bed for bitcoin and as a testnet
for bitcoin in it introduced a lot of things now it will introduce a solid uh simplicity so that is that
this was a bad gaffe but yeah there are some contact points uh simplicity but uh before you already
introduced opcat which is working on top and andrew postra in blockstream was the guy actually
discovering the trick uh with snore and an opcat in order to do covenants in a in a weird way so
the question to you is being that liquid can do covenant since a pretty long while even if uh maybe
not very efficiently do you see any activity in that direction i mean are people trying to use this new
uh tools to build something i know this question may not be entirely fair because the security model of
liquid is different because there is a federation so maybe the incentives are not the same but i think
it's still interesting to see if people are using covenants on liquid or not yeah so the original version of
covenants and liquid has already two methods opcat and check sync from stack so you can combine them you
can use them separately um and after some years of experience of people using that and looking at you
know the actual programs that you write and how many opcodes it takes to do things we added a revision so we added
about 30 helper functions to do with serialization a few years ago and that enables you add sort of
streaming hashes serialization things like that so it's uh improved the programmability um and yeah
people have tried things so some layer twos have implemented themselves on on liquid basically becoming
a layer three to show the test that works you know like one of the arc implementations actually
implemented liquids another interesting point with the different opcodes you can implement model liquids
using cat or csfs and so you can sort of for example the original lightning opcode which is not exactly
a covenant but it's similar they were able to implement that in in some number of instructions using csfs so
there's some kind of turing completeness of covenant opcodes where it's interesting to show but you know this one can
implement that one but that one can't implement this one that tells you something about the power
the flexibility of the of the opcodes so some people have tried that on liquid as well
thank you very much a big round of applause for our panelists our time is over