["I'm Gonna Make You Feel It!" plays in the background, then fades out to the end of the video.]
Good morning, good evening, and good afternoon to all of you who have joined us here today
for another AMA with Artier Reitman. Good evening, Artier, how are you doing tonight?
Hey, good evening, I'm pretty good.
Great, so just to introduce myself, I'm Prenoja, I'm the Community Associate for Arts
and today I'll be giving Artier the questions that you all have given.
Thank you for giving your questions on Reddit, on Twitter and in the Discord as well.
There were some great questions and I hope we'll get to all of them.
If you have some questions during the space and we have some time at the end,
I might answer some of those questions and I'll keep my eye out for those.
Predictor Dao had three questions.
The first question, are there any discussions with third-party bridges like Wormhole or Layer Zero
to integrate Tezos or Etherlink with their existing cross-chain bridges?
Yes, absolutely, I don't think Etherlink would have much legs as a product without a very high-quality bridge
at its inception between Etherlink and the Ethereum ecosystem as a whole.
So there have been talks, I can't say right now, for now, who will be talking to us,
but yes, that's been from day one, one of the important considerations for this ecosystem to have legs.
Great. So their second question is,
what are your views on account abstraction at the protocol level in other protocols like ZK Sync, Luxo,
and can similar AA changes be done to the Tezos protocol as well?
I have to see. So account abstraction is important and is useful.
I was interested in it for a while.
I don't remember being an early advocate, at least for Tezos, of the Argent wallet,
which I think did a lot of great work.
I made some proposals around social recovery.
One of the first wallets for Tezos was also the Cortis wallet,
which was a mobile wallet that featured account abstraction and security signatures
that was in 2018 or early 2019, so really ahead of the time here.
In terms of what's been done for account abstraction on Tezos,
obviously you can create a smart contract wallet and we've had wallets that do this in the past.
In addition, there's a Tzip that's been worked on for proof of event.
So it's not a fun way of talking about attending a live event.
One of the difficulties around account attraction is that you can't directly sign a message
and many dApps today are requiring users when they identify with a dApp to sign a message,
like, yes, I control this account.
How do you generalize that to an abstract account that's a smart contract,
which could be a doubt, which isn't an app that's necessarily like one person is a knower,
and the limitation that we found was the most generic is a proof of event.
So essentially you have a standardized event that the contract can emit
and when the contract emits that event, it's similar to a signature.
That Tzip was, I would say, an important piece of work
in order to support account abstraction on Tezos.
That said, I feel like I'm not getting something,
like everyone's super excited about account abstraction, account attraction,
this account attraction, that is going to fix everything.
I don't see it, so I may be missing something.
I need to look into why people are making those plans.
I think you should have it, and it's better than using just merely private keys.
It's definitely a better experience,
but I'm not seeing this as the NBO that it's being hyped at.
So either I'm missing something or it's just another overhyping.
And then the other thing is that, which I find super interesting,
there's two types of account abstraction you can think of.
One is you use a very generic smart contract,
and you get full account abstraction via a smart contract.
But another thing that I've seen in the design space,
which I find quite interesting, is account abstraction via something
that looks a little more like a Bitcoin script,
where essentially you have some programmability.
Well, you could even have some readable states,
but you come in and you provide it with this.
And so I could imagine something like a smart contract,
but with a beautiful state, with various entry points,
but that you call with a witness,
and you call it with a witness,
and you have a bound in the gas,
and it cannot be called by something else.
So it would be a smart contract,
it cannot be called by another smart contract,
only a user, like an externally, it's kind of an externally old account,
but essentially you generalize the signature
to any type of computation that could happen within the gas bound.
That, to me, is a really interesting possibility
because it's not as flexible as full account abstraction
but the little bit of flexibility that you lose
gives you a lot of benefit in terms of scalability.
So I wonder if some type of account like this couldn't be implemented.
