then let's do it
[Music] [Applause]
[Music] welcome everyone to game of nodes you
can see that it's in a completely cut crew from the game of notes crew just myself usurper
leading us off The Fray will be joining us shortly uh let's see null has the sniffles so
can't join us and uh schultzy is I
believe in an airplane so here we are and with us today we have Simon from noise Network we're going to talk a bit around Randomness and the importance of
Randomness and all that kind of good stuff so yeah welcome Simon thanks for joining yeah thank you very much uh thanks for
having me it's a pleasure so the freight yeah so The Fray should be joining in a couple minutes why don't we
uh why don't we kick it off and just tell me a little about yourself and your background and and kind of how you got
to this point to launching noise and we'll go from there sure uh so yeah try to make it short
um my software my background software development um I studied mathematics but I was never
good enough to be a scientist so um I studied
software development with always an interest in cryptography so
um in 2018 suddenly everybody was coming around the corner with some sort of
blockchain project and I ignored this whole blockchain stuff for a long time because I was never a fan of
um proof of work in Bitcoin and all this energy burning stuff so I missed a lot
of the development that were happening in the meantime and I was quite excited to see all this proof of work delegated
proof of stake uh proof of stake delegated proof of stake this this kind of development that that happened in the
meantime um so I started um working at ioe which is
um developing this product called starname was a developer there met a lot of very
good people in the blockchain space especially Martin and Ethan who
co-founded configure with me um in yeah almost three years ago we
started um conceal as a company that was
bringing cosmosom from proof of concept to a production ready product
and this is what I'm still doing maintaining cars and wasm and cosm Jazz
at configure this is what I spend most of my time with um and yeah along this journey I've
always been interested in Randomness and making Randomness available on chain and
all this kind of stuff so um instead of noise as a side project
something that I do in my nights and weekends um but given that there's a lot of
synergy with my day job at configure um Turned out it's uh it's a nice
combination to to do both and so do you still work at confu yes
yes I do um so noise is not a configure project um I want to get this clear this is
important to to separate um but I'm working at a configure four
days a week okay it's cool and when did you when did you start um when did you launch no it hasn't been
launched yet but why don't you first start like putting code down towards it and kind of working towards these test Nets and everything else that's been going on
um so the recent development is just two months old um however the team that is farming
right now is based around ideas that I published in blog posts two years ago
um where I figured out there's this Durant project which is a random number
generator um a decentralized random number generator on the internet that has nothing to do with blockchain however
from this Durant project you could get random beacons which contain a digital
cryptographic signature and those signatures can be verified using
cosmosum on chain so I yeah made a proof
of concept and wrote an article about that two years ago a couple of months later someone brought that to
um to Terra um using um terrand as as the name of the project
and build a build a lottery on top of Terran and this guy today is within the
noise team one of the four co-founders and yeah so it started very early it was
never a priority for me to to work on that but recently a bunch of people came
together and wanted to make this real and now we we go one step further than everything
that has been built before so we're going IBC now so which is a the big
change that was not possible one or two years ago or at least the tooling and
the Frameworks and everything was not ready so before because I do have some
questions around that component of it like why is it a chain and you know kind of what what does that allow from an IPC perspective
and we'll get to d-rand as well because I the way I want to understand that that relationship between noise and and that
as a project right like the dran structure so maybe maybe for me is being
a light in the structure tell me about why this is important like why is Randomness important why is it so
difficult um and and and why is this special in that
situation right so I see Randomness really as a basic
building block that every application developer should have available there
are use cases the most obvious one in blockchain is is lotteries or any kind
of casinos however there's interesting um use cases in the field of nft shuffling
nft creation but also governance processes where you want to do a random sampling from yeah
imagine you have a group of people and you don't want to select the ones who are the loudest but you want to randomly
select certain people and perform subgroups for for certain decisions
um so there's a lot in the governance field and also I had recently had
conversations with people explaining to me that um Randomness is an important factor for
um more advanced cryptography especially in the field of zero knowledge technology
um so um there's probably more use cases emerging
um now I think Randomness should be as easy as
querying a local random number generator for a smart contract developer however
Randomness is very hard in a blockchain environment because
um everything that we execute on a blockchain is executed in in a sandbox
um the moment exactly yeah so the moment this uh executable software is able to
access environment data um then every node will do that and every node will provide different
results um to feed into the computation and then the final result of the computation will
be different and everybody falls out of consensus yeah so it's not only the case with Randomness but also with times for
example if I access my local time of my computer as a node then I
will get a different result as someone else's note um maybe not in the second part but
certainly in the fractions of the second so uh accessing local times is
impossible with smart contracts or any blockchain application um
just for times this problem has been solved because tenement provides a time as part of the block creation it's not a
perfect time but it's certainly good for a lot of use cases but for Randomness
this is not available from tenement there are a bunch of other consensus
mechanisms consensus protocols where Randomness is needed for blood production
um but in tenements we don't need Randomness in blood production which on the other one hand is good because we
don't need it to produce blocks um no dependency on Randomness
um but on the other hand we don't get Randomness for free as a application developer because tenement does not
provide it yeah so uh yeah we we want to solve this lack of this primitive by
providing that on an application layer
and I see uh so um I assume that's because you were watching even with this
cold you saw the shitty host job I was doing so what happened you launched yourself into here
I couldn't sit there and just watch you fly all by yourself
I appreciate it I I do have a cold and I I might be sniffing a lot but um I just
jammed myself full of drugs and jumped on here for for support um but also you know I feel like uh
potentially I've read up a little bit more on um Boyson and generally on Simon
so I'm not saying you're doing a [ __ ] job at you know read between the lines
also block paint has joined this because he also felt bad on my hosting ability so now now we have tripled the number of
hosts in this call because of that and but to be fair Todd probably is
better uh you know suited to make comment on any kind of you know any
security or anything every single topic on this show yes it's
true delegate with buckling it's exactly right that's cool um so so Simon uh the oh
sorry welcome Todd did you have did you did you want some uh some introductory comments from yourself
do you have a were you thinking that usurper was doing a [ __ ] job oh you know I only watched for like two or three
minutes and then uh kids so uh uh I'll be nice
um next time so so the uh Simon the the application layer is derand right is
that I think you were saying that earlier and you know I was getting my stuff in order
yeah so exactly and the randomness is generated um in Durant Durant is a protocol and
there's an instance of it which is run by the league of entropy and the league
of entropy is a Consortium of um of companies that put their name uh
out there put their name on at stake without any blockchain and they do
multi-party computation and make sure that the randomness that is generated is is generated in a fair manner way that
no participant has information about the randomness until the point where it is
published so how do they agree on the
