Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
🌱Chia🌱 Royalties guaranteed 🌱 Lots happened this week in $XCH 👀
Recorded: Aug. 6, 2023 Duration: 2:34:13Space Recording
Full Transcription
how's it going guys hopefully my mic and everything's working today
i've had problems the last few uh the last few times but it looks like it's all good
yeah i hope everyone is well good to see michael taylor in here as well i'm hoping you're going
to come up and talk to us about uh the stuff that you're doing it's uh intrigues me and i'm i'm not
a dev so um i need it i need it explaining to me in simple terms so uh yeah hopefully you'll come
up uh andy as well good to see you buddy let's get michael up here let's get um see if we can get
andy up as well who's um been working on pretty cool as well how are you michael i'm good how are
you doing yeah very good very good yeah i think it's the first time we're speaking um i think last
time you came on um gordon was hosting for me so um yeah no it's good it's definitely good to to
speak to you um let me say hello to andy as well he's just popped up how are you andy
yeah very good thanks very good very good yeah it'll be interesting to uh to dive into what you're
doing as well let's um let's just give it a couple of minutes so we can uh get a few more people in
grant i'd love you to come up as well um so you can maybe talk to michael about what he's doing as
well on the on the technical level if you're if you've got time would be great if not it doesn't
matter let's have a look see who we got there we go let's see if we can get grant up as well
hey grant how are you are you still uh are you still um croatia way oh yeah for a long time
cool how long you how long you're out there for
uh an indeterminate amount of time but a little while right cool that's the life to lead that's the
life to lead just uh just pack your bags and off you go for as long as you need sounds good to me
so um yeah let's uh let's jump in and talk to let's talk to andy first uh if that's okay michael
i just want to talk to him about um he's he's he's just released a token on uh obviously uh tibet swap
and um yeah i'm just wondering andy if you could talk us through it and um i'll maybe pin up some
of your posts that you've um you've been putting out as well if possible
monkey i'm not sure if it's just me but like i can't hear a thing andy's saying yeah you're a bit
quiet andy i don't know if you're you're near your microphone maybe you'll maybe i'll pop you
back down and then you can come back up again something oh yeah hang on i'll pop you back down
and uh yeah if you drop down let me drop you down um then come up again there we go i don't know if
your microphone might be a bit quiet that's uh there we go let's uh let's see if we can invite
you back up there we go can i make you a co-host grant
i might only have a couple minutes okay no worries no worries um well let's um let's uh let's talk to
michael first then very quickly if grant's only got a bit of time because um grant's probably got
better questions than i have so um um yeah so to stuff on data layer then uh michael i've seen bits
and pieces about um decentralized social media and all sorts this week so um where's it all heading
uh well where's it all heading uh we're gonna we're she is gonna take over how the internet is built
uh that that's where it's all heading so let's see like how how does it work then i mean give us
give us a rough idea of what you mean by it's going to take over the internet or how the internet works
is it is try and explain it to me someone who's not a dev um you know in sort of simplistic form
uh yeah so you know right now you know internet you know the way the way you know we we consume the
internet is uh is very centralized we usually get some sort of a server whether it be like uh you know
go uh aws or or go daddy or or or so on you know different servers and dns uh that you know
ultimately you know we have to pay to to to host our stuff up and in in other words like you know
the whole internet is controlled by by corporations and and what chia has the capability of doing is
actually building a true peer-to-peer uh internet where you can you can host uh you can serve your
your content completely permissionless like peer-to-peer like from me to you uh without
going without using any of these uh you know central uh you know companies and and really what happens
is it gives us back control over our data and what the and what data we put out there
i think one other thing that's worth mentioning too is there's been a whole host of like secure
issues have been introduced due to the way that we consume content on the web so there's things like
cross-site scripting and all this kind of weird stuff because you don't really know that you're
getting the data that they intended to publish or like what form the data was supposed to have
and so like when you do it through data layer there's like a different way that data gets to you so
that might be worth talking about too is just like how different it is to get data this way
because instead of it being like client server you're sort of getting known good data that's
coming out of a data stream that's been replicated to you in a way that's verifiable that hasn't been
possible before yeah and and you know one of the big things that uh one of the big epiphanies that i've
kind of had this past week is how you know we we look at it as data layer you know and we we think
you know we automatically think database uh but i kind of realized that what data later really is
is a peer-to-peer communications protocol like i can i can basically anything that i put in my data
layer is all is ultimately broadcasted out to anybody that's subscribed to me uh or additionally
like if i if i if i utilize permission stores and let's say like i create a permission store that only
you can that i allow only you to uh to subscribe to and you provide provide a permission store that
only i can subscribe to now we have a a secure peer-to-peer communications protocol uh where you
know we're we're we're talking back to each other and just posting our our last you know our messages
one data layer for the other person to to sink down right to their computer uh and and consume
um and the these abstractions like really had me super excited right now as a matter of fact like
uh this past week i built a a web2 gateway on top of data layer that essentially that mounts
uh a you know a whole file system into your data layer uh and and then uh reconstruction reconstructs
the file system uh through the web2 gateway uh so for the the for example now through the web2 gateway i
can basically put in any store key and view that content and if it's an nft uh now i can view the
content of the nft like in my browser or if you but it also works for like entire file systems so
like uh you know a uh single page application uh javascript program you mount all the files into
data layer and now i can serve the entire uh single page application in the browser uh through through
the web2 gateway uh and consume the application uh so now you got a situation where uh grant like let's
say you you put you mounted uh your application uh into your data layer and i subscribe to you
i now own a copy of that application and i get updates as long as you push them and i consume
that application locally um straight out of my data layer in the browser
so it's it's really really cool stuff that you know that that's capable uh with the right abstractions
on top of it i think one of the things that uh in my mind i when we first started like thinking
about data layer and kind of you were pushing us down this path content makes a lot of sense right
like i could see you know the people who have been doing things along the lines of only fans having like
a direct way to do this without having to you know go to some brokered centralized entity i think when
aceville is talking about how does something like mint garden continue to exist where he is
sort of providing a shell beautiful access to what is already ostensibly somebody else's data
that one's a little harder because it's like he's already serving up the content of others
but like putting it into a nice format in a way that like makes the interface good so it's it's almost like
something like mint garden you'd you'd need to make like a subscription model or something i i'm not
sure that monetizing data layer that way makes as much sense but like if you're doing something
so it actually makes a lot of sense like uh so one of the one of the things that i would do but you
wouldn't be paying for the update to the app you know unless like he did a new version of mint garden
and then you're like well i want that new version of mint garden because once you have mint garden it's
just the shell like you don't need a new version of mint garden it just does what it does so so
imagine uh so imagine that we you know mounting uh your your applications on today later and consuming
them becomes mainstream you know out of the box uh you have uh you know everybody you know it's
basically a a permissionless like sort of network where anybody can subscribe to anybody else and consume
their applications but data layer has uh you know the these data layer plugins called uploader plugins
downloader plugins that you can utilize to build uh you know permission to stores and you can very
easily uh build uh some sort of permission system where you only give allow somebody to subscribe to
your store if they've if you've detected that they've sent you a certain amount of xch and you can even do
that on a monthly basis so yeah the first first time someone pays for for to get access to your
permission to store and that pulls it down you know they own that copy forever but if they want to
get continuous updates you know you know software as a service model it's constantly pushing out
more updates improving the app over time if they want to continue to get those they keep you have to
keep paying month per month and uh totally but but i think it breaks down like i'm just i'm just
going to challenge you on that one because it's like you know if you build it and then you
distribute it and people are like well this is good enough i don't need your latest like
whatever minor thing that you did then you're kind of stuck and that's why i think it's i feel like
that's good that's a good thing though like i feel right now the fast model is forced to
whereas i feel like you know everyone's always like i hear a lot of people saying like i liked it
before where i owned my software okay so it kind of goes back to the model of you pay for the software
you own it and you know if you if you have photoshop like six and you never need anything
else more than that then you know and you paid for it then fine um but you know if you want to keep
being seven eight nine ten you know you you would keep paying for it and maybe maybe you skip a couple
versions so so i think it's like a kind of a happy medium uh where there you you also you have this
permissionless distribution uh system which is basically just subscribing to data layer uh and
permission stores and you know you have enough people doing it all it basically becomes a you know
marketplace especially if you have some other apps act as you know discovery uh uh software that
helps you discover which apps you might want to subscribe to it a decentralized app store starts to emerge
out of data layer in that scenario
but but can't someone just like copy the website code and then share it like to everyone and everyone
can just run locally
there there might be uh uh some challenges like that to uh uh to you know tackle um and i don't have
the answer now um but i think that you know that if we're in a situation where we have to tackle that
challenge data layer has matured to a very awesome level already
it is pretty genius stuff um you know especially like um you know grant asking these questions and
certainly while grants here we should give him the floor to ask these questions but uh i think once
grant's gone i'll i'll i'll go back to try to break it down to really simple terms unfortunately the
time is now i gotta go so i'll talk to you guys later maybe i'll even jump on so we'll see you soon
fair enough grant fair enough yeah so so so let's um let's take it right back to the basics then
of data lab because i think you were what were you involved with the you know creating the uh what
the carbon credits are essentially going to be sort of retired on was that was that something to do
with yourself uh yeah that i was the the architect and uh lead developer for the for cad t cool so um
going back then so let's just let's just start a data layer right from the beginning so what what
essentially is data layer data layer is built on top of the chia blockchain um and we like i said
like i think um um might be yourself or grant just said we we sort of think of it as like a data table
on top of the chain um but i don't think it is quite as simple as that is it uh yeah but before before
we go down that rabbit hole um grant and i did have a two hour deep dive that he recorded one data
layer uh i don't know if we want to rehash the same subject when when we could just go listen to it
yeah no completely agree yeah more than happy to answer some questions but uh but um i don't want to get
down into the rabbit hole and what i really want was really excited to talk about was what i've been
able to release recently that makes data layer development more accessible to the rest of the
community go go for that then go for that let's um i'm sure we'll pick up bits as we go um but let's
go for it dive in and tell us tell us you know exactly that and and and again like i i didn't want to
shut that down but if you are interested in the deep dive uh grant posted a uh you know a recording
it's two hours long like i i don't have that kind of attention span but if you do uh there's a lot of
interesting information there that we talked about uh go for it so so the the tool that i i had just
released uh on yesterday actually uh it's it's called sprout it's a it's a developer tool
uh you know you can install uh using a you know node package manager uh once you install it you
get a command line where you can do a couple different really cool things uh like one of the
things that i noticed that a lot of people had trouble with is getting getting their content on
the data layer in the first place um and what this tool does is provides a very very easy mechanism to do
that so so when you install it when you install the tool let's say you're um let's say you're a nft
creator and you want to get your entire uh collection on one data layer uh you install this this uh command
line tool uh you you navigate to the to to the directory where your uh nfts are located in your
your computer and you type in two commands sprout um create store to create a brand new store and then
you you type in sprout deploy and you you just wait and all your nfts are meet are pushed into a data
layer store ready to be consumed and you're done um and so that that's step one of the tool uh the step
two of the tool is you know how do we how do you consume these nfts uh so uh one of the things that
i'm working on i built i i embedded a development version of this in the tool the uh the web2 gateway
it is a way it's a way to access any any content on any data layer store that you're subscribed to
or your own you can view it in the browser so as long as you're subscribed to the store you're pulling
it down or you own the store itself uh through through some easy easy uh ur url schemes uh you can
you can load up those nfts that you just loaded on the browser and i demonstrated that the other day
uh i i asked uh chunks uh i was like hey i know that you you put your collection on the store you
know what's your store id and and they gave it to me uh and i i just i subscribed to it i plugged the
store id uh into the browser and it loaded up uh one of the chunks right there uh you know as an image
it was um and that's one of the things that i think that right there just made it super super easy