I think, I don't think this is like a tremendous limit right now,
that I want, but when we're going back to the drawing board
in some sense for what is the structure for accessing something
like this too, that should be definitely a point of reflection.
And maybe if a predictor DAO can comment
why they are so excited about it
because they seem pretty excited about its future.
They have one more question.
Is adaptive issuance necessary
and how does adaptive inflation help in improving the user experience
Well, necessary is a very binary word.
So I don't think anything is necessary.
I don't even think it's necessary to have security on a disaster chain.
You could have it in a secure chain
and then it would be a lot less useful.
I would say the better question is necessary.
It achieves a lot of things.
I would say the main case for it is that it's more compatible
with the real world than delegation.
It's more compatible because it doesn't put its most strength
on the bakers in terms of the regulation you need to abide with
in relationship with your delegates.
I don't think it's been a pragmatic issue so far,
but I think it's something that would happen on the horizon.
And so it's basically like skating where the pockets going to be.
We've already seen difficulties that exchangers have had
in offering sticking programs.
And so in a scenario where the rate of inflation is fixed
and then you make it batch by delegating,
which is what happens in Tezos,
essentially it's going to punish pretty harshly people
who use exchangers who can't for a variety of reasons offer delegation.
And we've seen already Coinbase in New York and California.
They're still baking, but the rewards are just being sold
and not going back to the users.
It's also from a physical perspective,
taxation is not symmetric because taxation is not symmetric.
From an economic perspective, the idea of like, well, you have inflation,
but you make it back in theory works.
And in fact, on Tezos right now,
a delegation pays about 1.6% a year more than the rate of inflation.
So if you look at it as like 1.6% effective deflation,
that's actually much better than the so-called ultrasound money on Ethereum.
However, once you account for taxes, it doesn't look as good.
So both in terms of like infrastructure lending,
a lot of people in lending don't understand that you need to add
the delegation rate on top of the interest rates
because it creates friction in lending markets,
creates friction for exchanges,
and it creates friction in terms of like tax liability,
creates potentially friction irrigation for bakers.
So all of these things I think are a good reason
to not have this mechanism where you have high inflation
but then everyone gets paid back.
And instead of that, just basically cut the part that's supposed to be offset
and just go directly to whatever is the excess that's needed
to pay the bakers to do the work.
So from that perspective, it's much more compatible with the real world,
But specifically, how does it improve the user experience
I mean, it depends what type of user you are.
If you are a user and you're keeping a small balance
and with your small balance, you are playing some game,
shopping for something in FT.
It's not going to make a huge difference in terms of your experience,
I would say, one way or another
because essentially you can ignore the effect of sticking
or delegation on your balance.
However, if you're using this more for payments
or as a store of value or for other applications,
then it makes a big difference in the user experience
because the user experience of saying,
well, I need to use an exchange.
I'm not comfortable using my own wallet
or I need to use a custodian.
And all of a sudden, the custodian is charging you lots of money
for you just to make up inflation.
That's a bad user experience, like being diluted
because you don't have access to delegation for a reason or another
or having to file a complicated tax return
with potentially taxable events every three days for an entire year.
That's a bad user experience.
And so in that respect, it does help improve the user experience
It makes it easier because I think right now
you have to think about delegation.
Like if you're not delegating, you're losing money.
Whereas we stick with adaptive issuance,
essentially it makes it so that on the margin
you shouldn't really have to think about it.
And that's, I would say, the main benefit.
I hope that answers your question, Predictor.
Now we're going on to Ollie Stewart who has a question
and also some suggestions he added.
He said, what more can Tesos do to incentivize use of the network?
And as examples, he mentions XTC rewards
for users of DeFi platforms, like Polygon has decided to do it,
quests on a much broader scale,
and funding open source development.
So just to clarify one thing,
this is a blockchain, this is not an organization.
So Tesos as a blockchain does incentivize some things.
Equity baking, for example, is incentivized by the Tesos chain,
as is baking is also incentivized by the Tesos chain.
I don't think the Polygon chain incentivizes things.