um like Randomness so D randers It's like a network right dram and then so these
participants they they generate some Randomness and then all agree that that's not [ __ ] randomness
yeah so the way it works um at high level is um they do a threshold
signature using BLS um so they all sign
um the same message which is defined by the protocol which which messages to be
signed and they have a shared public key and as
soon as a certain threshold is reached they create a combined signature a
signature of this whole multi-party um and this signature is unique
um so we can use this signature directly hash it and have the randomness
so on what frequency basis are they creating new randomness
yeah so the randomness is generated in rounds and those rounds are counted and
one round is produced every 30 seconds okay so pretty often
yeah it really depends on the use case if you have um something with governance that's
certainly more often than you need if you have a higher frequency game then that can already be not often enough
and then we're not with noise utilize the existing dran Network off of another piece or with or or those validators
that would be part of say as an El as a as a chain here would they run would they CR would you create
a separate dran Network that's only for kind of within the cosmos type of idea yeah so in the first iteration we'll use
um the existing instance from the league of entropy because um it is working very well
um it is actually powering filecoin so the moment this random number generator
has issues a file coin blockchain will stop um so there's a lot of economic
power behind making sure those those companies produce the randomness
continue to do so um and yeah we could
um create our own instance if we feel the need to do so um and there's certainly
um Improvement potential but for now we will use the existing one
so is um is the league of entropy is that like open for membership or is it
just like a closed off uh like a Walled Garden of just trusted people
um it is um a proof of authority kind of system
so you will have to certainly have to um to go to them and explain who you are
explain uh why you are able to run a high availability uh highly secure
infrastructure but they are open for new participants they also want to increase
the multi-party set so um I think
um it's reputable companies um that run infrastructure
um other than Durant as their as their day job
but they will be most likely be open to you so I had some questions I'm not sure
if uh usurper already asked about this stuff but not everyone May understand
the significance of Randomness in games and you know the history behind
some of what people have done wrong in the past um can maybe you talk on that a little
bit so what I um explained before was
um why Randomness in blockchain is hard to get um I could go a little bit more into
what people do to try to get Randomness which which often causes issues
um so a typical example is people use the
block height and block time for Randomness which is something that has
very low entropy so if I am an attacker and I am on the edge to decide do I
still want to join this game or do I not want to join this game do I submit my transaction or not I can pretty much
iterate through all the options for the next block Heights which is a handful of
possible options where where the finalization might take place and also
the um the block times have very low entropy so I can just look at all the options
and see if the results the computation results would be in my favor and if yes I join if not then I do not join
um so that's an issue um there's other um
options where people use signature schemes to provide Randomness in some
challenge schema where they say hey sign up please sign this message which I just generated
for you and then the sign a signs this message and then they use the signature for um
for the randomness however the two popular signature schemes that
we are using which is SEC p and ed25519 they both produce signatures that are
deterministic but not unique which means the signer can generate as many
signatures as the signer wants and can iterate through signatures until the
signer finds a signature that generates an output that is in favor for the
signer because the signer might as well be a participant in your
um in your application so those are basically the two
ways to not generate randomness um yeah Regarding why Randomness is
needed I think there's uh many many use cases um the most obvious ones are lotteries
casinos of of any sort but also I'm building
um a demo application just to prove the point where I'm doing Monte Carlo simulations so I'm basically estimating
Pi by dropping random dots on a square and calculate the the distance between
the origin and the dot which is a fun
and yeah just one of those things that that you can do if you have Randomness
um at the cosmos conference in in Medellin the last two
weeks um I spoke to a lot of people in the gaming industry
um who need Randomness to Define all sorts of user interaction which I'm not
an expert of but I can certainly understand hey um if a blockchain
should feel alive and dynamic and like some environment that is coming with
surprises and is exciting then you don't want everything to be able to
pre-calculate
how will this be used across chains and you know if you know you're doing a
packet across a bridge isn't that something that's revealed in advance then
yes um so the important thing is um that you close your round
um and make no interaction possible at the moment where you commit to a certain round of randomness so what you're doing
is saying hey um I want um a Randomness which is published after
point x in time so I know okay my round can lead up to point x
um and noise ensures that the randomness that I will get is not yet published when when the point x is is reached and
then it does not matter if you get the randomness to make your calculations
um within one second or five or even 10
minutes um of course faster is convenient but it's important that you close your round
whatever that means in your application um while there are processing is taking
place awesome thank you for explaining my pleasure so
so why um why then do we need an app chain why can
people not just access directly from the league of entropy
did I miss this already no no actually we're just talking about we're getting to that question because if it is
through League of entropy today it seems like it could be maybe contract based but maybe that's wrong and maybe as it
grows go ahead Simon tell us sure
[Laughter] um it has been shown by Tara and on Terra
that it is possible to implement that as a smart contract um on a single chain
um so strictly speaking you can have a verification contract on every single
chain where you submit the randomness um you that the chain cannot access the
league of entropy directly because it cannot do Network calls um in the league of entropy Durant is
not within any blockchain uh space but you have to submit the beacons from there into
um into uh smart contracts blockchain um
now the issue is that if you have all those instances for contracts that verify
Randomness first of all they all take a lot of gas so what we have to do is BLS
signature verification which is heavy heavy computation um
so on some chains that is feasible um it might cost you a dollar or two to
verify such a Randomness but then other chains uh the block space might not be
sufficiently big to provide those randomness
okay so basically it comes down to if you if
you're doing it from NSC it did Simon freeze
no oh well oh sorry was he was he still talking when I was talking no his video's been kind of coming in now but
his audio's been good go ahead oh Ken Simon hear me Simon can you hear me
I guess he's going in and out did he did he say something I got here
no yeah this this platform sometimes is a little bit strange where you can like I can see and hear other people but not
like someone else if that makes sense or I can see them but I can't hear them anyway you're on the Starlet uh
yeah yeah I'm on the I'm on elon's teeth it's working well
so he's busted off it's funny like this this this platform's like oh you got to
use um you got to use like wired connection and
[ __ ] and it's like I'm on I'm on wireless to a satellite
still seems to work okay you're a little latency there but that's it's really not that bad though
oh yeah yeah that's good yeah I mean it's probably just me thinking though it's probably not really latency
it could be I mean it's good oh man I actually I was gonna ask Simon a
question I bloody forgot it so now I'm actually writing notes Here on a
piece of paper oh he's back can you add him up yeah I sure can there he is he's back uh sorry
there was my internet no worries where are you um situated what um what country
um I'm in Germany in the middle of Germany okay a little close to the notes
I didn't want to say like you know where are you located exactly like you drop that on The Fray and he'll tell you what
Pub is that it's uh I didn't want to throw you under the
docs bus uh yeah so the company is in Berlin but I'm not um I think Berlin is a bit too hyped
