CVA Podcast: Omar Syed, Shardus

Transcript

[Music] hello hello this is your host trash ahmed ready to kick off another bite-sized episode of the crypto valley association podcast a special thank you to celebank for supporting this initiative as they pursue their mission to redefine finance for the new economy now let's discover together the growing trends of this fast-moving space and try to shed some light on the mysteries it contains today we are joined by omar said co-founder and blockchain architect of shadows on a mission to create highly scalable distributed ledger technology so let's dive straight in to uncover the hidden world of blockchain hello amar

how are you doing today i'm doing great thanks for having me straz oh it's a pleasure and how about we start things off today by hearing a little bit about how you first got into the whole blockchain world okay well so i was writing an economic paper back in 2011 related to sound money i did a search on google for digicash which was a electronic money system back in the 1990s and found out that that system never made it but then i ran into bitcoin and uh it was immediately taken by you know what the technology was behind bitcoin and i read the white paper for bitcoin and you

know having a background in computer engineering and masters in electrical engineering and you know having worked in i.t space for several decades i you know fully understood what was going on with that technology and it just like blew me away and i think i probably didn't sleep for about a week after reading a bitcoin white paper i was just searching all over the internet to figure out all i could about bitcoin and it was just you know amazing to me so yeah that's that's that was my first introduction to you know cryptocurrencies in blockchain that's incredible and i'm

sure that there are many listeners today thinking wow they wish they wish they were you back then and found out about bitcoin in 2011 that's quite uh quite incredible i i hope you you held onto whatever you were able to buy back back then well actually at that time it was like very very difficult to buy any bitcoin because i think mount exchange had just gotten started but it wasn't even a year old and um i mean the first thing we tried to do was try to mine it with my son's computer and that didn't go over well i mean even at that time you know the mining difficulty we thought was like astronomically too hard but i mean

you know we we mined for a week and it generated like point eight bitcoin but at the time the price of bitcoin was four dollars so that was like we're wasting time you know uh trying to mine this so yeah we we dropped that right away but that was a good um learning experience because it taught me the value of bitcoin and i realized that hey somebody else is having to spend some electricity and you know computer resources to actually generate these these are not like created out of thin air right so no it's uh it's very very true and and it's wonderful to to to hear and see that you're still involved within this space after i mean 10 years now and that

that you are the co-founder of charters maybe you can tell us a little bit more about this this initiative this project and maybe even potentially the origins of the name yeah sure so you know going back to like 2011 after discovering bitcoin and understanding the you know the technology behind it i immediately was taken by the fact that this is really not going to scale if if it really got popular and i remember posting in the there's a bitcointalk.org website where everybody hung out and talked about bitcoin back in the days and the core developers of bitcoin their response was basically that oh yeah don't worry about it you know moore's law

will you know get get us faster computers and more bandwidth and all that and by the time bitcoin gets popular you know the computers will be faster too um and and then you know i i having a background and having built scalable systems for uh large companies you know like yahoo and zynga and you know built uh scalable systems i could you know really see that this technology really isn't designed to be scalable and it's going to face problems in the future so i kind of bought that

vertical scaling uh thing for a while but then in uh 2015 you know the core developers started having issues where you know they realize that hey bitcoin is not scaling it's getting very popular and they need to do something about it and so we started having those conferences on how to scale bitcoin right and so that kind of you know showed me that you know there's definitely some work that needs to be done in this space and so that you know kind of led me to start thinking about how should you know how should we how would i redesign you know things if you were starting

over so that it would be much more scalable and and that's kind of what eventually led to shardis and then so like in 2016 i kind of left my consulting work that i was doing and then focused on designing shardist locked myself up in the library for a year and a half and then in 2018 we started putting together a team and then starting to build the software for shardis and believe it or not i thought i would be done in about a year and a half but it's been a lot more detail and intricate work than i ever

imagined it to be initially so we're but we're getting close so and and and so the name shardus um also kind of uh refers to the way that our network works where we're trying to use sharding technology in order to allow for scaling and achieve what's called linear scaling where as you add each node to the network you get additional transaction throughput and additional capacity that the network uh can hold and so forth so so the word chartis is kind of us been on the word sharding okay

interesting i mean and thank you thank you for for all of that backstory there i mean and and interestingly enough i mean charters operates on a different let's say protocol different proof of than uh than the other let's say cryptocurrencies out there such as bitcoin ethereum and and many others i mean we've heard a lot recently about proof of work and even let's say the migrations to proof of stake uh however you're using uh something called proof of quorum uh would you like to tell us a little bit more about that yeah yeah definitely um so a lot of people don't recognize proof of quorum but it's actually like some of the

initial consensus algorithms were really basically you can refer to them as proof of quorum a more common word that might be used is uh classical consensus algorithms so these were like originally developed like in the late 1990s but basically it means that you know you have a known set of nodes that are voting on something that they want to you know agree on and if you you need to get at least you know some threshold of nodes to agree on that before everybody says okay this is accepted so that's just proof of quorum and and these were

you know developed there's algorithms like paxos and raft and pbft so those are the original foundations for this okay wonderful um great and i mean what you do you oppose any of the other kind of like proof-of-work proof-of-stake i mean do you believe they're viable as well or well so yeah definitely um so those are really needed in a scenario where you have kind of an unbounded number of nodes in your network and you're not able to know what other nodes are participating in the consensus right so that's where you need proof of work and proof of stake in you know and and that was