So I think it was Polygon Studio or Polygon Labs
or some entity that did this.
In terms of like getting rewards for DeFi,
many people have tried it.
Pretty much everyone did it in 2021.
You saw Avalanche do it, you saw Polygon do it,
They spent a very large amount of money to bring TVL.
And the only network where that TVL stuck was Polygon.
So Polygon was successful, as I would say.
It was not successful or cost-effective
in bringing sustained amount of liquidity in other ecosystems.
So I don't think it's a valuable pass
to getting more DeFi on Tesos.
I would say Tesos DeFi has deferred to sale
by being more innovative in terms of product that are offered,
more innovative in terms of the financial primitives being used
or in terms of the product being offered.
Like that's how it deferred itself.
And then maybe incentives can help.
But by and large, they are not typically sticky.
A quest on the broader scale, I think, is a great idea.
We have a Tesos quest platform.
The point of getting it on being a gas spender
is basically to prevent farming.
Because if you just say, oh, do the ascent,
People will create tons of accounts
and just exploit the system.
And you won't actually get a sense,
unless you get a sense of identity,
people can drain it and it just, that's what happens.
people will pay incentives for TVL.
They'll just have a couple of funds,
just like putting a bunch of TVL,
collect the reward, double the reward,
and then when the reward goes away, they're done.
So you're not really driving adoption
or many users doing that.
The beauty of doing it with ease gas spenders is,
in some sense, you can't cheat it
because you have to spend way more in ease gas
to be eligible than you would make for it.
So it's not like you can profit from it.
The other thing is it targets already power users.
If you're like a power user, you like these issues,
you like maybe lots of transactions,
probably you spend some money on ease gas.
So it's a way to provide civil resistance for Quest,
as opposed to asking for a full KYC
So it provides some civil resistance
and it also provides some civil resistance
that's super targeted in terms of users.
I mean, it's not the only possibility.
We could look at other ecosystems
where gas has been spent,
but it's nice as a filter.
Fund more open source devs.
That's also always a possibility,
but the work has to be really impactful
A lot of open source projects have been funded.
A lot of projects have been funded over the years
by the Zillow Foundation.
Not a few have been very impactful,
but many will end up languishing.
And again, the problem is you'll get a mercy selected.
That always happens in the sense of,
there'll be some great stuff,
but as soon as you start putting an incentive on it,
some people just get attracted by the incentive
and you get mercenary behavior.
I think like retroactive funding
where you seem to be missing
and you just come and say like,
hey, look what you've done, here's something,
can be more effective than call for building
or just like having people apply for grants and new grants.
And then last week we saw the launch
of the EterLink Discord and some of the roadmap as well.
So I had some questions that I thought would be good
for the community to know,
starting with EterLink is coming closer.
So could you explain a little bit,
what does EterLink attempt to achieve
and give us a little bit of a timeline of the release as well?
Yes, so EterLink is a layer two solution for Tezos,
which is fully VM compatible.
What is it trying to achieve?
First of all, it's a way to showcase the ability
to build complex and performant rollups on top of Tezos.
EterLink will not be the only one, of course, on top of Tezos,
but it's a good way to demonstrate that.
It also serves as a code base that can be forked
if people want to build custom rollups
based on the VM code base,
which has been a popular way of building rollups
It also provides a layer of compatibility with the VM,
which has been often asked by people who want to build on Tezos.
Maybe they've already built a prototype on Polygon
and they're choosing a network.
If you come to these people and say,
like, hey, why don't you rewrite all of your code
It's not going to be easy to sell.
However, if you can bring them into this ecosystem
without them having to rewrite all their code
and using something that they've already developed for,
So it's also a platform for compatibility,
an option and interoperability.
And finally, I would say even on its own,
A-Selling is basically gearing apps
to be pretty much state-of-the-art in terms of performance
and the features as far as Elto solutions go,
whether we're talking about latency,
whether we're talking about AMD protection,
which is not going to be part of the initial release
but is a topic of active work.