it was never my city oh no that's okay it's it's nice like how do you say that
noise yeah so the name is just coming from noise uh like the ocean like this your
the signal stuff but you can never predict what what you get
um and we just dropped the E to make it a fantasy name but you can just pronounce
it pronounce it like noise yeah okay
noise yeah no it's cool um so the the noise noise network uh
come is it a company like in Germany uh no we have no no company we this time uh
we go fully activism mode with just a handful of individuals and no token
sales at no company and no structure um so I mean I'm I'm I'm not hiding not
hiding myself but um I've been through the process of
setting up a foundation and trying to get a bank account and trying to get a payroll up and everything
um I've done that and doing that once is enough
um especially since we don't have the demands to higher up a large team we
don't need um Fiat in the bank in order to to get this project started
so we'll probably operate by just dropping tokens
um and get started away the foundation that was started was that
um the confio uh
yes yeah configure is the development
company in Berlin where the team is employed so we have 25 people in in
configure right now um there is a Dutch Foundation which is stitching ocean blue that's the
foundation behind the launch of the tigray token and this Foundation is the one that is
was doing the token sales and this um yeah Contracting continue for
building team Freight so why is it so important to do that um behind a foundation
because the moment you issue a token you get into the business of financial
instruments and you probably don't want to issue tokens and sell tokens as an
individual I mean you can but somebody has to pay the lawyers and
um make sure everything is nice and compliant and um
so it's all pretty sketchy yeah you want to try everything
it's okay it makes sense um it's just a lot of work
um and if you're planning a big big project then this certainly necessary
but if we can get noise out of the door without anyone employed without
um paying any external contractor without having to sell a single token
before launch and just make sure the product is out there people start using it people enjoy it and decide whether or
not it's worth founding a company then to me right now for this particular
project that's the better way so for the for the launch of the network
then is it gonna like do you plan to just do a pretty low-key thing um and just sort of you know have a have
a decent you know set of validators and then just throw it up like a smallish smallish set of validators get it out
there drop a token and then just let people use it pretty much uh yeah I think
that is more or less the plan um to make sure the focus is on the
users um and the communities that want to use
the product and not so much on um how any kind of marketing or selling the
token or this kind of stuff um I mean we are planning a bunch of
cool things for the launch of the token um but um yeah
try to to do everything within the team of the four co-founders without that
much external resources needed so who is the oh sorry okay okay
so what before we get to his story
um the what's the what's I assume there's a a payment or a fee structure
that that these the app like the clients that use the the network that would basically pay into that is that that is
that the idea and do you think that like the noise token itself will have a value to it and would there be a minting type
of structure you think it's just really based on a fee structure because people are using the network and they're paying fees to be able to use the network
yeah so there is an income stream for sure um which is
um applications using our service uh the noise service via IBC to get Randomness
they will have to purchase the token in order to pay for for that service
in the beginning it will be very simple as a pay-as-you-go payment you pay for
every Randomness that you request but we are also thinking in the direction of
subscriptions where especially applications that use a lot of rounds pay more like a flat fee
um or yeah somewhere between uh per usage fee and some flat fee
because we also want to make it attractive for games that need a lot of
Randomness as well as governance applications who might only need a Randomness one for quarter
right and so an evaluator Network itself obviously they're you know dealing with
the IBC structure and the transactions that are coming in through that piece and they do have they do have a role in
the in the the D Rand um integration right so there's a
request or are there there's an integration there between the drad network and obviously this this network
that's generating that random number and then there's a consensus from the validators to say this is a random number for this specific round right
that they're they're agreeing that this is exactly what the red said right so there's a there's not there's a a sense
of um of um completeness to say that this all these this validator said has agreed that this is the number that we
all see from dran so so actually the validator set itself is an oracle to d-rand
is that right they don't have to decide on that um and they don't have to check that
because the verification if a particular year-end round is correct or not um is
done in cosmolism and fully automated fully executed because each um a
year-round Beacon comes with a signature and that can be verified in cozumelism so it's not like
um um um Oracle where different validators
would have to double check if the data is really correct and Vary some external
endpoint at the moment they get the Beacon it is added as a transaction into
a block um it's automatically verified yeah
okay so yeah and we have Bots who do this
um bridging between reading Randomness from Durant and bringing that onto the noise chain and we have an
incentivization model for Bots for bot operators which is basically just
external clients that are running and observing the dran network and every time a new Randomness comes in they
submit a transaction to the noise Network and make that um Randomness available as fast as
possible cool so so the network itself the validator network itself obviously I
mean if it's just running wasm it's not really much of a of a value there so
I guess or could be able to grow and it could bring D Rand in-house and it you're basically building for the future
in that sense right because everything you just said right there they understand it it sounds like it could be a contract in Juno
right or or any other sort of cosmosom type structure um but you're building right away for
for future capabilities right so there is the the MVP can be built as a contractor
on Juno um there's no doubt about that however there's um
tons of optimization potential that we already have have listed when doing that
um on our own chain so the whole verification process could be speeded up
by using a different wasm compiler we could have um
crypto apis who are providing the verification functionality for BLS
verification in a way that is much more efficient but right now the most time we
spend regarding our chain is um thinking about block space how we
utilize blog space how frequent our blocks should be how much block space we
need um how small we can make blocks in order to get higher frequency blocks because
we want people who who submit or end their their rounds and are
waiting for this Randomness to get this Randomness very very very fast because
the time that it takes to process the randomness is the time that a user is sitting in front of a computer staring
at the spinning wheel waiting for a result and we do fight for every 100
milliseconds that we can get in order to to speed that up and being able to have
full control over our blog space right now just what is in there but with
upcoming prioritization of transactions and and
even yeah with ABC I plus plus even more
flexibility on how plug space is utilized um we see a much more potential for the
future there so Frey welcome
let me let me let me update you on the on the what's happened in the last 41 minutes this this started as a
two-person call and then null jumped in because it's doing such a bad job then Todd jumped in because Noah
was doing such a bad job you've already explained everything
there is to be explained then at this point I'm guessing well that I mean that that piece right there around why a
chain versus a contract was very helpful I think I think I get it for this in this small period of time right here
yeah that's cool but we've basically just been here like slamming Simon with
questions this is really what it's been I've been the most educational episode I mean that
kind of sounds like if anything the podcast is improving without my presence so
I I will say that when I when I launched it I couldn't remember what our tag lighting was so I just skipped it I was
like hey hey gaming nodes I forgot I don't think I even said game of notes it's just like
yeah hi we are you can fix that way were you nervous you said were you having a little nervous Nelly moment I wasn't