for every nft creator to at least get their uh get their collection onto the data layer and now and then
we're now building uh tools in order to get everybody else to be able to view their collection from from the
data layer so i don't know so essentially is that is that um an alternative to ipfs
yeah so data layer is an alternative to uh ipfs i i think they serve similar yeah similar roles
uh so but you know one of one of the things i i like about data layer over ipfs is if you're if you
have a chia wallet you have data layer and it's it's you know and it's it's it's a you know it's a
primary you know data layer is part of chia so whereas ipfs is you know another system so you know i think they
both serve similar roles but i think being able to natively store your nfts on data layer and then
making it you know very easy for everyone to consume it uh i think that's really powerful uh like
ultimately you have unstoppable um you end up with unstoppable nft collection like even ipfs
you know one of one of the nft collections that uh i got involved with uh about a year ago uh they
they paid for some sort of ipfs promotion i mean a subscription or hosting service and then they
stopped paying for it and now i have like literally 500 broken jpegs even though ipfs was supposed to be
some decentralized you know storage system uh whereas um you if you use data layer anybody
can permissionlessly act as a mirror to the nft collection so let's say that i i did sink in a lot
of money into into this nft collection and i you know and i have incentive to uh to make sure that that
data you know stays alive uh i just be set myself as a mirror and then uh you know i'm serving the
whole i store a copy of the entire collection in data layer myself for myself to consume and everybody
else that uh that uh it you know everybody else that wants to consume because now that now that i'm a
mirror uh you know broadcasting it elsewhere so i think data layer while still not as mature as ipfs
can be more powerful when it comes to you know this kind of scenario yeah definitely i think um
yeah that that's and it's also with you know contained within the same ecosystem as well which
um and i think if you're because because the cheer nfts have these multiple uris then you know why not
why not have have that option as well i think that's i think that's great i think that's a great idea
um where i mean where does where does it lead on to then to um i mean is that did you say it's called
sprouts uh so this the cli tool that i built is called sprout um yeah if you it's on npm.org uh the the full
name is chia sprout cli um if you if you install that as a package uh as a global package on your
machine uh you have you can now use that command line tool on any uh on any um folder in your on your
computer awesome so how does that how does that potentially lead on then to i know you've talked
about um uh sort of decentralized social media is this this the same sort of premise as what you
was talking about earlier on about how um essentially you can use it as a a messaging service would it
would you would you utilize data layer to um i guess be the be the behind the scenes to uh to a sort of
twitter looking front end yes uh so the the so the sprouts i developed sort of the sprout tool i
developed uh because i started uh thinking about what is the best way to go about building this social
media platform and and uh and then i realized like i need certain tools to just to make my life easier and
literally took me a day to write after i figured out what i was going to do so i was like why not
i wrote the tool uh released it but that is also i've been using it since i mean it's only been a day
but i've been using it like profusely to help continuous development uh continually develop additional
stuff uh that's going to lead up to this social media uh um platform but calling a platform is kind of a
a misnomer because um you know chia data layer already has this concept of following it's called
subscribing to someone's data layer uh so you know and cad t actually already did this actually when i
thought about it you know this social media you know application would work very similar to cad t
in that uh when when you start the application you you create a data layer table that has your
you know social profile on it and uh and then you you know have you know another data layer table that
holds your content and uh and then you can subscribe to anybody else's data layer and as long as it
follows the same standard in terms the same social standard uh it's detected as a social uh um profile
and it can be you can basically build a a front end that ties everybody's social profile together
completely off like decentralized like not not on a server not on a website like everything's like you
know running through data layer as a communication protocol not through like traditional means
um there's a couple challenges in terms of being able to reply to different people's posts but uh
you know the the the premise of uh creating a social profile subscribing to other people's social
profiles and putting posting your you know your own content and everybody who's subscribing you know
automatically gets updates that concept has already been proven in cad t that's exactly what cad
cad t is so you just take cad t and rework it to be around social profiles instead of organization
profiles and you you basically have this decentralized social platform that that's served directly from from
data layer you just need a good a good uh you know front end to uh to make it easy to navigate
so like basically all the all the pieces are there it's just building a standard in a in a ui
and some um stinking tools that are that are that do things for you behind the scenes so do you think
it's um do you i mean it's decentralized um i mean can you would you be able to link it maybe to
you know your wallet or your did uh you know is that is that sort of you know is that is that sort
of achievable that the way that you can really decentralize and really sort of own your identity as
such yeah so i mean uh don't okay so one of the things i got a bit clear is this isn't like a website
like twitter that you log into uh you're literally like this is software that's running on your machine
more akin to a desktop application uh yeah there might be like ways to get it into the browser
but it would all be like local host based like that's one of the biggest things that a lot of
people uh i think are missing is this is not like a a website service that you log into like this is
your software that you run and you're just communicating with other people's software
so what i'm trying to get at is you know if i can if i can communicate with data layer
there's nothing stopping me from communicating from the wallet um and i mean there's nothing to
stop me from communicating with any any of the chia capabilities so yes with enough development time
you can build like you know a very comp you know robust uh uh front end that allows you to talk to
folks and send you know one click button like you can send money to you know funds to somebody's uh
profile and they receive it and i mean you can do any anything that she can do you can you can build
into this front end like there really isn't a limitation there yeah because i think i was
talking about this in a space earlier on about how you know world coin essentially they're sort of
trying to do this proof of human type thing which which essentially is what she is doing with the
dids and you know verified credentials etc it'd be it'd be interesting to see how um you can sort of prove
that the person behind the um profile as such is the real one if that makes sense so you've got
this this proof that 100 this is the person you know if they haven't got this check mark
yeah i um
i gave that a little bit of thought and i asked a couple of i asked a couple of folks on the chia side a
couple of different questions that led up to it uh i haven't gone down the whole rabbit hole but i
was saying like hey you know if someone wants to prove that they are in they are them because you
know when you build a social profile you can build any number of profiles you want there's like really
no way to stop somebody from making 30 profiles if they want uh and and that might be a good thing like
different profiles for different uses uh but if you if you want to be an authority and you want to say
i really am this guy and you want to attach your your id to that profile i was like okay well
i think it's really easy to i'm not easy but i think it's there's a path forward to create a proof that
uh you this this xch address owns this data layer profile but then you need to and then it's a matter
of saying how do you prove that this uh this the per this person owns this xch uh so anyways i've
given it some thought i don't have it this solution yet uh but it's something that's on my mind i don't
think that's if i'm going to tackle that immediately uh but i do think that that problem is eventually
going to need to be solved uh to really you know makes such a platform such as this uh really powerful
yeah definitely that's it because i think i think that's with the birth well i say the birth but with
the recent advancements of ai and deep fake videos i think i think a platform that can sort of almost
prove that that person who's making the videos or the content or the tweets or whatever is you know
a hundred percent that person um is probably going to be the next the next big thing when it comes to
social media because i think ai is going to you know it already is isn't it deep fakes are really really
going to be a problem sort of in the next couple of years so you know if you can get that if if that can be
solved that way of proving that a person is behind a wallet then you know i don't quite know how you do
that either but it sounds like this this could be a solution or certainly a step in the right direction
well verified credentials i feel like should should solve some of that um i i'm not i'm not involved in
that initiative but uh when it's ready you know i'm happy i'll definitely be looking into it to figure
out how i can you know build that into the app in the applications um but yeah i think i think we're
there's going to be i think chia has stuff in the pipeline that that can help solve this problem
and i think that these problems the this technology that's being built absolutely can be integrated in
such a social platform let's open up to the floor i mean josh uh lucas and anybody that's got any
questions um and thoughts about how how sort of certainly that side of things could could go feel
free to chirp in by all means um does one actually have to run a full night to use data layer or is that
not needed no uh well it's preferred but it's not needed so so basically you could create um
a website which basically can show all the nfts basically like a marketplace through utilizing data
lay and also just getting all the nfts through the full node theoretically which then runs locally on your
device so so and i i don't know if i uh understood correctly but nfts would not
come through the full node they would come from the data layer uh the the the only the only thing
that's actually one chain at all is uh the proof of inclusion the proof that the data in your data
layer was in fact uh uh put there by you or your your wallet at least uh but uh with that being said uh
you don't you do not need a full node to run data layer but you do need to run a wallet uh there are
some quirky edge cases where it's like you know like the cad t participants we've told them we prefer you
to run a full node because there's some scenarios where you might not get information well well if
you're on a full node you're guaranteed that the proof of inclusion it is is a is 100 correct where
if you're running a wallet you're kind of trusting somebody else to give you the right information but
am i but these are all theoreticals in my experience like 99 of the time you don't need to run a full
node okay so currently let's say i want to use those chunks or show those chunks nfts in like a
local website i'd create a website then i'd like host it through data layer and then i would need
that website basically to subscribe to the chunk store and then i could retrieve those images and show
them uh yeah so um that the what you want to do in this scenario is you want to you want to take take
the chia software and express it as http server uh and that's exactly uh what the web2 gateway uh that i
built did uh the the web2 gateway can either run look on your local machine if like as a service uh i mean
well that's the plan it hasn't been built to do that yet but it can run as a as a win as a service
on your local machine so that any any links uh to the data layer can be can be opened up but you can
also take that web2 gateway and you can put it on a server and basically and what what you do is you
basically have a a a a server that's linkable that that has access to to your to your chunks so so the
the the the infrastructure would be you know uh you know hypothetically um you know ubuntu server
uh with the with the chia blockchain uh installed uh with the wallet and uh and data layer uh daemons
active um all the ports uh you know the only portion that should be open are you know like
port 80 or 443 uh and then on a lot running alongside uh your data layer uh you would uh
you would run the web2 gateway and bind that to port port 80 or 443 and uh and now uh you can access
that you know access that server from the web and you can access all the chunks that are in your data
layer now you have to be subscribed to the store for that to work like you don't get access to
anybody's data layer but everything that you're that that machine is uh it's subscribed to
is now viewable through your through your web2 gateway
it's it's it's it's incredible tool and i guess you know you'd know better than most of you know the
future use cases and um what this could be used for i mean you know potentially is there is there scope
for things like medical records and uh you know all these things that blockchains claim to be able to
solve or are trying to solve um is the is this sort of data layer the way to go for all that sort of
thing you think i i think that someone with enough funding and time and interest could absolutely uh
build uh something along those lines uh personally i i i know that it's there's so much more involved
than just building the software to get medical records to be widely adopted that i i generally
avoid the subject but i but it's totally totally doable uh and and it's i think it's actually a
great use case for those that that want to tackle it go on lucas go you know that is um that's a great point it would be
interesting if somebody um started a movement to have like a personal copy of your um of yours
um to to to to begin the concept to get to uh i don't know not be the official record as that builds um
i was also thinking about like andrew's file system or or like having um uh you know like like the
ability to basically have my auxiliary accounts or data sharing from my desktop with a node to like
my phone and whatnot do you is that something you had considered or thought is possible
sorry say that again so i can think it through basically like you know like your your the cloud
systems that they have for for your preferences on your on your applications from your your desktop to
your phone like rather than google and whatnot it would be nice if you actually hosted that yourself
and then had devices that could phone in and get that information i mean if you're running your own
infrastructure uh and you have the right you know security precautions up uh you know you can absolutely
you know stream that information directly to your phone um whether it's data layer or or
chia or whatever i mean you can i mean if you i mean the rpc the chia rpcs are are just
you know direct their their api either http you know requests the libraries um so you know you can