There's many VM solutions out there, right?
But I would say it was going to be state-of-the-art
in all aspects, not just performance,
but also decentralization,
which is where we've seen people be pretty lazy
in the past and in execution.
And so if I'm an Ethereum-based developer,
what would make me choose EtherLink
over another L2-based solution?
Well, it could be a very clear reason.
If you're hardcore about Ethereum,
maybe you're not going to make it.
But a lot of people sometimes when they choose an ecosystem,
it's not just a technical decision.
It's also about other builders in the ecosystem
who they interact with, community.
And for all of these people,
if they happen to already be doing EVM
and they like it as those in the ecosystem and community,
That's the most obvious thing.
For someone who's not specifically interested in it
as a community and who's just completely agnostic
and just shopping for an EVM-based L2,
well, we still have something to attract those people
because, like I said, it's going to be state-of-the-art
but also in terms of decentralization.
So if you care about not having a single sequencer
that we promise will open it up to other people,
but right now it has to be us,
yeah, that's not going to happen.
There's a sequencer, but it's an open set.
If you're looking for something
where some company has a multi-sig
and can all of a sudden override everything
and steal all the assets,
again, that's not going to happen.
If you want, permission is proper.
So decentralization is an aspect.
And if you're completely agnostic,
you don't care about it as a ecosystem,
you just want some performance EVM solution.
And you also don't care at all about decentralization.
I would say then we're still quite competitive,
but I would say, and again,
this is not part of the initial release,
this is active work on AMD prevention
or a strong AMD mitigation,
and that I think is something that people,
specifically in the device space,
And this is something I have been reading a bit about lately,
but I've seen some community members
The data availability layer that has been announced,
could you maybe explain it in short
and maybe give some use cases?
So if you look at the traditional blockchain,
like the original Tesla's blockchain,
the original Ethereum blockchain or Bitcoin,
you have these block creators,
so miners in one case or bakers.
And if you look at what they're doing,
typically they have three functions.
One function is to order transaction,
decide in which order are transaction going to come.
And that is the main point of consensus,
and that's how you solve interval spending.
You have to have some agreement as to
which transaction came before which,
so we know which one is valid.
The other thing that they do is they execute a transaction.
They compute the outcome of transaction,
it allows them to not include transactions
So they only include valid transaction,
which is a nice prevention against spam.
Because if you don't know if transactions are valid,
you don't know if they can pay their fees.
You know, it takes also valid transaction
is one that can pay fees.
It could be doing absolute garbage,
it could be calling in existence smart contract, whatever,
as long as it can pay fees, it's a valid transaction.
So they order the transaction,
they verify that they're valid,
and the third thing that they do is,
and that's a weird thing,
they make the data available.
Well, you could imagine a blockchain
where the block producers would say this,
and they would not tell the blocks to anyone else.
Like if you wanted to know what's inside the blocks,
Well, this network wouldn't have censorship resistance
because if you don't know the content of the blocks,
you can't create, you can't replicate the change,
you can't replicate the change,
you can't join the network,
you can't join the network,
you can't even make a transaction
because you don't know the state of the network.
So making the data available is a key part of this.
And the word data availability is actually,
I think this expression makes it more confusing,
and it's one of the things that I was,
in my journey, like understanding how the systems are built,
it's one of the parts that confused me for the longest,
and I think if someone had told me early on
that a better word for this data complication
would have made more sense.
It's about making sure that when a transaction is included
in a chain, that fact is broadcasted,
that fact is known to anyone who's connected at the time.
So a lot of the way you can scale blockchain
is by unbundling these aspects,
which doesn't mean, when I say unbundling,
oh, we're gonna have three different shims do it,
and they're all going to have their own token,
It's just a ploy, that's a modular ploy
to sell people more tokens, no, no.
I mean modularity in the levels of the code, right?