nervous I I know I could keep Simon entertained for at least 25 minutes I was worried more that this would be the
shortest episode that's ever happened and then he kicks he kicks Simon off
after an hour you're just there by yourself just like that I would just I would [ __ ] talk two of the three of you
for the next hour that's what I would do I would just I would just rail into both of you Todd gets a pass Yeah well yeah
I mean yeah fair enough I've I've I we we moved back an hour because Noel was
ill and and I had other responsibilities and then just blew straight through that time limit anyway
yeah I was I as everybody knows I am a very big football fan and I was at the
football let's go with that is that definitely what I was doing is that is that definitely what you were doing I
was 100 at the football um football that's that's me
love love the football um
is in the air both of those times love love what they're doing there
um I love that I love that scene in transporting what are you doing football
what about you shopping probably had no idea what I'm talking
about thank you I think this may be like one time ever that a movie reference has gone over my head a little bit I don't
recall really and the girls come up to them at the
club and they're like what are you talking about and the boys are like football oh we're not talking about things they were talking about the girls
uh okay yeah and the girls were talking about the boys and then the boys get what are you talking about they're like shopping okay
yeah I guess I only really remember the dramatic traumatic moments from that it's not nearly as funny when you have
to explain it yeah it's kind of yeah kind of sucks that now
yeah sorry about that um so you know uh you know don't let me
interrupt like should we should we continue asking questions well I feel like you really want to know some
some Cosmos and [ __ ] are you interested in the in the randomness or the or the cousin wasm well I'm wondering like so I
put some questions in spreadsheet I don't even read the Press read the spreadsheet but I I I kind of thought it might be good
um for for the non-devs to actually like do some kind of foundational stuff on
Randomness but I don't know if you guys already kind of covered that yeah more or less that was Todd's first question
he said for the plebs out there why do we need randomness July
28th of July was actually a test of Juno's ability to create a random Beacon
via the process of app hashing and it works it turns out it does work yeah yeah
um so how about really hard to do because you have to use auth Z but you do get a random hash out at the end uh
this will be interesting um for the phrase why don't you recap your questions and I'll see if I
can answer them okay I love that right so now
why why is Randomness important in in a blockchain context
honestly because reasons no so so uh from what I understand so we
need um you know well things need Randomness and it's just hard to generate well
difficult to generate reliably randomly in a blockchain so you know casinos
obviously uh games need Randomness other things need Randomness and it's just
hard to generate is it was that is that the question and right okay so we've got some some
decentralized applications might need Randomness yeah okay um so you're saying why specific
why is Randomness so so hard to come by in a blockchain context what's the challenges why is it something that we
even have to talk about rather than just like generator and you're done because of lack of entropy in a
blockchain and because the smart contracts can't reach outside of their little contract environment their little
uh you know VM so they have to rely on things that are available to them and I
think Simon said the things that could possibly generate Randomness is the
block time and the block higher like they're predictable and thus you can
work out a handful of combinations of what they may actually result in giving
you the advantage of being able to predict it somewhat and then cover your bases in you know if you're doing
something that requires that answer so
the true Randomness can't be generated inside the blockchain right so why would I not just uh import
a library and and just like do it in a smart contract
so reason that was So when you say a library what do you
mean by a library well you can generate yeah it's obviously you can generate any number of
ways I know what you're getting to I know what you're getting to so
to do that inside a inside the blockchain to to get
Randomness that is truly random then that will give everyone different answers and you can't commit that state
to the blockchain or your app hash everything is that right did I win yeah
you've won the game yeah you've won the game of explaining about the situation very very well done very well handled
Simon must be an absolutely excellent teacher because that evidently only took the the 28 minutes that I was gone to
explain all that so he's a good teacher to be fair Simon uh explained it in much
better detail and context and less stupid words
I mean I could I feel like Simon's just sitting there cringing a little bit he's like it's not really what I said exactly
yeah it's pretty good um I mean it captures the most important
points if all we do a game of nose is we have the Geniuses we bring on their fire and
then the the the the the the general public that I we're just a lukewarm water in between then I can look that's
right just a little bit of that rubs off I think that's that's pretty good right
yeah 100 um you know I actually wrote down the main points on my piece of paper for my
uh article on the whole situation you you knew that you knew that the phrase quiz was coming the phrase Chris died
somebody's gonna ask me about these so so my
my notes don't consist of much so it's like it literally says D Rand
uh uh League of entropy and then under why app I have like it
can't be implemented in a smart contract There You Go properly and then I was
like I hate writing notes I'm done yeah um
[Music]
no I don't um I mean I usually write show notes down
in the in the thingo but I feel like they just get ignored or are too far off to the right to be seen where do you
rate them that's a lie yeah I write it on the right hand column
of the you know the the row certain column yes XXX yeah
yeah yeah I was gonna say like H H keep scrolling he's gonna keep scrolling
and you'll see all the hard work okay I believe it when I see it um
so uh have you talked about cosmosom at all we haven't no we haven't but before
we do it oh sorry you have something else we haven't got like that yeah so I just actually I wanted to talk about the
token just a little bit um because yeah it's not tokenomics hour here we go DJ no it's not it's not gonna
be a stable coin calm down it's not going to be a stable coin
I don't think so I think
like anyway so with your token you say you're just
gonna airdrop a bunch of [ __ ] out there right just spray it around
well yeah so is that what you wrote down is that what did you write did you write
spray it spray it around in your notes over there I don't think that's exactly what came up well I I had a couple of
pointy questions I I might are you going to do inflation like do you need inflation are we trying to
encourage people to like do regular users even have a use for this token or
is it for me it's just going to be people who want to run apps and stuff that want to use the the noise of the
network um will want the token right so I'm thinking like
you know what's the value proposition for users is is the demand going to be that great that
um you know it's a good a good punt for people or like do you even need to
distribute it can you just not sell it and then you know have the like can you not just
create a pool and people can just buy it out of the pool and and then the validators collect
um you know uh rewards and then sell them back to the pool hey so I'm just
wondering like is it is it the type of thing that actually needs a wide distribution and staking is is the basis
of my questioning um so right now we're really focusing on
the product um and the token is a little bit like yeah tokeno makes you do that later
um and I think there have been a number of decisions that
um Define the tokenomics and first of all as we do don't do any pre-sales
um second is there is um a native demand for the token
um through the service that we're offering so in contrast to many other tokens there's actually
a business case for it so that also might create a secondary a
speculation case so you can buy the token in the hope that it is worth more
later um we yeah will not we know yeah I already
said that uh yeah you said you said you've not been doing Styles yeah yeah um and one thing that we want to make
sure is we haven't an inflation model
um that rewards all the network participants and those are not only black creators
um there is a inflation that goes to validators but there's also inflation that is going to the league of entropy
um because they are doing a fantastic job in running this infrastructure and they um should have a share even if it's