you can always call the rpcs remotely through http to interact with your full node it's really all about
the security and locking that down like if the if the concern security security wasn't a consideration
it's like super super easy you just open up your ports and then just tell your application your phone
to use uh you know your rpc your your full node uh ip address uh to call rpcs
is there is there is there a cost is there you know a um a network cost to every time you change a piece
of data on data layer yes uh because changing a piece of data on data layer requires a new proof
of inclusion to be added on chain anytime you interact with a chain you gotta pay a fee
uh so every time you make a change that is a blockchain transaction that's involved um
it's i mean it's it's the fee isn't a lot uh but but is there so you do need mojos to be able to use
data layer yeah so um you know obviously fees are tiny at the minute uh i guess i guess eventually they'll
they'll they'll probably have to move certain parts or certain pieces of data on to some sort of layer
two maybe uh it's funny that you say that um the first thing is uh um the the nice thing about the fees
is it's not the fees are not proportional to the amount of data that you're pushing uh it is simply
the fees are simply to cover the proof of inclusion you can put like the actual data itself like does
not affect the fees but speaking of layer two um i'm actually like 80 percent done a i guess you could
consider it a layer two uh solution for data layer uh this was something that i needed uh i felt was
important for um for uh uh uh this chia social to to be realistic uh so i have a project that i'm
working on uh that i might actually get released before too long that i call uh data uh chia data
layer high input output uh and the way that this works is uh
uh it it it creates a staging table on top of data layer uh and allows you and so when you when
you're your application is interacting with data layer it actually doesn't push the transaction
immediately uh what it'll do is it'll batch up all your changes and on either on an interval or manually
it'll push all the changes at once uh that uh because and this this what i think this is really
important because applications can't really use data layer currently in its form as its main application
database because every change requires um you know a transaction and transactions take a little bit of time
to uh to uh to propagate uh so it could get confirmed i mean uh so what happens is you end up
like forcing the user to slow down and you have to block them from further activity until your
transactions are done um whereas with this uh new abstraction building on top of it your application
can just run as fast as it needs to run and then it just kind of does a checkpoint and pushes the
checkpoint to data layer all at once uh whether on an interval or manually uh so that's actually
the next thing that i'm going to release fairly soon that's uh that's music to my ears that is um
because because the way i see certainly with the monkey zoo project is um essentially using data layout
as the way that different games can look at assets so for instance a game you play a game with your
character um pick up an object uh you know equip it to your character it updates the data layer to
say that that that this item now belongs to this character um and then you take your character to a
new game and the new game looks at that piece of data and essentially says right well this character now
had this this item you know attached to attached to their character so um i guess that was one of my
concerns was having to do all that those updates every single time would get expensive after a
while but what you're saying is you can sort of bundle those together and make make one change every
so often which seems to make sense yeah so so the cad t already uh proved the concept um i'm just uh trying
to build it in a smarter way now that i'm older and smarter and actually know what i'm doing
uh but with the cad t we have this concept of a staging table and uh and the the user uh you know makes
as many changes as they as they want to their to their carbon registry uh and then when they're ready to
commit it uh they you know press a button in the ui that says commit and it pushes it all through the data
layer all at once so the the the data layer high input output abstraction that i'm building uh kind of uses that as
as inspiration um the the interface is actually almost the same so you can enter you can just swap
it out uh to existing applications um just you know point to a different port and it'll all of a sudden
use you know this layer two abstraction instead of without any additional coding uh the only thing
that you would have to do is you know set a few uh configurations for how often do you want to push
the new updates uh and then if you want to programmatically invoke a a commit to the data
layer manually you know you would just have to write that in um but yeah like anyways long story
short the concept's already been proven i'm just making it better and easier for everybody to use it as a
package brilliant brilliant as yeah i mean if anybody down below doesn't follow michael taylor then um
you should certainly follow him and take a look at his tweets because it's um it's blowing my mind
it really does blow my mind but um hey it's just really clever stuff go on josh hey thank you thank you
uh michael so have you one of the things i talk and talked about a while back in this space so you've
got the web gateway now which is awesome you've got this concept of um uh what we were just talking
about which is abstracting further abstracting the the database into its own little fast database which
then gets synced to data layer over time uh because that's the that is i guess the problem with data
layer currently not not the problem but the the the working it is it's a problem it it it lowers the
feasibility the applications that you you can use like i want to run a minute yeah yeah yeah it's about
every minute or so so that definitely limits it from that perspective to um what i call or it's almost
like the uh write once read mini worm the old worm drives i don't know if you remember those in the
old days uh it's like a write once read mini that was that's the the concept is you want something
that you know it may take a while to write but you read it lots of times after that because it never
changes its static data and that's where you get your your performance improvement but anyway um what
here's here's one of the things that i talked about a while back and i loved and i think you're so close
to now which is um i would have you ever worked with azure uh just the azure files or the blob storage
or the tables or any of the queues or just the basic like the base uh primitives in the azure system
microsoft system cloud system i i have i'm more of an aws guy uh okay that's and i know i i know based
on some of your other work that you've done a lot of aws stuff too so just just quickly this is actually
and i'm sure aws has um a common uh a similar concept but in microsoft's world it's it's azure
you've got um tables queues blobs and uh files i think um blobs and files are similar but you can
do like a directory structure and files and things like that but anyway these things are the very simple
primitives of the azure uh back-end you know file system that you can use uh and you can serve static
sites directly from it you can do all this stuff because it's all just really you know web accessible
but there are these very basic structures that they've defined those of course are pluggable so
you can implement your own back-end system that that looks like an azure blob storage service or an azure
file service or an azure uh queue storage or q service like an msq the old you know queuing system
um where it's just like pushing you know popping and pull uh things like that but anyway so the
idea here that i that i wanted to implement a while back i'm sure amazon has something similar is similar to
your web gateway but let's make um let's do the implementation of data layer hidden behind what
looks like to everybody else in the world a common azure front-end uh system so that all of the code out
there that we've already written that talked to azure common primitives just magically works with our
data layer system they don't care that it's now on data layer but you see what i'm saying like i love
that so that's what i so that's what i well i sort of did that uh with this um with the sprout tool
uh so i mean it's obviously not azure but uh you know i i basically i abstracted data layer to to
have the same interface as aws s3 uh so love it and that so what i so basically uh
i i i treat the data layer as a object store um and and then i use what i did was the sprout tool
that the sprout cli tool that i just released uploads data into the into your data layer as a object store
as a aws uh s3 bucket it it pretends it's a bucket and it pushes the data in the same way
and in the web 2 gateway consumes the data layer as if it was a aws s3 static web host or uh or just
object lookup depending on how you structure your data layer and i've actually been able to serve single
page applications in data layer uh because of that uh it's it's like so cool because like you know
you build your single page application you know the the entry point to it is like index.html
but then you know it has a uh you know you might pull in style.css and that style might be located in
assets slash css slash style.css well by in the same way that aws treat doesn't have folders it it just
the folders are virtual it's basically the entire path is the key uh that that's how i store it in data
layer so now when you load this index.html from data layer and uh and it wants to load you know styles.css
it makes a http request to the web 2 gateway saying give me style.css at this path and it finds it
so to the point of view of the browser it doesn't it doesn't know any difference it's just it's just
requesting a resource and it gets it uh so i i think i guess what i'm trying to get at is i think i did
exactly what you just said except i i did it for aws s3 instead of azure yep that that's almost exactly so
the next and of course what you're talking about now with this uh staging table is almost if you
took that a little bit further so so azure also has the primitive concept of just tables you can
actually define um very simple uh columns and rows and then it's then behind the scenes it's still all
just blob storage on the on the system of course but but again you've abstracted um right right now
data layer is literally key value hex right like but you you now have abstracted this further up
where you can have uh tables and columns with different types and things like this like a like
well i i didn't do that i want to i should say i should say that's that's what you're going for yes
absolutely okay so basically so basically the the the the chia uh uh uh data layer high input output
uh uh project that i'm working on um i originally actually was building what i call chia ql
i wanted to build a a sql lite um a sql lite uh database that ultimately stored itself on data layer and
then could and basically you would you would read from you would read from the sql lite so it's all
queryable and all but you were right to data layer which then sinks back into sql lite uh so what you end
up having is a uh abstract you know an abstraction or data layer that's completely queryable even though
like you said under the hood it's all blob storage and that's exactly what we did in cad t actually
um the problem is when i started thinking about it i at the moment i got stuck on how you denormalize
sql in a generic way so it works for everybody and not just my one application
and um i kind of i got really stuck so then i thought to myself you know what i'm just going to
keep it as a key value store but just make it make it the staging table like key value so so even though
we're not queryable the same way i wanted to at least we still have this staging table concept so
you still get the high input output so anyways long story short i'm thinking about how to do that i don't
know how yet but i'm very interested in solving that problem yep that's that's uh that's a great thing
and well and this also does go back to uh somewhat goes back to the the whole only update every 60
seconds thing because even you know like one of the things that we could do is say well let's let's
start doing like games and things on data layers so like even imagine a chess game based on data
layer where every time you move your piece you are updating the data layer that somebody else sees and
and then makes their move i mean you can imagine how boring that's going to be especially for the
opening moves right like as you yeah so the one minute per move you know i know you know when
ethereum first came out i'm sitting there i i thought i was like you know what this is gonna be awesome
because now we can have like we can like build pokemon where like every where your location is like
cryptographically verified and and you can trade potions as nfts and all that stuff and then i realized
like every every move is a blockchain transaction and it's the same way uh in in data layer if you do
the same thing every move is a blockchain transaction now you can build this l2 on top of it but that's not
it's not going to be cryptographically secured until it's promoted to data layer so
i don't know if we're there yet i'd be interested in to see what people can come up with with this
layer 2 solution that when i get it out there uh but at the moment i still don't think we're we're quite
there yep i i agree i actually i see a lot of potential and data layer as a uh uh peer-to-peer matchmaking
service whether it's for games or even applications uh very similar to how you publish your own uh
mirror of somebody else's singleton by updating the somebody else's store by updating their singleton
with your ip address um you know that's how you kind of decide you let the world know that you're
also a mirror of that other store you could also imagine a world where similar to that you are updating
uh a singleton letting letting another letting this chess game that we're talking about know that
you are at this ip address and so other peers can discover you and and challenge you to a match
and then then it connects directly peer-to-peer on a on an l2 type of channel a side channel and now we
can play a chess game you know at the speed of normal internet so you just described you just described
how i want to build chia social instead of but instead of a chess game it's it's messages yeah
exactly by the way did you read my uh paper from last year i think it's very similar um
i did not send me a link uh input i will yeah i think you'll really like did you just say inbox
outbox like in terms of like an email system yeah yeah that's what that's so cool because like two
nights ago like i was just thinking like we could turn we can abstract data layer into an email
system and oh yeah you actually put out a tweet like about that uh so it's so cool yeah i meant to uh
i meant to also say about that the the email system one of my i got so excited about data layer last
year early on when it was announced um and i was kind of making guesses at what it was because
one of my early back in like the early 2000s i kind of designed out a base very basically just
kind of a proof of concept uh of a distributed uh system based on email based on smtp and pop3
so you instead of sending around email messages to each other like we did back then anyway and still do
today uh this is this is back in early 2000s instead of us sending emails to each other our apps would
actually send emails to each other but the apps themselves would just be would contain snapshot
delta snapshots of the changes to our our shared database and whoever was cc'd or whoever wanted to be
in on it to subscribe could actually get in on it and we you know now our apps would be sending out
so we're using we're freeloading on the existing smtp and pop3 systems uh for email to to establish our