Well, with rollups, first of all, you ask for execution,
like a single rollup node can execute all the transaction,
and now the validators don't have to worry
about executing the transaction,
they just have to worry about making sure
that they've paid enough to be included on a chain.
But that type of ability, that becomes an exponent,
because if every transaction,
if the content of every transaction
has to be posted in every block,
you're limited now by the bandwidth of the validators
who have to replicate everything.
That type of ability layer is an extension
of the type of ability that's already available
within these blocks, but this one uses sharding.
So it means that instead of every baker
downloading every single block,
you'll have the bakers and also the nodes in the network
download a small fraction of every block
with error correction, which allows a dramatic amount
of scaling on the network ability layer.
And so if you combine this with rollup,
you essentially have people that made much
about the blockchain trilima to security decentralization
and scaling, and it basically gives you all three.
Like you get scaling, you get security,
and you get decentralization.
So that's the other component.
Rollups alone won't do it.
You also need to scale your data every year.
There are no use cases for it specifically.
The use case for it is scaling.
And there I say, it's the same thing for rollups.
Sometimes people think that rollups do something different.
Rollups do not allow you to do anything
that a blockchain doesn't allow you to do.
That type of ability layer does not allow you
to do anything that a blockchain doesn't
But what they bring is a lot of efficiency.
I think I understand it a lot better right now.
So we have some questions from a familiar
community member, Tesos Texas.
He says, it seems like a lot of positive things
There's an increase in communication,
new talents being hired, upgrades coming along,
new projects launched in gaming, et cetera.
His question is, what makes you excited
And what bothers you still or needs to be
improved to gain success?
I mean, I have to say, what excites me the most
is not necessarily what's the most important
videos, who's the most relevant videos, right?
I get very excited about technological stuff
because I just have natural information for it.
And so that's the excitement.
It doesn't mean that there's always the most
There's a lot of, like you've said,
there's a lot of positive things happening.
But I get very, personally, on the emotional level,
I get very excited for things like that.
I mean, you're launching and roll-ups working
and the whole work towards Tesos the most.
That's what excites me the most.
Very pragmatically, I think we're going to see things
like the Essendon launch moves a needle probably more
than all the stuff that excites me.
But again, there's a difference between the two.
In terms of what bothers me, I would say probably
And there's always going to be some limitation of that.
I wish we could go at a speed of ideas.
And that's not realistic.
But I started thinking about, I started getting
a full picture of how to scale in 2021.
And with mid-2021, I started getting the full picture.
And I wish we could have done it faster.
And yeah, the thing that bothers me is,
oftentimes, speed of execution.
The other thing that needs to be improved,
but that is, it's not clear how to get there,
is there's a perception gap in Tezos.
Tezos doesn't have as big of a community
as some of the top blockchains out there,
like Ethereum or like Solana, doesn't have as much users.
So there's a gap in terms of fundamentals,
like how active is the network or all of that.
But there's another gap, which is the perception gap.
And the perception gap is that Tezos is still perceived
far from where it actually is in terms of adoption
It's not sufficiently part of the conversation.
And that is something that bothers me,
the fact that you'll still have people compare,
oh, let's look at 20-year-old solutions
That's something that's actually being worked on.
I'm not sure how it will be cracked,
but I do think that there's something there,
because there's definitely a disconnect.
We're not where we need to be in terms of fundamentals
to really, really be knocking it down the park
compared to some of the top five or top 10 blockchains.
That also needs to improve,
like the fundamentals need to improve a lot.
But beyond that, I would say there's also
a perception disconnect that's worth fixing on this one.
All right, and as a follow-up question to that,
we've seen some gaming activity coming onto the chains,
which is specifically from Coinblast as well.
What has been happening behind the scenes
and what are you excited about on that field?
Well, there was a gaming vertical created at ShuiTech,
which is a Tizos adoption hub in London.
I would say the gaming verticals started in 2021,
but it didn't really start kicking off,
I would say, until like 18 months ago.