a small chair in the rewards generated by this project and not only the rewards
not on the the monetary aspect but they should also have an ownership aspect and
a participant um share in in noise to make sure that
whenever there's anything ongoing on dear end and the Ring of entropy they make sure they
have aligned interest with noise and make sure we are um talking and we know what's going on
well in advance and they are not making any crazy changes that they are
discussing between the league of entropy and filecoin as the biggest user and we
only hear about them when it's too late um so for example um the Durant team is working on shorter
blocktimes run times for the randomness in the order of magnitude of five seconds
instead of 30 seconds and all this kind of stuff um is good to to establish a close
relationship with the league of entropy um as long as we we use their service
and collaborate with them and there's tokens that should go to the bot
operators that provide the bridge between the dran network and the noise
blockchain so inflation should go in in multiple pots and that's going to happen
for sure um we will make that token available on some
decentralized Exchange in order to allow people buying it who want to use the service or who want to have the token
for whatever reason um and yeah that's that's basically how
far the thinking is for now is it possible to have a
non-inflationary mechanism and only have like
the distribution of the um like transaction fees for for the use of
the network go to those separate pots the reason I say this is so I'm I'm
pretty like I just like for example the way kujira did it even though it keeps every
validator broke um in that they're not sort of
for now but they're not trying to like um uh be something that
yeah they're not if that makes sense like the the inflation is like
it's it's the um you know it's basically writing its own
destiny like your your inflation pumps your coin up to the users because they
want to get that return on the coin right when it's really like to me it's a pretty disingenuous system
because it doesn't value the coin based on its its use cases and its actual
usage at all it only you know people want the coin in the
short term because it's got inflation and they can get a return on it and dump it in the back into the um right the
pools and and it's got like a it's got a limited um lifespan you know lifespan because
once the the pool rewards and stuff run out um and once the inflation goes down the
chain it just ends up being a dump coin and then you know the the hype around it Fizzles
out whereas in my humble opinion I feel like it's better to not have the hype to
begin with and then just have the the token based on its usage and the actual
income generated from it based on a demand for its usage not an inflationary
like um mechanism that makes people want to because it's it's it's a [ __ ]
mechanism that just pumps up projects so people can dump on the pools yeah but then you're implying
no you're completely right you're you're implying that the inflation goes to
delegators who do nothing else than uh buying tokens putting them somewhere and
waiting for more tokens to come back um the inflation model that we are
thinking about is making sure the inflation goes to those points where
value is generated um inflation should go to a development team inflation should go to validators
who run the infrastructure inflation should go to the direct Network inflation should go to bot operators in
a way that um the inflation is much closer in line
with the value generated by the particular parties and I prefer a token
model where you start with a very minimal initial Supply because why dump
a massive initial Supply on top of anybody when there is little
value created but increase the supply start with something small and increase
the supply as the value increases so
I mean back I guess that sort of goes back to the question is why can't you just
distribute the fees to those parties instead of actually have inflation to those parties uh yeah because fees uh
everybody wants zero fees and we have fees that are barely enough to protect
spam and it's a race to zero and we are
seeing that all over the place and um uh yeah the feet are not
um at least not the at least on not the the network fees
maybe you have to differentiate between Network trees and product fees and then right that right that's what I mean
that's what I mean the usage of the product yeah yeah because because that
makes sense then that means that like you know if the chain sees usage it has value and if it doesn't then it doesn't
and it should just die yes absolutely uh 100 I think you can
implement it with inflation by burning the um the product keys that you earn as a
change um so you have income and and um and
yeah tokens that are taken back so possibly a dumb question here then this
sounds like if Staker our stake is going to be rewarded at all is it still going to be a proof of State chain or is it
going to be uh a kind of custom inflationary model that isn't really proof of stake that is something
different the tokens are in the tokens that are in Supply are liquid and they're and the validator bonding
mechanism is something more like a foundation bonding or something and it's more about just Direct Security
and no we'll be working with uh Cosmos staking module um pretty much 100 just not all the
inflation goes to um goes to the block producers and the delegators
um right now I think in the module the distribution module that gives you one
parameter where it says okay you can give like five percent to your community pool yeah it's like community pool and
then everything else is just like yeah right yeah and default means uh 95 or
even more go to block producers and I question that 95 of the value is
generated by block producers um we love our validators we also
appreciate the staking mechanism that's that's all great but why not 50 of the
inflation go to by producers and delegators and the rest to to someone else
um about 60 or 40. a naive question uh would be given that why what's the isn't
there like a uh might Okay so
you can obviously modify the staking module which sounds like what the plan is but you could also presumably
um write that in cosmoasm and implement it as a privileged contract within
something like uh tiwasm right that would be another way of going about implementing a custom staking system
that that didn't really have to rely on um the way the SDK does it if you know
what I mean yeah absolutely uh we have done that in configure we have done that uh 40 great
um that all works um but that is a lot of work um it it has been implemented in in
proof of Engagement it works entirely different it completely swaps out the staking module from Cosmos SDK while
um continuing to use everything else um and I want to focus on the product
and not on the token distribution so I'm more leaning towards ideas that have
been developed by the stargaze team where they have a distribution module that is more flexible but overall works
like any other customer SDK chain they just say yeah we
also have an inflation share that goes here and an inflation share that goes there
um and distribute it differently um which I think makes sense it's a good
compromise between a little bit more flexibility without having to reinvent
this whole security aspect of staking as well as yeah the tokenomics and everything that
that users are used to right okay that's interesting because we I I I'm I was kind of interested to hear
you go in that direction just because I assumed with the with the kind of t-grade connection that it would be like
oh well there's actually two but by the sound of things it's just it's too much distraction from the product
yeah exactly so I'm completely open to those ideas um and I'm talking with
um all kinds of people about how to use the flexibility that cosm wasn't
provides to um replace those core modules of Cosmos SDK and for each of those modules for
each of those slots have a variety of plugins that you can choose and say yeah
I want to have this type of staking and I want to have this version of the bank
and I want to have that sort of vesting accounts um and have
yeah mix and match system however for noise right now it would really just
delay the launch of the chain which is certainly something that we would absolutely avoid right okay that makes
sense um because it I guess so we've obviously we've been working on this this howl
thing for a while and something became really obvious to us like really quickly is that we needed ipfs nodes
and then you kind of go all right but they need to be paid for and the
validators I guess are going to have to run them so like what what we the the
kind of thing that we started looking at was was tiwasm um for that because obviously we're all smart contract
developers mainly like that's the rest of the team I'm the only one that's really done any SDK work and
unlike the SDK is I think quite hard to work with um compared to cosmosm so it was one of
those where we were like okay well probably um tiwasm represents a pretty good