decentralized uh database network so i thought that was a cool concept but yeah then i got i saw data
layer i got really excited about it because yes exactly we could we could we could rebase email and
everything on this eventually that was the big epiphany the payment whatever you want to call it that i had
when i realized that you can build and you can you can express chia the chia ecosystem as an email server
that's when i realized data layer isn't a database it's a communications protocol and once you once you
realize that you realize that we can rebase all the the traditional internet infrastructure
from http servers to databases to to email servers to everything we can we can rebase all that to use
data layer under the hood and now you have a decentralized peer-to-peer internet yeah right and
and they don't the the apps don't even care right because like for example they're just talking smtp and pop3 but
behind our custom smtp gateway just like your web gateway it's a data web web 2 gateway to data layer
we're storing those physical email messages inside data layer and and syncing them to our other notes
and see even internally this is the cool thing too for enterprises so maybe you don't necessarily and
this is i guess getting into more like verifiable credentials and stuff but maybe in the enterprise space
you don't care about uh other people syncing your database necessarily you just like to use these
features to sync all 15 of your servers across the world at different locations together right like
so even just from an enterprise space that's technology or stuff that you need to pay for
before even if you did care we have a concept called permission stores
brilliant so see yeah this is a virtual private blockchain right the bpb type stuff that gene likes to
talk about where yeah basically um yeah and and you know before that was to do geosynchronous uh
databases you know you you pay a lot of money to azure or amazon things to truly keep that going
but if this world i think i can't go into details but i will tell you that there's a lot of uh
uh a lot of proving of this what exactly we're talking about in the enterprise space right now
and everything that i've seen up until now has been working splendidly
that's beautiful yeah uh check out my my diesel paper by the way i also talk about plugins throughout
this because one of my big goals with this with the decentralized uh social media service is that the
the user all um there should be several hooks and plugin points in the in the social media
app and framework itself so that the user can um pick their own algorithm their own timeline
algorithms their own everything i'll tell you this about the social network stuff uh my goal
is to get the base system proof i need i want to prove that it works
how i get like basically mvp out there and uh just to say hey look she can do this
uh unless i get some sort of funding or something beyond that i don't think i'm going to work on
beyond that but i will be but it will be open source for the community to pick it up and turn it into
the most beautiful network that you could possibly think of
okay we'll definitely check out the the thing i read there's some interesting ideas in there one uh
one thing one epiphany i had by the way this is just kind of a cool i think in general you've got
to use some sort of id for each post or tweet or whatever you're going to call it right like we have
to decide on some sort of id whether that's the the hash of the thing or whatever it is but my proposal
was to use what twitter calls the snowflake which is their custom uh basically a good if you think
of it it's like their custom format for generating a new id for a new twitter post so use their exact
format uh because they actually they actually open stand open sourced their their id format and call
it the snowflake that's what they that's called its name and um it allows for other vendors to create
their own snowflake numbers but it's all still matching so anyway the reason you do this is
because what we want to do with social media in my opinion is not try to say hey we're the new twitter
like like meta like come on you know that didn't work uh threads or whatever they call it uh what we
want to do is we want to say no no we're not we're not trying to compete with twitter we we want to
we want to integrate with twitter we want to start bringing your stuff over so you you the very first
thing you would sign in the data layer you'd sign into the the the chia version of the social thing
and it would import all of your tweets from twitter using their exact existing ids and everything so
that as you you actually import all of your social over from that place and all of our links and
everything still match up because all the same ids are used and everything but then going forward we use
our own snowflake ids from that point um and then you even maintain sync right like so you can still keep
using twitter and as you make tweets over there we'll we'll pull those in as well we'll use gateways
for all this stuff right and no one will even know at first we'll we'll slowly uh slowly uh make our way
into the social networks uh that way so not just twitter facebook you know anybody whatever other
network comes along we'll say okay well here's our next one for that your side comment kind of
hit the nail on the head though when it comes to this in order to use the social media you have to
i mean this social network you would have to have a chia wallet or some sort of
application interfaces with the chia wallet um so in in the beginning i kind of i think trying to get
mainstream adoption is going to be difficult it's going to be more like a closed circle of really cool
people uh but i think one you know i i think i think it's good i think getting it out so it's accessible
to everybody is going to be the next big challenge uh once we actually get the system working
definitely definitely sorry lucas i just saw your hand you just just speak up man
all good not at all i don't want to interrupt y'all y'all keep talking as long as y'all will and then
then i'll talk oh good go for it well what i was going to say uh to the mail comments also um
i was talking with dustin and cbdc token in that room the other day and i said to them i said you
guys ought to be thinking about wallets like post office boxes and uh and uh your jpegs as political
advertisements um because uh you know this that's what it is it's a post office box it's a community
this is communication um so i had been i had been wondering along those lines and thinking about
data light data layer in along that line too i'd also been thinking about it in terms of people using
it um uh with also like putting limits as to the like the you know like so many mojos anything that
costs less than these amount of mojos can't come into my wallet and that way you could you can get
rid of uh spam in or at least what people make sure that people want to pay to send you a message well
here's a here's uh the the challenge well maybe a challenge maybe not but here's one of the features
of the data layer that uh uh you gotta remember um we we don't have a way
to build stores that anybody other than the owner can write to um a grant called them community
stores like we don't have the ability to create community data stores the only person that can
natively write to a database to a data store is the person that owns it um now uh we're working on
i'm working on solutions to create these community stores uh the the data layer that storage website
that i've been working on it is like an abstract abstraction although the only problem is it's it's
it's centrally managed by me uh you know it's not like native crypto like community store but i still
create the ability for people to through their credentials to modify the stores that i'm hosting
um but i guess the point i'm trying to get to is when when you think that realize that you i can't
actually send you a message but i can post a message on on my one of my stores and if you're
interested in reading my message you subscribe to my store and and you automatically sync it down
so in terms of the email system um i still gotta give it more thought but my initial thoughts were it
would be one of those things where you you start off by setting a communications channel a communications
channel is between you and one other person or or a couple other people uh you you and in uh in
determine number of parties and if you put a store a permission store that says only these people can read
my store or one person and then they put they create a store that's saying only you can read their store
so every time you post to your own data store everybody else that's that's allowed to listen to
your your everyone's everyone else is allowed to listen to your store will automatically get the
message and when they post to their data store since you're allowed to listen to your theirs you get the
message so there's a little bit of a handshake that needs to happen but the super cool thing about this
is if you can figure out a way to do this in a way in a way that's invisible to the user
you've just solved the problem of spam mail because no one can send to you without your permission
well in but in in one regard i think there's some distinctions here too and we were talking
again and i think with monkey zoo um about uh you know i studied literature oh michael did we lose you
uh studied literature or no drac came up and um and fun fact uh when you write a letter and send it to
somebody you can write them a letter later and say send me back all my letters because you own your
letters and um it's interesting to me that the the internet in the copying of things um changed that
notion in some ways but i really like like one reason i really like telegram is i can pull my messages
back or i think i can i don't know if i can or not um so i actually i i think we were talking about
this in regards to using mirrors the other day too in in order to uh you know like broadcast your
uh your content that you wanted to share out to other people so then you could you could delete the
source and and um have all your messages back another thing though with it with a post once someone
just subscribed to your store pulls their updates they have it forever like it's theirs they own it
like they have this physically sitting on their machine now you might make an update to the store
it says delete the message that that they pulled down and if they look at the latest uh uh merkle root
of the store they're going to see nothing there but the store is trans very transparent
and auditable and you can actually roll back and look at any merkle root uh in the history
i think i mean yeah i i actually like both of i like that like both of those aspects because to me
like um i don't know i don't i well these these get to good questions but i i see where you're going
there yes so yeah so i don't know i don't know if there's a way to say and this would be cool if we
could figure this out because then you then you basically have a a really awesome like uh a sas
model you can build in terms of software distribution uh it's kind of what grant and
i sort of touched on the very beginning is uh you can you know you using the the sprout tool that i
built you can build an application and mount it to the database to the data layer and then you can
build a permission store that some in order for somebody to be subscribed to your store they have
to send you xch and now they have access to your store and they're able to pull down a whole copy of
your store and they and they use it when they run the software it's running on from their machine not
from yours um unless you put up a web 2 gateway which they can access it through like a traditional web
server but now the problem is what if they stop paying well in the currently uh you have to consider it as
once they paid for it they own that version of the software forever because you cannot force them
to to delete it uh but uh but if we could ever solve that uh i think that could be very powerful
in terms of software distribution because you can say you you do have software as a service at that
point they if they uh they they buy a subscription to your software and as long as uh they're paying for
it they're able to sync it down and use it and once they stop paying for it it's you basically do a
recall um that's not possible yet uh but i love the fact that we talked about it because i'm going to
think give it some thought i think it uh it actually could be today on chia if you used uh nft ownership and
nft signing so now assuming that the application itself does something on the chia blockchain so
if you built into your chia list your custom chia list puzzle i assumed that your application would
use if you built into it hey this must when you spend me you must be signed by an nft from this
collection as well then the owner would have to have the nft from that collection in their wallet
that gives them the license so to speak to your app so i can use on-chain functionality of that app
anyway yeah i can totally see that being built the problem is that
an spa app which is a you know single page application that's built in javascript
it's nothing stopping somebody from going into the source and commenting that code out
however uh if you didn't still if you did not distribute the software as a spa but just
but distribute it as a just a binary download so the data layer actually stores your uh some sort of
executable um that where the source code is hidden i totally see that possible as a matter of fact i
wonder if we can go interface we run the executable directly from that data layer instead of running as http
easter that'd be cool well yeah with uh chat by the way my we also need to talk about that some more
um that's offline maybe but my the whole chat framework that i'm building instead of an executable
it's actually a dll but it's that's how you would build apps for this this framework that all be stored
as uh dll's that would get loaded up um but but sorry back to so it's not just the so the app itself
you obviously if you own the bits of the app then you're going to be able to execute that app on your
computer but if that app is a blockchain app and it writes to the blockchain to do its work
then at the time that it writes to the blockchain it can say hey are you licensed to use me oh absolutely
you see what i'm saying so i'm i am only thinking yeah yeah i'm only thinking through the lens of
i just pushed this javascript application to the data layer and somebody else is consuming it uh but
yeah that's awesome because yeah one of the things that that that i have to always be careful of is
not to be limited by my lack of imagination when i say things can or cannot be done
you're right well and and that's always been the the problem the classic problem is with with
software licensing is that again like you said earlier once you once you give the user bits then
they own the they can they can crack it they can they have years to crack it figure out a way to do
whatever they want so if it's just about running the application then yeah we haven't solved that
but one of the classic ways that currently uh or not classic one of the the newish ways that
software companies realized is subscription model right because we have the internet
so what we can do is we can make some features of the software dependent on an internet connection
back to the server and so that way if you don't pay for us then we'll cut that off and yeah you may be
able to start microsoft word but it's going to come up with a message that says hey you don't have
all the features enabled because you don't subscribe to our cloud service yeah so you're saying do that
with a puzzle so yeah yeah basically you know if it anything that's anything that's actually
communicating with the blockchain like if you don't have a license you can't communicate with
the blockchain exactly if it's useless to you possibly right because it's like okay you can