That vertical has been driving a number of deals
in the gaming space to bring more games to Tizos,
specifically focusing a lot on casual games.
And we're starting to see,
and it took a long time to build this pipeline,
when we started to see this pipeline now
finally shipping and bring users.
And it's very exciting that, you know,
we're seeing the numbers that we're seeing,
we've gone by in terms of adoption.
it's long work that's been going on,
so critical to Jeremy at ShuiTech
for doing this work and bringing more games to Tizos.
So, I'm excited about that.
I would say one of the rationals for gaming
is that this product market fits, you know,
oftentimes, blockchains are a solution
in search of a need or a problem.
And they are genuine in real use case in gaming,
And number two, in terms of like bringing many users
to a blockchain and bringing a lot of transactions
and activity, gaming has always been probably
one of the most promising one.
You know, if something is going to bring all of a sudden
like a couple of million users on Tizos,
is there something that's deeply cultural
And I'm excited about seeing more games
and starting to see cross-pollination.
I think when you have a user playing a game,
that's one thing, once they start playing a game
and then they discover other games on the Tizos platform,
they stop just being a user of that game
and they become a user of Tizos.
And that's really the conversion you want to see.
It's good that those games have a lot of users.
It's even better when they start becoming Tizos users as a whole
and start, you know, using another application,
maybe they get it in a game,
trade it on object or some other platform.
Like that's when really they become
unboarded into the ecosystem.
So we'll have to monitor like how this conversion rate happens,
how those funnels basically drive broader user adoption on Tizos.
I'm very much excited as well to see
what games come down the pipeline.
Thank you, Tizos Texas again.
And next up, we have Mr. Tastas,
who has a question about PR.
His question is, when will there be a budget
dedicated to getting articles in major medias?
So there is a budget for that.
There are people working in PR
and there are people working on it specifically.
Obviously, the budget is to pay people to do this job.
You can't pay the major media to place articles.
Some media will absolutely work,
So yes, to that end, there is a budget.
Traction has been a little slow in getting Tizos covered.
Sometimes it's a bit of a chicken and egg
in the sense that people will say,
oh, well, we don't necessarily have any readers,
but we are working actively on this.
Next, we have L9393, or LOL, I guess.
I didn't totally understand this question,
I want players to be able to run their own servers in my game.
Could I use Tazos' roll-ups
slash data availability layer
to create a bridge server
that tracks, validates, and coordinates
changes in a persistent world
shared between independent peer-to-peer servers?
So I'm not sure I understand it 100%,
because it's a fairly technical question.
I have an understanding, but I might get it wrong,
and so if I get it wrong,
feel free to contact me later on
and ask me your question in more detail.
But what I understand is that in some sense,
you have a multi-player game,
and you just connect in a peer-to-peer fashion
but obviously, they are taking action in the same game,
and their actions are consequences in this game.
So, for example, if you're playing a city-building game
and a building somewhere,
the building at some point has to show up on your screen
And the way this is typically done by games
is that users use a centralized solution,
like they have a centralized server,
everyone connects to the central server,
and the central server attaches to everyone.
Or sometimes they'll just use a leader,
so they'll just say one player is basically hosting the game,
and the other players connect to it,
and they become the server.
Basically, what you need to do to solve something like this
is you need to solve consensus, right?
If you want to avoid sheets,
and the way people avoid sheets in this case
is like people don't bother cracking the software
or doing those other things,
but in practice, you could try to cheat.
You need full consensus, essentially,
if you want to have this.
This is exactly what consensus solves.
You have a bunch of people who want to modify the shared resource
and they need to have the same view of the shared resource.
But at the same time, you also need very, very low latency.
You probably want to have something like a sequencer
for this to work well on top of rollups.
and the sequencer will give you the low latency
that's probably required for doing something like that.
Now, obviously, that being said,
the rule is not that very different from a blockchain, right?
You'll have better performance.
Maybe you can squeeze a lot lower latency
and more throughput and all the things.