compromise from where we're standing albeit you've got to kind of start from from scratch right with the things
you're implementing what what is what is 2000 can you explain for people uh well I mean Simon directly worked at it right
so Simon could also probably explain it okay yeah so I did not directly work on it um
but it's basically a module in tigrate that is T great specific that is part of
the proof of Engagement um implementation and proof of Engagement is a reward or inflation
distribution mechanism that is based on the idea that there are no delegators
um and only validators um stake their own tokens which belong
to them um and there is rewards distribution based on engagement points which are
collected by any sort of contribution to the ecosystem um and then are assigned by a committee
um on chain and those engagement points um then allow you to earn part of the
the rewards um so you can see how that would be generalized that if you're providing
storage Todd you could say okay well it's kind of the same in the you go you have a set validators who are providing
storage you can then I mean are we even all these sort of folks have already long
solved the kind of prisoner's dilemma element of how you prove the quantity
the game theory component of how you prove the content is alive that's quite trivial within the kind of Economics of
it and so then it it's it's it's actually a lot simpler than proof of Engagement to do infrastructure based
um stuff that sort of relies on the validator set but is outside of consensus I think basically
um yeah there's a lot of change that have that problem right now yeah exactly so this is it's this kind of that like
oh I'm just sending proofs that this stuff exists oh that's doable in an end blocker right it's that kind of uh use
case essentially which has been interesting from the start and I've been bugging Ethan with questions about it
for some time but it's not been a thing that we've it's taken us so long to build what we're building that
it's not really a bit of problem until we're like we're like about to launch and we're like oh yeah that problem that we were like we still don't have any
ipfs nodes what we're going to do about that yeah yeah
do we know how to incentivize it no not yet but um well we do know how to we
just haven't made it yeah maybe
yeah now you're muted can you tell me something about tigrade uh with the
the validators can stake their own tokens what does that mean for the
validator like does that affect it doesn't affect anything does it like you can have no token state to yourself
right like it doesn't affect your Awards but it doesn't affect your voting
uh yeah it does um so very worth it you're getting is
um calculated based on both stake that you put at stake and engagement that you
provide um and the idea the base idea is that if somebody who's very established who's
sitting on a large amount of tokens and puts a lot of tokens at risk to make sure they do their job properly they can
get a high reward and a newcomer who's just coming in who's not having that much tokens but it's very very engaged
doing a lot for the community can also earn um a lot of a lot of rewards making
worth the effort of spending all this time and energy into this ecosystem so
it's really those two components and then there's a reward
calculation based on on those two inputs is there a formulation that's available
for that there has been a white paper that was published which contains all the basic
ideas and models um and the curves that I use to
generate the rewards based on the two inputs um
but I'm not super familiar with the details of how the chain is running at the
moment since I'm not not really into the t-ray team
so what's what's your main team within um confio
grew a lot in the last two years uh we are 25 people now
um and I'm focusing primarily on the public goods stack
which is uh cosm was a maintenance and cosm.js maintenance because the Muslim
is the Smart contract stack um and cosm.js is the low level client-side Library
for for app Builders um so that that's what I'm I'm focusing
I like the low level stuff I don't enjoy the product questions too much I
don't understand markets I don't understand users [Laughter]
same doesn't mean you don't want to talk let's not mean you don't want to talk to users that's a difference
I love the brutal honesty I'm just like I don't understand I like this code man uh all these
irrationalities what Twitter you never understand Twitter
um guys do you think you're at that time can we thank Simon and move on have you
said everything you'd like to say Simon or is there any anything else you want to bring up while you're here man you want to tell us you know
how pretty weird or anything no just uh thank you very much for the
invitation and um look man thank you for coming on um thanks for all your hard work like the the number of projects
that you're a part of like I guess like I also like from uh the perspective of like you know being on the core chain of
a team where we're kind of Downstream dependency and like we have to whenever Simon drops into whenever I get a
message from Simon I'm like oh [ __ ] that's my reflection because yeah I'm
just like oh no oh no I don't know what is it like I see it pop up my phone and I'm just like well my whatever I plan on
doing today I'm not doing it today anymore and then when it's sometimes it's just something completely harmless like oh by the way you did something and
I was like yeah sometimes we have security issues to share
I saw him at a conference and then he came up to me a little bit later and was just like hey do you have time for a coffee and I was like oh no
anytime everybody from configure wants to talk to me I know something bad is about to go down
it wasn't that bad uh well well we we had to quietly leave
we had to quietly leave and then go fix it didn't we um it was one of the things it was in May it was like one of the exploits that
came up just before version one we had to go back to our Airbnb and fix it and Pat [ __ ]
yes thank you so much for all the all the
work you and the comfio team do um yeah we wouldn't be yeah that that
the ecosystem wouldn't have even a fraction of the of the value it's got
now without um Cosmos and you guys working hard on it in the background and um I feel like there's not a lot of
recognition as well in that part so you can have it from us thank you very much it's really it's
really a pleasure doing that stuff and uh seeing so many people using it um so there's been one of the learnings
from Cosmos as so many people who are using cousin wasn't now who I never see
because it just works for them they don't have issues I don't hear from them so um it's it's really amazing how the
cosmos and ecosystem is exploding
thank you
that was good where's that thing
listen Professor let me just catch you up in the whole situation I was kicking ass on this thing by myself and then I
think null felt bad about it and got involved and then I asked Todd I'm like it's only just two of us I'm like you
what are you doing come on in and Todd joined knowing that I was kicking ass I mean you can go back to the tapes I'm
like we have this on this is recorded right so we can I mean I will I will go back and check it and and then make a
judgment first he was doing great I did forget the name of the show and our
tagline and our guest and your two note your two names but
other than that I kept it that's not bad going really um you know it takes a lot of time to get
to the level of slickness and professionalism that I've achieved where yeah just turned up and say the same line
every episode no matter how badly we've been rugged on the intro like it's a it's a real skill
um like it's it's right there in the notes it says
do you know the funny thing is actually that the one that knowle's written it I think in null wrote that in the notes
that was correct that's actually not the correct topper for the show it's not the correct intro well that's my weekly
podcast
and to also to to maintain the Integrity of gaming nodes I didn't even open that
spreadsheet so that I mean that's what anyway any of you would do right we just ignore that
spreadsheet completely and then afterwards oh [ __ ] all these notes are in there I actually looked at that
there you go there you go um very much so I've just realized one of the one of the
talking points this week is relevant to Simon um but he's gone now so he's gone we'll
never see him again probably who was who who was that guy did one of you guys invite sorry we'll stop making Jason
um but but so a talking point I want to bring up which actually I put on the the she and then I remembered Simon was on
and I was like oh when when shall we speak about that but is um Larry who was
on the show the other week um has been doing some preliminary work and has actually done a surprising
amount I made a surprising amount of progress because he's a shadowy supercoder who apparently doesn't ever sleep
um uh on cwsdk which is uh the can we
just have Cosmos modules as smart contracts and just and just run those suckers and talk to tenderness