maybe you can still so for example uh maybe you you purchase a uh a portfolio type app that helps you
manage your your portfolio in a nice way better than the wallet currently does maybe um so similarly
you know you could you could make transactions through that portfolio app um but only if you
have a license and that makes it much easier maybe for example a really cool portfolio app that i
dreamed of recently or a trading app would be one where i just put in an amount of for example space
bucks i want and it looks at dexie and at tibet swap and says okay you should buy this much from
dexie and this much from tibet swap to get the absolute best price on the amount that you want to
purchase and it just figures it up for me and then makes those two deals for me automatically
so you could charge a premium for that right for that little service right there um but only if you
have a certain that transaction and i'll do all the fun stuff so basically what you're what you're
doing right now is you're just making me like 10 times more excited than i was like a minute ago
once i people get the hands on the stuff that i've been building i want like to i i think that
that being able to build this this new internet infrastructure once again once once it becomes
these tools become mainstream in terms in part of everybody's development platform i feel like you
know people like you are going to build like super awesome things that i could never imagine and that's
just super exciting well and it's this is actually where uh i think you and i have similar goals we're
just starting from different sides of the spectrum so you're kind of at the at the data layer deep
end and i'm at the higher app dev end where uh my that's my whole concept of chaps right so
the idea is is that an app developer should be able to do stuff in chia very easily without
having to worry about how to distribute that app but still make it all on chain and licensing and
all that very much like apple's app store or the windows i'm interested in that because you know
i'm uh i'm day or layer through and through but in all my day or layer work i've had almost no reason
to touch key list and uh well and and i've read about it and i want to use it but like
nothing that i've been doing requires you so well right yeah exactly the the primitives that they've
given us and this is what i try to tell people too who want to jump right into chi list it's like i mean
feel free to learn it for sure to understand what's going on but you probably won't be doing
a whole lot there i mean unless you're writing you know you're yak and you're running next to
back swap or something like totally novel from that perspective you're going to get 98 of your work
done with all of the stat the feature stack on top of that stuff you don't even have to get down to
that layer yet right um yeah so anyway uh but yeah if we can we should definitely work together on on
on the this kind of app dev framework to make it easier for devs to to just use all of this stuff
as abstractions and easily deploy things um well for example if you did want to like to bet swap um
you want to deploy you make some really cool fancy new chi list how do you actually get that into your
into your user's hands you have to somehow distribute an app like uh you know a custom
like freddy did with tail database or some sort of custom app that runs on the client side that does your
custom chia list or knows what to do with it um there's there's one of the developers out there
i can't remember if it was enderton or or cacman or somebody has talked to they were very interested
in using by the way what's that i said i'm in your town i didn't know who was talking this okay
i don't recognize voices yet i only see everyone through twitter twitter yeah i'm josh uh endertown
oh gosh okay i think you were the one that was telling me about it
i'm basically going to regurgitate your it using data layer to this to distribute um uh uh
chia list uh you know puzzles yep that's exactly right yeah and not only that um but running those
so data layer is basically not only our replacement eventual replacement for github because if you think
about it github uses a very similar uh git itself uses a very similar merkle tree type concept even
just for source code um but so data layer obviously is a great replacement for github eventually but it's
also a very interesting execution environment itself because if we if we did fully move to lisp as a top
of the full stack language then you could actually execute lisp directly from hex data structures in uh chia lisp i
mean in in from data layer but are um uh verifiable through the chia lisp on chain so your programs now
that they're all in lisp could actually you could use the the outcomes of lisp running on the client to pass
into the chain naturally it would all just be one big beautiful merkle lisp or a merkle tree uh chia lisp
uh program if that makes sense yeah yeah but that's that's way way down the road that would be like
some sort of you know compiler maybe trans compiler transpiler that would uh translate code from one to
the other and all kinds of fun stuff anyway yeah i don't i don't i mean it might not be as far as you
think it was like i've been i've been sitting on this web 2 gateway idea for about a year now
and i was kind of afraid to tackle it because like i got so much work during the day like trying to do
like regular unpaid work in between is like like my biggest challenge and then so like i've been kind
of sitting on this web 2 idea for for a long time and then like literally like a couple nights ago i had
like really really bad insomnia and all i was thinking about was chia and then all of a sudden
like i i like jumped out of bed i was like i figured it out and i went downstairs and i wrote the whole
gateway in like an hour so so what you said here michael is in between your um your normal job you're
just going to recreate the way that the internet works
i'm trying to figure out a way to get one way or another i'm trying to figure out a way to get
funded to work on chia all day every day nothing but chia all the time so like uh you know i i'm
working with uh you know the the the chia the company uh on a part-time basis uh and which you know
oftentimes turns into full-time but it's not on paper so like i don't know if that continues
tomorrow or not so but because of that i i have to maintain my other list of clients which are
completely not blockchain related at all and i hate working with that stuff like it's just to pay the
bills and i'm just sitting there thinking like if i could just figure out a way to only work with chia
all all day all the time like like i just need a year of that and like i feel like i can like do
so much cool stuff i'm always tempted to self-fund myself i'm just gonna say i'm quitting everything
for a week for a year and i'm just going to work unpaid and then come back a year later and see what i
had what to do with myself uh but yeah you're right i i am trying to squeeze all this in in between
and uh it's very challenging but the you know the only thing that that's kept me going is how
exciting this is like i i as a developer like this is like the first big passion like i'm 20 years
into developing and this is the first time a big passion project i've gotten everything else was
just like okay where's my paycheck i'll just do what you say and i didn't care but this is like so
awesome yes it is fascinating um uh bubble god uh you got a question um i guess one can see like
cheer lisp as like the operating system programming language that you never actually touch and everything
on top is sort of what you actually use and creates yeah that's actually a great uh analogy um yeah and in
fact as an as a developer think think about how often you write apps versus um submit pull requests
for the linux kernel right like that's that's probably about the the amount of time that you'd
spend writing chia list versus just normal code probably yeah exactly but yeah chia is like um once you
see everything you can do as chia you just like you can't fall asleep anymore like so many ideas just
going through your brain but it takes i haven't been able to sleep in a week because all of a sudden
like a bunch of different concepts i've been thinking about separately just kind of all came
together at once and it's kind of why you've seen a big development push like i've kind of pushed a lot
of stuff in just a couple days because a lot of ideas just kind of all came together at one time
i have to admit like um uh you know i visit a lot of spaces and um i come out of those spaces and i
just think how how far behind are these people you know the people obviously there's some fantastic
developers on ethereum polygonic etc etc um but i just come out of them and i think they the little do
they know you know you know how far this is going on cheer and um you know i think if i think if they
spent um you know a couple of days just looking into it they'd sort of realize this this is sort of
it seems to me that it's so far ahead of everything else that i can't quite comprehend it i mean
is that do you see that like that michael
i i gotta i gotta be honest i'm not i'm not very tribal when it comes to you know the different
coins and and you know i see i i still see pros and cons on on on different things now now chia has
definitely caught my attention and i that i mean obviously the fact that on that i'm doing all my work
going chia right now you know says something but but i think i think what what what what's happening
is every you know every each you know a coin has their own set of challenges and and once you become
like an expert like you know it doesn't matter what the state of the coin is if if you're somebody who
can make the coin better you know you know you're putting your energy into doing that so you know let's
say like tomorrow like i don't know like some some other coin like just completely like was so
futuristic that even chia couldn't couldn't like tread water against it like i'd probably still be in
chia and i i just want to improve chia uh be uh and maybe and try to get it you know to bring it up uh
you know just as good so i i imagine that these other developers they sunk a lot of time and they
understand their coin a lot and this is very niche knowledge that it's i think it's hard to uh to
to really pick up um and and once once you get in like i think it's like okay like if if let's say
she is you know even if if i was a theory of developer and i i agreed that she was better
but i already had my foot in the door of the theory i might just want to try to
continue improving ethereum to try to compete with chia i mean that's the way i see it that makes
sense yeah yeah completely completely agree and um and i think also there's um there's like a huge
financial um aspect to that as well as in the the amount of money that people have put into developing
on uh ethereum predominantly um is a difficult thing to to move away from as well but yeah it just it just
it seems like the tech is you know the technology and and certainly the the fundamentals of the the
chia blockchain just seem to tick a lot of boxes for a lot of people yeah i i like the first time
that like so so like the first the first year i was on chia it was more of i wanted like i i had i've
been dabbling with ethereum for a long time from a developer standpoint like from a user standpoint i was
heavily involved but from a developer standpoint i was dabbling into it and i really wanted to get
my foot in the door with blockchain uh buddy of mine had a contact he's like i got blockchain uh
project you want to join and i was like i was like who is it like chia i was like who are they
uh and i was like okay i'll do it i don't care but like that first year like i didn't really get it
all right uh but then like after i i kind of realized that um you know what the thing that i built was
actually really awesome i was like i'm gonna go ahead and i'm i'm gonna actually go in and learn
it and and i i read about chia lisp i read about the coin set model and i instantly was sold
i was like i was like it was like a brain explosion i was like holy shit like
all the stuff that i've been complaining about ethereum this is all fixed right here
um they went they went they went the only thing is how do we the only thing i've been wanting to do
is how do we build the same applications in in chia and i think i've had some pushback from like gene
and stuff where like well we're not trying to be the same you know we're gonna build our own path but
you know there's still some base uh you know you know base uh uh uh applications i think
are drivers of growth and and tibet swap is actually what like in line with what i was saying
like before before tibet swap i was always kind of like i hope one day we can build something like
tibet swap but until we do like you know i'm kind of like i'm still kind of a a a two coin guy
uh but then tibet swap came in i was like yes yes he showed that it's possible like now we can do
everything and then i was like okay 100 like now i'm convinced if we can do in chia and she is better
than all the other aspects like this is a really awesome coin
here yeah yeah it certainly is and um you know i think going back what you what you guys were saying
there about the sort of all the primitives are are there i mean certainly from the nft point of view
it's just you know there's no there's no contract writing it's just sort of almost plug and play but
you've got this this sort of fantastic security and um you know this peer-to-peer nature of the
blockchain it's just it's an incredible it's an incredible tool to use and uh and i'm really i'm
really pleased that someone like yourself is is you know not only building on this data layer but
but able to articulate it in your tweets and sort of coming into these spaces and talk about it because
it's it's something that i think is a little bit hidden away really that certainly isn't talked about
a lot when it comes down to people talking about cheer it's a lot of it is you know it's about the peer-to-peer
aspect and the decentralization but but i think so i've seen a lot of people interested data layer
but they see it as this like magical or shiny unicorn that they're afraid to touch and that's what i want
to change i couldn't agree more yeah i couldn't agree more so so like for example like the the the
sprout tool uh you know i i really built that for deploying apps for myself but i i think that could
be a boom for for nft creators like it just you haven't like there's no excuse now not to put
nft your nfts on on the data layer now you that might not be your main uh consumption method yet
but it takes 10 minutes and you can say hey my nft is now on data like i mean so like i feel like
like i i feel like that tool uh can be huge uh for ta nfts yeah i'm definitely gonna try your library
or your call it the mpm package and just see what i can do with it because before like i sort of knew
about data layer and stuff but i i briefly looked at it and it looked a bit hard to like set up was
not that much to read about it but now i'm definitely going to give it a try just to have
tried out yeah uh like i said like uh it takes it takes 10 minutes to push a for a folder of files
which could be a folder of jpegs to to your data layer um you just run a couple commands against the
right folder and and then you wait and then you're done like so you don't even actually need to know
anything about data layer other than the fact that you're pushing it and you you can now say okay
like if you want to subscribe to our nft collection on data layer here's the store id and later
we're going to find a way to map you know uh um you know the the the xch uh uh name uh nfts
to data layer singleton so then you don't even need to pass your your your data your data
layer idea anymore now you just go to i don't know to bet swap to xch on uh on the right browser
you can also create like a link tree dot xch you know where you have your links to certain stuff maybe
yeah absolutely