But a rule of law at the end of the day is like a blockchain,
so it needs to receive messages from the server.
It doesn't by itself connect to the internet
or do any of these things.
Okay, I hope that answers your question, Lowell.
Next up, we have Cody Smith with a question about marketing.
What is the general plan for marketing and exposure
to the average person who is a passive consumer
that mostly sees misinformation?
Yeah, that's a tricky one.
I think you need to do more information
than there is information.
And probably content marketing is one thing
we've been quite weak in the past,
producing more content, pushing out this content at all level.
We've been decent at having some strong,
technically-written content,
not a lot of content from broader consumption.
So I think that, let's see,
we're looking at different things,
and right now we're in a space of measuring things.
And Varun, who started as a CMO recently,
would be a better person to answer that question than I am.
But it's looking at things like content marketing,
it's looking at things like channel,
seeing which channels are most effective
which channels are most effective at driving engagement,
what do people respond to running user service?
Like right now, we're very much getting a lay of the land
in terms of what moves people,
like how does it get information,
how does it form information about functions?
And it's only once we have this picture
that we can have a very strong way to approach it.
And I also hope we get Varun up for an AMA soon as well.
Much Apartment has a question about recent engagement.
What is the reason for engagement going up big
in December and January from the Tesla Foundation?
Is it Alex Bond or a new marketing strategy?
and again, we've had an AMA with Alexis
and I don't want to speak for him,
but I would say when Alex had started,
he wanted to have a look at what's working,
what's not working, what can we improve,
and communication and engagement, one of those.
And so it was a concerted effort to do better on that front.
And a lot of it was also fascinated and enabled
by some of the new marketing hires.
And then last question from me.
This year has been quite exciting.
We've seen the 5 million block passing
and this year will also be the 10 year anniversary
of the Tesla's white paper,
which is pretty big in a blockchain's life cycle.
So what are you looking forward to
in the next 10 years for Tesla's?
I'm looking forward to 2027,
for sure that Tesla's is still running.
There's a lot to look forward to,
but it's like I would say a stabilization of the protocol.
I mean, you know, I love the developments,
but I would love to get to the point
where it's like the protocol just works so well
and it's so performant that all the work
that's happening on the protocol
is just basically like doing formal proofs
about the protocol, hardening it,
chasing any remaining bug,
chasing any remaining source of performance.
Right now I have like this long list.
we probably shouldn't prioritize implementing this,
but if tomorrow we were to wake up
and this was implemented in the protocol,
we'd be happy that it is.
will it be nice if this were in the protocol?
Not that we would want to do it right away
because there's so many things to do.
I look forward to that list going down to zero essentially,
so that every good idea that,
because it means the chain has already been implemented
and the chain is just like generally very,
very stable in terms of our features.
That's one thing to look forward.
The other thing I would say is growing adoption.
There was a time where, you know,
like this was called a ghost chain.
It was basically like almost no activity at all on chain.
That definitely changed in 2021.
A lot of it was driven by NFT and art.
And we now have an active chain.
You know, there's a lot of transactions.
I think the other day we had like more transactions
happening on Tezos and Ethereum,
I am getting two levels where Tezos is one of the most
And I think it's possible because the absolute number
that we see today are still quite small in terms of users
So it's still very much up for grabs,
like who's going to use these things that,
you know, a lot of people using these things
is still very much for grabs.
So a little bit of the same,
basically being done with any tech roadmap
aside from hardening and proving things
and also like having a bigger community,
striding ecosystem, millions and millions
and millions of users every day,
that would be the things to look forward.
But obviously it's not going to happen on its own.
It's going to take an entire community
And I'm excited with this community to work on it.
And I'm excited for the next 10 years as well.
I was looking through some of the comments
and I didn't see any more pressing questions.
So I think we can wrap it up for now.
Thank you, Arthur, for joining me.
And thank you all for listening in.
Hope to see you all in the next spaces
and have a good morning, afternoon or good night.
All right. Thank you very much.