talking about that just the other day right yeah yeah right we were talking about the timelines yeah yeah we're
literally talking about like hey is this possible is this the thing that seems like a thing a lot of Debs won and he was kind of and then he just like woke up one morning was just like well I
guess time before lunch yeah exactly and half
an hour before lunch so he just implemented we asked that but he said like it would take two years right or
what have you got or maybe it was you I think 18 months obviously we have a difference and he was like that's a
morning five five minutes yeah getting to Future parity with the SDK but whether or not
you need every feature I think is the question and that's the that's I think the the premise here actually is that a
lot of decentralized applications that might want to move into the interchange
don't need all of those shiny features and actually given what we've seen of
discussions in our mesh security um which is to my understanding my kind
of Layman's understanding of it we can sort of call it sort of adjacent to what people were talking
about as ICS version two okay um so you know kind of like layering of
layering of uh security rented from other places right that kind of idea right
well as I said I asked somebody to come back on because I don't understand any of this [ __ ] so it would be good to yeah
I keep wanting to understand I keep wanting to talk about mesh security but then I'm like I know that you you're trying to get him back on and he he will
at some point actually come on the show and then it would be a bit but in fact we could just get we could actually get
Jake and sunny back on probably really easily we should just do that yeah yeah I think I understand it at a conceptual
level I think the actual like some of the intricacies of it and how that actually works and everything else I think would be interesting to I mean
it's nice that it says a presentation level but when you actually get into it like how would this work right there's a
lot of there's a lot of stuff in the weeds isn't there right yeah exactly so yeah I think I think there's there's
some details yet I don't know if you've seen the implementation you know they they knocked out a a an initial like uh
proof of concept at cosmiverse if you look at the read me and the documentation for it there's like a list
of like hey you should be aware that there are not solutions to these problems yet and they need some
discussion and there's a big bullet point list um it's really interesting reading and actually it's like I say it's well it's
actually quite readable it's well structured if you're a cosmosom Dev whatever you can just have a peek in it
and um that's that's pretty cool um it's pretty cool oh yeah it's super cool
um and it's super cool that Larry is just like yeah not started knocking out this uh the first part which you know kind of tender man
um connection part right so so that means what so if
if that Cosmos SDK can be replaced by wasm right and if
mesh Securities other types of things are wasm driven types of solutions the
way I understand it then what is the value of the cosmos ecosystem is it purely is it it's kind
of centralizing around IBC protocol it begins IBI and ends with b c but that
but that's also going to expand outside the cosmos right because there's a push to say hey screw this whole thing around
Bridges because Bridges [ __ ] Bridges right so if IBC then turns into a global
standard of how ecosystems run and the tenderment piece goes away then what
we're really getting down to is a replay like a replay a blockchain replacement of TCP with IBC and then whatever tool
set and structure people want to build as a as an e so everybody's in L1 it doesn't matter as long as they talk IBC
they're able to be able to take advantage of that and there's no winner
IDC is the winner yeah like there's no there's no ecosystem winner I mean it dies yeah I think I think yeah I think
if you go all the way back to like literally episode one or episode two of gaming news I'm gonna I don't know I
don't watch this show uh yeah well I'm gonna call this the the the the there is a there will be a cliffable quote of me
really a time ago a million years ago in crypto years just saying the only bit of
Cosmos that will survive is IBC and it will come to dominate that's that's my
that's my bet on the cosmos ecosystem and it always has been about but I what I'm surprised by is how quickly now the
conversation is moving to taking that chunk taking the IBC bit and
really generalizing it and you're seeing that with you're seeing that from one hand being pushed by like configure and
these sorts of people who are sort of working very heavily on the cosmosm side of things you're seeing it from Strangelove and Jack's team and all
those guys who are trying to bring other ecosystems into the IBC verse and you're seeing it from obviously informal who
are basically you know built IBC rust which is a standalone imported into your rust project that is
going to be a blockchain and you can do stuff with it um and and and that's like there's
there's a number of different angles there which are going to become different ways of building an IBC stack
over time uh none of them correct all of them very different but like you say like I
mean I was having this conversation with uh Jake the other day about you know how
far we think it is a way that you're going to be able to um have like a viable IBC chain
that sort of bootstrapped from a team who are in the cosmiverse now that
doesn't have tender men as the consensus engine but it's sort of recognizably sort of a cosmos chain in in the sense
of how we kind of think about it but then when you kind of go and look at the GitHub you're like oh wait no no none of this there's a different consensus
engine it's different this it's different that's different the other [Music]
wouldn't it be more likely that an existing chain adopts IBC
and so therefore it meets that same criteria just from the opposite direction that's going to be yeah I think that
will be the big onboarding next into the IBC first will be other chains sort of what we now call bridging but it
will be actually just joining the IBC verse um and that's gonna be really exciting I like I don't know it's going to make the
atom price pump but it's going to be cool for the uh the ecosystem of IBC I
suppose um uh and that's really cool but nobody benefits off that right like it's not
like uh well the end user I guess let me rephrase there's no centralized entity
that that 100 benefits from the use of IBC like like right which is how you can
tell it's probably a good thing right yeah no I'm saying it is a good thing yeah for sure people benefit from IPC
yeah end users benefit and nobody and few people other than developers who are
experts in IBC are really going to make like loads of money off it so it kind of suggests that it definitely is actually
what decentralization is and not what people talk about when they put up gov props wanting to increase the valve set
by five um right which was possibly my spiciest comment of this episode
um so there we go um I I definitely beat your ass with the uh
spicy comment oh yeah don't miss one to miss a good one I just
General okay so chain economics
I mean your your econ well we've we've got a little bit of spice to cover as well actually it kind of news a good
week so we got um do you go for okay I get okay well let's go for let's go for
spiciest first [Applause] Tombstone of the week
yeah but but I come into it completely unprepared because I've forgotten who it
was there was oh [ __ ] it man there's there's literally been so many [ __ ] tombstonings lately that I've stopped
scaring I saw no details it was on the machine I'm not on though I know that yeah
was it on Lou or Sith [ __ ] hell it might have been on Sif it was on Sif right
yes yes so obviously like you know we we are taking the piss a bit but like we
sorry about delegators funds we're obviously like I think as we covered in Pre in the previous thing about
Tombstone it could it could happen to any validator like you know there are things that we don't understand about
the stacks we use but two percent of the week we've seen a validator go down and see if uh they
didn't I think initially noticed that they had actually been tombstoned um which is I
easy common Trend amongst the tombstone is well I guess maybe because maybe the
thing is like again like because I remember when I I have Tombstone to validator on testnet and I remember what
happened was the alerting went you're not producing any blocks and I was like hmm interesting
frantically why would it work I just like yeah just
restart cosm advisor I just while I go check the memory doesn't seem to be any memory pressure because it's not doing
anything um I was like well that's really weird and because I think at that point we
didn't have testnet mint scan as well because it was like a starty or something like that right so I was just like what could it be what could it be
what could it be I think it was just Todd actually who's gone because he was just like hey has anybody else been