fantastic stuff uh track you jumped up as well you got any any thoughts about all of this i don't
i don't know if you're still there god yes there's so many my head's cooking here um i had questions
but you guys kind of answered it along the way i was curious about subscription model uh the user
permission uh store and i was thinking about if it's if it's immutable you know how do we do upgrade
downgrade is that even going to be possible sometimes people change their own account level
access say based on needs if you're providing a service in some sort of sas product so but you guys
danced around that in a whole bunch of ways and i think there's still unknowns and development there so
kind of don't have a question there but my head's spinning
you have a simple i do have a simple uh pattern in terms of upgradability um so
and i wrote about this in the medium articles if you if you uh are familiar that i had some
medium articles out there on data layer um so a data layer store is just a key value
store and uh and there's nothing stopping uh the value of one of your keys to be the store
store id to another store so uh we'll use the example of a cad t um
every but like when as soon as you open up the cad t you create what's an organization uh so you
create an organization store think of it as your profile store the uh it has you know your name your
icon uh a couple other pieces of metadata but then it has a a key in there called registry id
and that uh that key maps to where all the carbon credit information is if you want to upgrade
to a different store or or say you somehow uh you want to deprecate the old store
all you do is you swap out the id that's referenced in the in in the top level store
so so as long as your top level store may remains the same you can actually create a hierarchy of
stores that you know it's just pointers to the story and the pointer changes uh so that way uh you can
actually change the store that's served by the same store id even though it's a different store uh you
can change as many times as you want and you know it'll just automatically get synced down
there you have it
so let me uh i don't know michael i want to also wanted to get your thoughts on uh what i'm calling
chaps uh i don't know if you've seen any of my tweets about that but i actually i have seen your
tweets about i'm convinced that yeah that if you say it enough times everybody is just going to
it's just going to be part of every day everybody's language and i think that's what your goal is
of course that's what that was the and i even i have uh ch.app domain so that's that's ready to go
uh i've got uh but anyway the the idea here is that so just so in case not everybody's uh familiar
with what we're talking about here the the the idea of chaps is that they're chia apps um they're
or daps as we have called them decentralized apps um but in my kind of web3 vision everything should
be running on localhost right we shouldn't actually be just like michael has been saying with data layer
we shouldn't be necessarily browsing out to third party services or sites you actually should be
looking at everything local first on the blockchain if you can that you've already verified all the
machinery of the blockchain has already verified for you and the consensus model has already you know
sinked down to your machine and if you could look at all of that locally for 90 of your work or 100
ideally then not only would it be secure but it would be extremely fast right you don't you're not
having to reach out across the internet anymore because it's already all mostly there on your machine
um so in my in my view the the view the the future of web3 apps are these these daps or these
things that i like to call chaps because they're actually dependent on the chia blockchain um but the
idea is so imagine a world if you will uh it's it's almost like a little companion app is how i like to
think of so normal you have a normal chia desktop wallet running on your pc or mac but then this would also
this app would run right alongside it and it would be very similar when you open it up for the first
time it would look similar to maybe like an app store he'd show you lots of different apps that
you could download or chaps and these are all represented as as nfts by the way so the creators
of chaps will distribute them as nfts um if you and in once you connect it to your wallet it's actually
going to show you here the the chaps that you own and maybe you got those for free from the creator
maybe you bought them maybe you traded them on dexie whatever but here's the chaps that you own
you double click one uh it opens up that chap and this is just an assembly uh in dot net terms or in
cstart terms that gets loaded up dynamically all pulled from the normal machinery of the nfts and
verified hash hashes and everything that way so um the idea though from an app developer's perspective
is you don't have to worry about anything anymore except your actual ui and your logic
um chaps by the way the the shell itself is written in dotnet maui a multi-application user
interface multi-device so it runs like i said on windows uh mac ios uh android and even tizen i guess
which is like a tv i don't i don't know but uh it runs on lots of different platforms it's all open
source it's microsoft's uh latest dot net stuff um but it uses web assembly uh which is cool so
even in the browser this this could work in the browser with limited permissions or you could have
download a full chat shell that would run chaps that would have full local permission to talk to
the full node directly data layer directly uh in between abstract databases uh staging databases
directly um all of this fun stuff that we've talked about and you have a app developers now
have a complete framework very much like apple's app store where you don't have to worry about
distribution anymore you don't have to worry about how do you get your users to trust to download your
github repo uh release you know or whatever or talk them through that they just it shows up like a
normal icon and like in an app store and everything just from their perspective is is at least the current
state of the art but they're used to in other app stores right um but we are using all the fun stuff
behind the scenes so anyway that's the that's the pitch you just gave me an awesome thought and maybe
this is already what your thought was um you know what if uh one of the nft itself has a verifiable hash
hash so that way you know when you pull you you know and also a pointer to the application so
that way you know you're pulling the right with the right application um oh of course yeah that would
that's that's that's why i love the so just like github where you can do sign you can actually sign your
code with with code certificates and things this is how we do it now but that's exactly right by using nfts we
don't have to do code signing anymore like we did in the old days because we're just we're just
piggybacking on all that machinery now so you have to explain this more josh because i'm literally
sitting here doing code signing on an ios app ready to slit my my ring so explain this how does this work
well code signing is all about just so you've created a binary package of some sort or any sort of
you know distribution and now you're going to make a hash of that and then you're going to use your
public key to sign that hash and say okay this this is you're going to get this binary and if you
decrypt my public you decrypt my uh little thing here the hash with my public key then you'll know
that and then you hash what you get and you end up with the same hash then you know that what you get
is exactly what i'm telling you that i've signed that you should get if that makes sense so um that's
that's that's the normal state the current day code signing certificate so that that also code
signing or certificates in general ssl search and everything our course are ultimately centralized
right they they depend on a chain but but at the end of the chain you're trusting one of five or six
major uh companies that create these certificates you know so anyway in this world in the new world
you would just by packaging your app whatever your binary whatever the output of your app is whether
it's a binary whether it's html and javascript uh zipped up together in a single zip file or it points
to a single spa you know uh deployment of some sort whatever it is you can hash that and that's going
to be the hash of your nft already and that's already protected by all of the chia blockchain vids and
everything so now we we don't have to do code signing anymore because that's what we're already doing with nfts
right so it's all the same stuff
hopefully i think that made sense yeah thanks josh what what what is what is the the
when you build your chat what what what what's the build output look like like is it like some sort
of uh what i guess when okay let me let me rephrase my question like the thing that was super cool
about the web 2 gateway i built was the ability to run my app directly from the data layer just by
going to a url like it's literally like it doesn't even copy any like anything out of the data layer
it's literally running directly from data layer like can we accomplish the same thing with with chaps
of course and in fact how you would so in the chat world you would you could actually i've got a few
different concepts for plugin types so you could have a full app plugin that is that's got a user
interface and obviously lots of code behind it um that will all be packaged up and deployed as a
uh i believe they're ms msix's or something like that for windows uh there it's different for platform
of course right like the in the the physical file there's actually uh the build process outputs several
files depending on the platforms that you've selected to target um but regardless all of that can be
packaged up into a single nft manifest file at the very least so the the nft might point to a uh
uh in fact the metadata the json file for the nft itself might have a big list of files that it needs
to go download from somewhere those were chip 15 right to your exactly there you go um and this this
is uh so chip 15 describes a bunch of different metadata types uh but exactly so so your your app nft
could have custom metadata in it maybe in fact your app nft could still just be an image it could be the
icon of the app right but the nft metadata could have a bunch of uh links to resource files and
their associated hashes um that uh that you can go find based on the platform as well then really
some platform exactly yep and and this these links excuse me that could be you could go find them
either directly at your data data layer maybe there's a special format we use like you you were
talking about uh custom protocol handlers maybe there's that or you go through your web gateway
right you just uh you point them to normal urls but behind the scenes that's actually your web
gateway writing now your web gateway itself could be a chap in my in my model that people download and
run locally because your web gateway correct me if i'm wrong but uh it's running directly from from data
layer but you have to have this gateway itself as a process running somewhere right whether that's a server
it's a companion app and uh i want to admit i want to run it as a service on the machine like in the
background and i want to register a custom protocol called chia slash slash so basic so if your custom protocol is
set up anytime you go to chia slash slash and the the store id it will just reroute through
uh through uh through the process open up the browser and and uh and uh in the app okay brilliant
okay so check this out you could create a chat called uh chia data layer gateway or image gateway or
nft whatever you want to call it gateway um the user downloads you know find your nft somehow whether you
distribute them for free or or sell them on the app on the dexy or whatever they get your nft your your
your gateway your web2 gateway um app or chap installed into their chap shell and because
that's running natively by the way uh so on the web it would be running under web assembly but because
it's running on a local device it's going to be running in full native you're going to have full
user permissions on the local machine so this is where we can talk to the local full node
um we can uh on on both mac or pc uh i don't know if they support uh ubuntu yet but regardless that's
that's definitely one of their targets but um the point is you could you could use the chap framework
to deploy this web2 service as a chap and then people could install it uh update it etc but it would
be running you'd have full access to local machines so the first time it starts up it could create a service
you know on the in the windows service in windows or a mac demon running on mac you know whatever it
needs to do on on different platforms um to do what you want it to do you can install reverse uh
reverse proxy servers you could do you know whatever the user has permission to do of course on their on
their device um now you you know that may not work on ios of course you know so so the ios version of
your web gate may be may simply be a uh redirector or you know maybe it redirects to somewhere else in your
network or something um so that if you're on the ios version of chap then and another app wants to
open something from your web 2 gateway it knows to reach over your network and get it from the full
mode not on your phone of course but on your pc or your server or whatever you've got so this is the
really cool stuff i think this and and the point is this is all this is all solved in the enterprise
world as far as like deploying to different platforms and and deploying updates and app stores and all
the stuff this is the state of the art that users expect we've got to have at least that before we
can even consider ourselves you know competent in this world of of apps so that's like the very basic
uh what are the table stakes as as gene likes to say and others in my mind um so that's where we have
to get first and foremost and then we can build from there of course for all kinds of fun stuff
so i have a i have a related but different question that i want to ask it's a heavy debt question but i
feel like i have some smart devs on the phone so i want to ask out loud i'm trying to i have this idea
on my mind uh in terms of how to solve the issue of uh the the certificates um communicating with uh chio rpcs
uh so so so right now our common pattern is uh uh with the with with all the the chia data layer work
with even with the enterprise customers is we build a front end and the front end can either be served
an electron or it can be served um or or it can be served in uh on the browser but and it communicates
with with a localhost back end um and that back end is really what what calls the actual rpcs
hand you know loads and certificates and so on um i would love to be able to build self like self
isolated like completely sandboxed apps that like you don't need it could be a front end and the back
end you don't need to run the back end somewhere else uh in order in order to use it and for that to
be possible we need to be able to call chio rpcs from the browser uh i mean it won't cover 100 of our
use cases but it will get like 80 there like to build a lot of cool stuff so my question is
from a security standpoint let's say you installed my my web2 gateway as as a uh a service
so it's running on localhost and uh and you and you load uh you uh load a a chia app from the from
the app store through the web gateway
if the web2 gateway included a base 64 of the certificate and the key file in the headers for
the localhost app to use to communicate with the localhost chia node or a remote chia node
is that a security security concern whereas like if i if i go through all the effort of making that
work and is will will i basically like get shunned from security experts or is that actually a feasible way
to to uh to make this connection and and make the the applications that we we deliver over uh data
way or more powerful uh unfortunately no so you so i i actually have spent the last week banging my
head against this exact issue so it's funny you you bring us up um no and and the reason if once you
once you realize it is is pretty obvious in a browser the um the javascript uh sandbox itself is
what is in full control of whether javascript of any sort can interact with localhost and all of
them you you do actually have uh permissions to talk to localhost in in a lot of scenarios but when
you get ssl involved especially a custom certificate what you have to be able to do when you see this
in your own code i'm sure and and in other open source projects out there of all different sorts
node.js and and dot net and everything you see where they um they load the custom certificates from
the full node because when you install your full node by the way your your chia wallet and full node
it generates unique uh ssl certificates for you that are called self-signed certificates that only
exist on your machine and this secures the electron app the main front-end ui that you use secures it
talking to the back-end uh chia process that's running and doing all of its fun work so that's what we're
talking about here is when you write custom apps that we're talking about you have to use the same ssl
sorts that were generated uniquely to that user when you try to talk to the rpc and this is a security
measure um and then right now in all the software we've written the actual call that is being made
in the in the companion uh back-end app that's running on localhost but it's using super agent
which is like the same library that you can run on the browser and there are there are options to put
certificate and key files and is part of the request exactly but in that back-end uh uh process that's
running you are you are in the full you're that process is actually running on the local machine
already right so you've already deployed that process you are in full control of that when you're
in the browser though the browser sandbox itself will say sorry you specifically there's the the one
thing that we're talking about here is the ability to tell the ssl uh or tell the http request to
ignore ssl errors because in the local environment when you're doing your own environment you have full
control over it you're you normally the javascript browser uh sandbox would not let you will not let
you in any circumstance say hey by the way you can ignore the certificate errors i'm going to call this
website but but but don't worry about it you can ignore the certificate errors and still do it
of course that makes sense you should not from javascript be able to in any way tell
the entire sandbox to ignore the main thing that keeps us safe right like so that's why from the
sandbox you can't do that but when you're running as a local full permission to app you can and that's
this is how that'll work you can actually tell the the request to say hey it's okay that this is a
self-signed search you don't have to go fully look up the trust chain and verify it uh just make sure
the basic security works for it um and and continue so okay that's the difference so the super agent
library as an example has the ability to uh to to ignore uh uh it has the ability to ignore verification
of the certificate but you're saying the browser itself exactly if you ran that same library in
the browser then you would get an error as soon as you tried to execute that the browser you'll get
a console error that says hey that's not allowed you can't you can't tell me to ignore the ssl cert
okay let me let me follow that up with this with the fallback that i thought of i was like okay
can i do this but then i was like is this a security concern or not
what if what if i proxied the request back through the web gateway and the web gateway is
what actually adds certificates which you're essentially doing absolutely fine um but well
what what you're essentially doing though is you're you're you're making it so the certificates are no
longer needed to communicate with the qrpcs and now even though you can bind the web2 gateway to
only accept requests from localhosts and uh and you can um and your your app is running on localhost
in theory you should be okay but the fact that i'm basically disabling the wallet security is that a
bad thing yes because now javascript can talk to your local wallet without so you again think of it this
way we do not want under any circumstance just random javascript out there that somebody can put in
a web page that my browser downloads to be able to call to be able to tell my local host to stop
don't worry about certificates that aren't valid and go ahead and call the uh chia rpc on this known port
and give me the seed seed phrase right like obviously that's the that's what you could do if so what
you're doing is you're basically saying you're you're stripping away that that layer and obviously this
wouldn't be a huge thing because not everybody would know that you're that they're running your
specific little thing that's doing this but theoretically now javascript could say okay go
to localhost port whatever that i know michael set up and ask for the the seed phrase because i know
that that thing is going to proxy the correct uh ssl things for me and then it would work so yeah that
that would be a huge security opening if people know about it what if you found a way okay what if the
one of the one of the web 2 gateway signed a jwt token and provided it to the localhost app
and local so that way basically the every request will include this jwt token that has to be
verified when when a request comes from the localhost app that way you know that it's the request is
coming from a data layer app not from any other random web page and only accepts it only
accepts requests from data layer apps that were served through the web 2 gateway
here's here's where i come back to so we're doing all of this because you want to try to avoid
you want to some you want to be able to have somebody do it through a browser right i want this
i want the i want the app to be as self-contained as possible without having to run a hundred auxiliary
apps in the background because that is a problem that we have right now with cad t is it's we have
like so many auxiliary apps that all have to run at the same time and all their purposes are well really
just to serve as a way to make rpc calls so so again going back to so my point is though that you are
you've got lots of things running and you're saying let's solve let's instead of seven things
running let's have one thing running that does this stuff and and proxies and does all of that
naturally that makes sense in a closed environment where you know maybe you you have a lot of control
over the environment but not for like obviously publicly distributed apps and things but that's
where this chat framework comes in again because it would handle that all that for you right so it's
already figured out it's already been configured and figured out correctly to talk to local rpc
your app now that you're building assumes that it can talk to rpc regardless of where or how it's
running because that's part of the plugin framework context that chap is providing to it so it doesn't
have to worry about any of that anymore it just talks to rpc or any of dexy or any other services that
are provided to it in uh uh in an injection uh dependency injection by the way which is really fun
uh so all of that just happens the app doesn't have to worry about that anymore at all is the
point it's already taken care of you in one spot by cat when it was installed i would love to see i
would actually i would really love to see a proof of concept of this working because i think it solves a
lot of problems not just like in the community but also with what she is building internally
yep yep i've been uh really i'm having been a lot uh i've been having a lot of fun building it out
and doing proof of concepts myself i'm trying to find before i i publish anything or make you know
get people too much more excited i'm trying to find showstopper type stuff so for example one of the
first things that this what we just talked about uh uh browser you know the javascript sandbox and why
that's so important that it works that way and why we not only can't get around it but shouldn't
right like that's the fun i understand why it works that way i understand why my question was
is this work around is this work around like a bastardization or is it actually a legit solution
that was my question it sounds it sounds like you work around with a bastardization yeah my well just my
my um i guess my opinion would be that you are you're trading off one security mechanism which
is the built-in ssl at the very least for your own custom kind of security with a jwt token and all
of that which has always been a bit of a red flag for me like i i hate trying to build my own security
anything i would much rather try to figure out how to work around that's why i wanted to ask it publicly
instead of my own headcanning yeah yeah yeah well and it's funny that you mentioned it because like
i said it just so happened to be what i was like very deep in this last week because i was trying
to do that so because i want to work with web assembly right so web assembly is this concept of
of getting a lot more beautiful graphical apps and all kinds of fun frameworks that we can use in the
browser um but not just javascript uh but still compiled but again still governed by the same
javascript framework that everything else is so my last week was trying to prove to myself
absolutely that i could not get a web assembly app spa web assembly app running that would talk to the
local uh the local rpc i really wanted that to work again even though i knew it shouldn't and it couldn't
because it was all the same stuff behind the scenes and i verified that no that in fact does not work so
my my current thinking is actually it's a it's like the progressive of the pwa uh model where you
could theoretically you could go to the chap webpage ch.app and see a web representation of of how the
chap looks when it will download to your desktop um but maybe there will be missing like you know if
the chap only needs dexy to work then it will work in the browser just fine but if it needs to work with
data layer or anything local with rpc then it'll say hey you need to actually download this to your
local machine uh to run it and give it some more permissions and at that point you will be you know
running it with more permissions okay so okay i i i think a competing idea what you know would be
building a uh a browser uh a app loader from data layer in electron and then the electron handles the
the uh the rpc calls and then then any any data layer thing that you load into the electron browser
uh could actually talk directly to to to data layer same exact thing yep so i looked into
so josh let me just i hate i hate to put in because um uh it's absolutely fascinating but
we've got about half an hour left and and i've not seen birdman keep his hand up for this long without
shouting pipe down for well ever since i've made it i've seen i've actually seen it and i was actually
i have a stopwatch going here to see how long you can hold it because i i'm i'm fascinated by this too myself
i'm enjoying this conversation keep going guys like i'll i'll keep it up
i think what i wanted was a public chat conversation like for the longest time
it is fascinating but um i'm gonna wrap things up shortly and i want to i want to try and get to
andy because he has well patiently been waiting for about two hours uh to talk about something he's
been working on so fair enough fair enough let me let me see the floor
but it's it's absolutely fascinating and um and i'm gonna be pretty honest with you i'm not a dev
and i'd say most of the last half an hour i've caught about i don't know 10 of what you're talking about
but um it's you know it's fascinating to hear that you guys are on the same page and you're you're sort
of working to pretty much the same sort of conclusions which is brilliant so um let me just jump
to andy very quickly are you still there andy by the way um yeah i'm still here can you hear me
yeah we can we can just about hear you but yeah i don't know i don't quite know how you're gonna um
follow that up to be fair well yeah just sorry andy yeah real quick real quick i'm sorry i'm sorry i just
wanted to say you know i'm having a lovely sunday afternoon you know all is right in the world
sitting here sipping my whiskey smoking my cigarettes making beef wellington tonight for
the family it's gonna be great just very very happy to be part of the community
oh god oh god hold on guys there's somebody at the door here
oh god oh god they found me they found me oh yeah i'm coming i'm coming
okay don't worry okay all right whoa whoa oh my goodness they communicate tens of thousands of
new findings to each other every year oh my god oh man i'm getting out of here
i'm getting the hell out of here this is crazy
oh my god
i'm running i'm running that is one way to light in the mood
after me what am i supposed oh god here they come here they come i'm hiding in a bush
what are you guys told on me
i'll see you guys later i love you
what do you reckon to that michael did you reckon we could put that on data layer
i didn't know what just happened
well uh yeah bird man's on a different level i think he's just been um he's been caught by the
police for being uh being um a cheer a cheer troll
oh dear well done bird man um well maybe we'll see you in the churches here soon depends if you can
get out of jail or not i suppose but andy come on let's let's go back to some more serious talk
jesus that man honest to god okay uh so i have a token it's called uh u.s
uh it's uh it's as simple as ngbt lp the lp stands for level targeting um and it's um it
seems to be indexed to u.s uh nominal gross domestic product well actually it doesn't really matter if
you say nominal or not but uh the idea is that it uh it should be um it's like a standard of value
that's based on actual production that's been based on based on you know what the fed decides to do
some particular time
and uh
so i don't know i'm in a big trouble
i know yeah no it's fine um because i've seen i've seen uh some posts of yours and i think if
people will get better understanding if they're going off quickly for your timeline um did you say
you've you've you've you've put them on uh uh dixie and tibet swap did i see that yeah so it's on
dixie and tibet swap um what i do is like every quarter when the gdp report comes out i i put an
office on on dixie to um to peg the values to you know whatever the latest reported gdp or the
that's actually the first the advanced gdp report um and uh the the token is uh is that it's indexed
gdp minus five percent of the year so it's a calling tag um you know kind of like uh they used i don't
know if they could do that but they're going to be uh trying to see i used to pay for the dollar
with the phone but but um and and right now you know in the in the long run i hope to have uh you
know smart coins and oracles and stuff involved at this point it's just me and you have to trust me to
do the quarterly uh tags um this also traded on tibet swap and i i you know sometimes i put in
and sometimes i i make transactions into that stuff to make a price so closer to what it should be
um and originally so i mean usdp is you know recorded in dollars originally i was using um
uh that uh that's uh official stable plane um stable uh usds um as uh as a proxy sorted for dollars so
we should uh link it to gdp um exactly um but since uh since uh that's uh stable plane doesn't really
work um anymore so now uh and what i'm doing is i'm taking the the prices here in dollars uh from uh
what's that website there uh uh the lizard uh uh ghetto uh plain ghetto
um and i just use i express the gdp in terms of sia using the quote that is most recent at the time
in the gdp report comes out um so the idea the idea is it's supposed to be sort of a stable point but