Tombstone just questioning just just put it out
there you're like
that was then and this is now right like given the experience you have
in node operating between then and now I think if you've stopped producing
blocks you can immediately think of whether or not you've been
doing sketchy [ __ ] that may result in a tombstone and get worried about it pretty quick and check
don't you don't you think now like there's a difference between then and now that well this last one I think this
one on Sif I think that person brought up a new validate like they they were moving Hardware
they were moving hardware and they just literally copied Keys starting it up and oh [ __ ] I can't do that like it was like
one of those situations right I I don't think it was it wasn't anything it was no it was it was just them and it was
the chain was moving right it was just like uh oh [ __ ] I can't do that I I
would say that is sketchy [ __ ]
you know that type of [ __ ] checklist exactly but but what I'm saying is that like now we know if we're if if you are
all of a sudden in jail I feel like you can do a quick recount
of the last five minutes activities and determine whether or not you're going to be probably tombstoned right yeah right
so I guess for for us like we just we just take the hit and we go down we just we just stop the service I usually
disable it in system d before doing any maintenance and that's actually even now
we we've been on a remote signer now for about 18 000 crypto years which is like
I don't know six seven months or whatever um but that yeah even using a remote
signer like it you're just like okay there's a lot of moving Parts here I haven't ordered every bit of the code for tmkms I think I understand how it
works do I want to bet my life on that no and the fun thing was
I think I mentioned this near their episode when we were talking about tombstones was that like you're on a test net very recently
um uh I do have keys on a box because it's a test net box and I also run tmkms and
I cycled out what I so when you submitted a gen TX you need to have the
original key if you're reusing a box and you want to um use the same remote designer right so
I cycled in the old key to create the Gen TX because I didn't do on my laptop
because I my my I'm an idiot basically that was the that's my excuse I don't have a good
excuse I should have done it on the laptop um but I can't emphasize enough that most of our most of our non-test Network
I think literally all of our non-test net keys and backups are in a bank vault so this would have been even more of a
pain in the ass if I had to go to the [ __ ] bank vault to to then like sit in the safe route
for a test net yeah exactly this is why so test Nets my level of my level of
paranoia is reduced to be like look quality of life so anyway we're bringing out this day and we're recycling on some
other network and um yeah move the key onto the box that we were gonna that we were going to use
and just went okay gen TX then remove the um the file right and then when it
starts it'll actually generate a new one so dish bash Bosch you're done
um what I did was I I pieced This Together afterwards I had actually fat
fingered the command to do what I normally use I typically move the file while I just add dot old on the end and
then move it somewhere and then when I'm sure that it's worked I delete it um this is on a test net on mainnet it
is very much just not never on the box anyway so it doesn't matter um and we wouldn't reuse a box and that
way we bring up a new box because we have scripts that could do that but anyway mainnet is very different test
net test notes a lot less automated um so anyway we got this up and then went
over to I was like that's not signing that's weird and then went over to TMK ABS looked at the log and it was just
like double sign double sign detected prevented double-sided detected
prevented and I was like how the [ __ ] are there it turned out
yeah I went back over to the other thing and yeah the the command line instruction I'd run I just fat fingered it I just like could cross CD or
something and actually it was just the old key yeah so yeah I think it was like a bash group
that had a few a few instructions on it and it must have errored on a previous line and I just don't like the classic thing I'm just putting set e at the top
of the bash script and not thought well hang on if this line errors it's not going to move the old file right and
then that will still be in place and then it will restart the service and I will think that it's restarting the
service with the old file moved out the way generating the new file incorrect because it failed the because of my fat
finger on the my typo it in fact erred out and then I'm just like oh okay that
arrowed restart the service not thinking oh the problem is not the best yeah it's not just like the bash script error for
some reason I'm just like oh that's weird I'll just I assume cosmaviser had just like barfed on start so just
restarted it but it actually had never run because the script had exited because the file hadn't moved out of the
way we uh like that that doesn't add up to me that doesn't add up it it does it's
it's since it's scripted and it's like in a bash script you don't know was it a separate note
like were you trying to just change your note from having that key on it to Bringing on like to changing it from a
local key to remote Sonic key yes
it's one node because it was already a remote sign a key this is just a test net but the test net the network was
bringing up a new test map and I was reusing the old box but I'd obviously already taken off the uh the the private
vow key that was already gone somewhere into the [ __ ] never never sphere but then I wanted to create gen TX and I
was being lazy because it wasn't on my laptop so I found I put there yeah exactly I'm like I know the correct
version of the software is on the server what am I doing so I threw that I had through the through the privilege up on
that box because then it will do the correct Genty X right because you need to print valve key to be correct
so then issued it from that box and then ran this like generic cleanup script but at the top of the script errors out if
there's a thing I mean just didn't because again I I use Zed shell locally so I'm just so used to seeing like the
Red X if something if it didn't Exit Zero I see the Red X I'm like oh but then on on the servers I'm using a
different shell setup so I didn't see it and so there's there's like a bit of a comedy actually they have a comedy of
errors here with like the way I was using the the show in the on the server
yeah minor mistake in a bash script and you're like yeah that's how you tuned interesting because that's how you teach
that's exactly what I was going to say is like like we like I run a shell that like the if the Last Exit was not zero
it [ __ ] it shits everywhere like it tells you like it's red and you can tell like what the status is of the last
command that's run and I'm exhausted because I've been rewriting all of our Ansel stuff so it's
all up at our repo and all this stuff and I've been have pulling heavily from schultzy stuff and also from polka chew
stuff and like just like rewrite the way I would like I manage it right which is similar but different right which is no
big deal and um and one of the things I haven't done is I haven't rerun I
haven't re-run and reworked all my zsh stuff because I I run primarily zsh with
like all my zsh and I have like a specific um color scheme I use and like just
because it helps me when you're managing like we have I don't know 88 like 80
machines or something like that between VMS and a bare metal or something like that and and when you get in that
situation it's really easy to [ __ ] up like it's really easy to like I I like am I like right now I probably have 25
Shell stations open because you might be doing something with a you know you got three nodes of this and this and
everything else going on and like you just lose it especially in a VM structure because bare metal is a little bit easier but when you like we run VM
speaker for a bunch of stuff related to test Nets and everything else and and when you get in that situation it's really easy to forget what you're doing
and so as I'm running all these ansible scripts everything's moving to bash and I feel naked like that's exactly what it
is like you're exactly right like when you don't have that some of those catch-alls that are in there
um I think it's really easy to [ __ ] things up right and and that that like that that structure like zhls and like
showing like massive red X's when when processes fail that's not like something that is out of the box that's something
you learned from 30 years of [ __ ] other things up right um is really where it comes from
it's 10 36 in the morning like that was a good story for 10 36 in the morning I'm not I wasn't looking to rock you I I
look like it you were yeah I know you were that's okay over the [ __ ] thing he just waiting
for somebody to say something funny I'm not I'm not I'm not
my hands are on my hands