that's more in the sense more stable than sia um in as much as it takes you something real um how much
do you as actually producers uh that have been being pegged to you know dollars or whatever which
you just um you know you can't um you don't know what can happen you don't know when there's going
to be more inflation than expected or more growth than expected or sometimes there's going to be less
inflation than expected or less growth than expected so um that's the idea i also have an inverse gdp
coin which you can use to like uh speculate against uh if you if you think that uh you think gdp is
going to come in weaker than expected or inflation is going to come in weaker than expected um you can
you can speculate on that using the inverse coin which is designed to design so that if you would be the
the price of the price of the inverse coin plus the price of this direct coin would be you add up to
two dollars uh based on whatever whatever key you're selling at at the time when you're paid off this
you know yeah it's it sounds like you're you're right at the beginning of this journey as well um yeah i'm in the very
early stages i think yeah yeah yeah um so you know essentially you could be using like you say
some sort of smart contracts smart coins to uh to keep track of all this as well
ideally what i would like to do is like be able to commit to to do offers getting created
um but you know ideally you want something that's going to be able to commit to and say um when the
report comes out this is going to get the number from the oracle it's going to automatically create
these offers and and there's you know that's committed to an event um
that's right you know that's that that's a very long way ahead so right now it's just me
no i think that i think it's i think it's a great idea actually um i know josh has been playing
around with tibet swap and uh tokens and and um sort of the way the way it sort of operates i mean i'm
i'm no expert on finances at all um lucas has got a pretty good idea of a finance a financial world as
well so um but yeah i think it's something we need to dive more into unfortunately i think because
because of the the data lag or it's something we could probably talk about for a couple of hours
on this subject alone i would think um which is a bit of a shame because um
we're sort of running out of time but um so i've pinned i've pinned your tweet at the top there
um i think probably i mean if it's all right with you i know we've been sat here for two hours waiting
for you but i'd like to i'd like to do this again maybe next week or something um i'm not sure
maybe next week or the week after yeah yeah i'd like to i'd like to spend a bit more time and and
dive a bit deeper into it because um it sounds like something that's probably a pretty good idea i
mean lucas uh you know maybe you'd um take it lucas is is certainly my go-to when it comes to
financial sort of aspects of of crypto and and the world but um i'd like to get your opinion lucas
you know maybe if you've had a bit more time to look at it and and and we could dive into it a bit
deeper that'd be that'd be really good um yeah god i'm definitely down for that
i i um i was a little bit distracted there because uh well i got to cooking while while
they those devs were a little the y'all got a little dense for me josh but that was fun
um and then i've been so concerned about about birdman um but yeah uh we could uh we i you know
i love talking about all aspects of finance so um i'll i'll be here next week
um okay i i i'm not sure i'll be able to do this week so maybe at least a week after
yeah definitely let's um let's pencil something in it's uh yeah it's a shame we couldn't talk more
but i think i think if um if if certainly if lucas can have a good look at at what you're doing and i'll
have a look a bit deeper as well um let's try and pencil something in for a couple of weeks time
and let's let's dedicate a whole a whole show to it so that um you know we can get a lot of people
interested well i think i'm sure we can talk about it because i think it's i think it's a good idea
and i think um i think the more um sort of more cap tokens on cheer that are doing something useful
that's not to say i'm not a fan of the shit coins don't get me wrong but i think i think that's
that's that's something that's quite important so uh let's uh let's certainly i'd love to dig a bit
deeper into it and try and try and sort of work out exactly exactly what your thoughts are on your
thought process because it's uh you know it's important that that that some some actual thinking
goes into cat tokens as well uh in my opinion anyway but uh i don't know what the lake is what
what are the latest shit goings out lucas you must be all over this isn't it chump coin or something
well uh the cbdc is i wouldn't call um i mean that one seems to be a very like a political a new a
new era um there seems to be an oa something i i've seen over on clone on uh on cookies space uh that
seems to be some sort of response to uh pulse but it's actually pretty quiet
um bonk seems to be trying to rise again but i you know i like the idea of memetic tokens but
i'm not an art like i don't uh i don't speculate on them put it that way yeah cool
well maybe um maybe uh let's get to an island
yes it's that that sounds like a plan sounds like a plan um we'll set up a dm group with us three in it and uh
uh anyone else who wants to wants to get involved and um we'll we'll dig a bit deeper in there so i've got a better understanding and at least i'll i'll know what sort of questions to probe with as well um yeah that sounds like a plan sounds like a good idea to me um
um yeah like i say yeah and also we need to try and find we need to get you a microphone that's a bit
louder than the one you've got you're quite quiet when you listen back you'll um you'll realize that you're coming through quite quiet
yeah maybe that yeah well because i'm just using the microphone on my phone
maybe maybe although you might need to put it right next to your mouth or something i don't know
i don't know but no that's cool that's cool it's uh it's been a pleasure well it's a pleasure to meet you
anyway and uh and here you come up and talk and um you know it's one of these things that coming up and
talking in spaces can be difficult as well so you know especially if you've got something that you
want to talk about and people don't understand it so um yeah and i appreciate you coming up and
appreciate you waiting for two hours although to be fair the um the data layer chat uh chat has been
pretty um it's been really interesting actually it's uh it's one of those chats that i couldn't really
jump in and stop and i didn't want to i didn't want to sort of like wedge you in in the middle of
it to get forgotten about so um i would want to try and get it finished
yeah oh we'll start the dm group we'll start the dm group and um uh and then let's pencil something
in for a couple of weeks time and um we can do a bit of advertising for it and uh we'll get some
we'll get some more people in but um yeah so i'm gonna i'm gonna i'm gonna wrap things up very
shortly is there anyone else on stage that wants to say anything jay ron you've just jumped up i'm
sure you're um you're an expert in the financial world uh far from it yes uh hello everyone happy
sunday happy forward monday and tuesday and wednesday um tim enjoy your time away wherever you go
be safe um and yeah you know we started the cheer hub um i keep on uh sharing that information and i'd
like other if anyone wants to add their project nft or otherwise uh from the cheer group community
whatever dm me and i'll add it in um we've got the education part or the learning about cheer so if
someone's got interesting relevant blogs i'm happy to add that in there as well and um next saturday
i'm going to be uh on a another space doing a deep dive with sort of an ama about uh the mergals and
the wandering sailors so i'll i'll have that info up on wednesday and hopefully uh we can introduce
those people to cheer it is going to be polygon centric but um you know yeah just sow the seed
of cheer anyway thank you very much no worries at all no worries at all i think track did you run
uh did you want to have any final words
i thought i saw your hand pop up maybe you uh maybe you just uh maybe you flinched
no i was here but then i had to yell at my son for a minute because he was being annoying
and now i forget what i was gonna oh it was i was gonna say the data layer uh josh i was working
with michael earlier today getting my sba going and back and forth with him i'd love to get a space going
later um where we talk further on the stuff we were talking about because i'm salivating over all
of it yeah yeah totally um yep we can talk more about uh what i've found as well i've been doing
some r&d this last week or so so yep yeah get another get another space going after this one if you
want i've just got to uh i've got to jump down and uh see my wife before uh before she tells me off for
being on the on the phone for three hours but um yeah i just wanted to um just point out the pin
pin tweets at the top there the first one from uh quirky he was up here earlier uh he's he's one of
the original um nft collections on cheer and he is um very close to releasing something quite special i
think and um as a bit of a bit of alpha i know that we've created something for him as well and
i think he's got he's pinned up to the top there the first pin um that there's there's only going to
be 50 in this collection so certainly i would pick up some of these nfts anyway before um before the
rest of the world realizes what's going on because i think some of these original collections are going to
be um worth having in your wallets uh i would say that and then a couple other posts up there at the
top there from michael taylor again if you're not following michael i tweeted this the other day if
you're not following michael then you should be um because you know he's one of these um these people
have that has this very vast knowledge and a good grasp of what data layer can do and um you know i i've
certainly knew that what data layer is capable of but he's blowing my mind and i'm sure like like
drugatis has just said it's blowing my mind to how far it can go um you know us my my thinking of
data layer was how you can get multiple gaming people to look at the same piece of data that's
in a decentralized way but you know that the sort of applications he's going to be able to create
and we're going to be able to create on top of this he's going way beyond anything i could i could
think of so uh yeah definitely follow michael taylor uh follow andy as well and like i say um he's i've
pinned up the top there uh one of his tweets but give him a follow and let's um we're going to try
and get something with him in a couple of weeks time so we can actually dive a bit deeper into that
because i think it's worth it's worth investigating something that's uh that's worth looking at and jay
ron has sneakily put up there look up there uh his chia hub which is actually a pretty cool little
website um uh again people down below if you're if you've got an nft project or any other project
reach out to jay ron and he'll just add you to there um and it's like a quick way for people to
go in have a look at some sort of trusted uh projects that are being built on chia a little bit
about some of the uh i guess the architecture of chia and and the learning learning zone so
that's well worth a look at and and i like i like the way that that you've done that jay ron where
it's simple you know go down have a look click on a button and you can look a bit more it's it's like
a one-stop shop rather than rather than you know nothing too complicated where you've got to go and
start looking through different tabs and um you know going through different projects you know
it's it's a one-stop shop so yeah good work jay ron good work but yeah um yeah so that that's it for
today uh usually we only do it for an hour or a couple of hours so we're we've done two and a half
hours today um hopefully it's not too much that people won't listen back because i think certainly the
first well the first couple of hours the first hour or so is a good insight to what what data
layer is going to be all about i think the second hour got very very technical which is great and
you know i think some of the more technical people will get a lot out of that it's it's very difficult
to um talk about these technical things without going too deep into it and uh you know it was like
being a fly on the wall to listen listen to josh and uh and michael talk to stuff which was you know
it's fascinating in itself and uh sometimes it's a bit of a different language but you know you you get
the gist of it and um yeah again if you're down below and you you're not uh not sort of in the cheer
ecosystem i can see a few people down there that's that's uh maybe intrigued um you know follow a few
people in this room you know if you've got the little seedling next to their name then they're
worth a follow and uh you'll you'll you'll start to learn a little bit more about what the cheer
blockchain is and you know what it's going to do and what it's capable of so um it's well worth
looking into but um yeah final call from anybody up here on the stage uh appreciate you all coming up
josh especially for for handling some of that technical talk because i certainly wouldn't be able to keep
up with any of that and um and lucas as always uh drac j ron andy appreciate your patience for the
last the last couple of hours and i think i think we'll get something out of that and we'll definitely
speak more and bubble um you know i've got some i've got some talking i need i could spend an hour
talking to bubble about qr codes and things as well because uh we've got we've got some catching up to do
on that as well so um yeah i'm away for a few days but that won't stop me tweeting about cheer
because you know that's uh that's just how i roll uh birdman is he still in no he's not in but
let's face it it wouldn't be a space without birdman coming in and causing some chaos
for a couple of minutes what a legend he makes me laugh um sorry mr monkey zoo you know i don't know
how long i've what it's been a few months right uh that we've been uh that have been on cheer and
i only just realized now that you can do multiple offers with different uh cats um
this is a fascinating bit of info to me but um because i was just scrolling through the you know
the dexie and then i saw this and anyway i'm i'm doing some uh trio packs and stuff like that so
instead of getting one you get three yada yada yada and i'll do some in mz as well as in cheer
um and yeah that's a wonderful thing i mean we've also got plenty of giveaways now on cheer so um
yeah this is so much more dynamic than the other blockchains but i'm not throwing shade on them
and i'm just saying you guys make it so easy to do these things so appreciate that
yeah i appreciate that jron um yeah check out jron's project i'm sure he's got some links on his um
his profile but you are you're bang on the um when we talk about cat tokens for anyone that's not not
really um sort of with it as such the the cat tokens are pretty much like erc20 tokens on on cheer and
exactly what andy's um created as well so um yeah it's it's certainly uh it's a different world and
you know we'll try and do more uh i do try at least once a month to try and go go into some basics and
um you know aim it towards some of the people that that aren't within the ecosystem so we'll
we'll certainly dive a bit more into cat tokens and i think actually probably a good time would be when
andy comes on in a couple of weeks time we can sort of dig into cat tokens how they work what they are
and and um and go down that route but yeah but yeah appreciate all your time guys um appreciate it
and um you know feel free to to retweet this space um you know just to highlight some of the
important bits that have been talked about and um yeah we'll catch you all up next week
so appreciate your time and we'll catch you up later