satoshi's biggest breakthrough was being able to achieve consensus in the network where there's unbounded number of nodes each node doesn't know anything about what other nodes are participating in that consensus and you're still able to achieve consensus among all these nodes so yeah there's a need for that as well but with chardis because we have a network where every node kind of has created its own node id and the nodes know which other nodes are in the shard and consensus mostly happens just within the shard then you know it's a well-defined

group of nodes although the total network size could be unbounded but the shards will typically be limited in size even for like a public network you probably only need about like 200 or 300 nodes in a shard um and anything beyond that you're just getting a lot of redundancy without you know getting more throughput okay okay interesting and i mean following from from that point you can tell us a little bit more of the importance of some of the things that you're working on within charters based on for example uh you mentioned linear scaling but then also the ability to achieve auto scaling uh would you talk to us about that yeah so first um you know

for anybody that might know not know what linear scaling is that basically means that as you add another node to the network you get you know a small amount of increase in your transaction throughput and so two quarters ago we actually demonstrated a network running on aws across like 20 different data centers and we started with a network that had 20 nodes in it and we showed that it could get 100 transactions per second but then when we increased the tps to like 150 the network wasn't able to handle that it was you know dropping some of the transactions

then we increased that network size to 200 nodes and we were easily able to get a thousand transactions per second to that network so that demonstrates that as you add more nodes to the network uh you're able to get more transaction throughput now auto scaling is a feature that i don't think any i haven't even heard of any other project uh thinking about doing that but this is this really comes from like production enterprise networks like for example if you have a network that's handling a really popular mmo game or something like that or even a mobile app you might have

a lot of users during the daytime but not as many users during the night time and so for those kind of systems you know when we're when i was working at zynga and other mobile game development companies we would design our back ends so that they would auto scale so that as the load increases uh more nodes would be added to the network and then in the evenings as there weren't as many players our back end would shrink back down and there'd be a lot less nodes and and that really kind of helps you save on the cost of your network i mean of course it's important to these companies but i think eventually these kind of things will also become important to decentralized networks so that's what um auto scaling refers to

okay wonderful i mean thank you thank you so much for for that context and um i mean i want to touch on on one point here maybe it's a little bit of a sensitive topic uh as per say and it's discussed quite quite quite a bit actually with some of the crashing uh the network crashing happening but there is quite a large dependency within the cryptocurrency and blockchain infrastructure space on aws and running servers on it what do you have to say uh about that omar yeah so you know um when we're testing our networks we use aws um because it makes it really easy to spin up a large network and and do the tests um but in actuality you know the

shortest software is designed so that other developers can use this as a starting point and then add their own application layer to that and launch their own network and now when you launch that network it's really up to those developers to decide how they want to launch it but ideally what you would want to do is provide your code so that home users could download that and run it on their computers at home or you know run it from a vps server on any old you know web hosting company and so forth so and and we've even demonstrated a node being able to run like on a raspberry pi and join

a network from behind a home firewall so that's you know the the long-term vision is that you should be able to launch these networks and have like end users be able to participate in these networks and that's what's going to allow these networks to become a lot more decentralized i mean not only currently do we have a lot of dependency on aws but um even networks like ethereum and bitcoin which have like say 10 000 plus nodes in the network um most of those nodes are not really producing blocks if you look at like what nodes are actually producing blocks that's probably less than a dozen

um so so that kind of loses decentralization because of that because you know you're kind of limited to how how many nodes are really actually producing block so what you know the way shardis is designed uh we we you know would really want a lot of different entities being able to participate and as a network grows you get more entities participating in and that gives you a lot more decentralization and security okay no i see it thank you so maybe come come back to to to shadows as per say and maybe even the first application that uses shadows as a software which is called libernus uh maybe you would like uh to

tell us about that sounds very liberating yeah yeah definitely so you know as i said uh shardis is software that others can use to build their own decentralized networks and launch their own networks with but actually we wanted to create some networks that we would would have liked to see and that's really why kind of shardist got started too because you know we felt that that software wasn't there and if if we're going to build a software let's build it in a way that anybody can use it and so one of the applications that we want to launch which chart is on our own is is librets and really it's

it's just a simple coin that hopefully will be more stable than a lot of the other cryptocurrencies out there but still you know have some aspect of appreciation to it and then the other thing that we want to do with that is allow users to be able to communicate through this network as well right so would have like end-to-end person-to-person communication built in and and right now a lot of our communication software is things like uh whatsapp and telegram and so forth and even though these companies say that it's end to end encrypted you never know if there's a back door to it and and at the end of

the day it's just one central company that's handling all of that data ideally it would be nice to have you know this huge decentralized network with 10 000 nodes you know handling that and so there's no point of failure to that okay wonderful and you you also have a a shortest token i mean the ult how about its utility i mean what utility deserve yeah so it's kind of interesting when we you know launched shardis we decided that we don't want to do like an ico or anything and we decided that the shortest tokens would only be released and come into circulation whenever someone did some

work for chardis or provided some service and generated an expense that chartist needed to pay and so that's how the shortest tokens are getting uh put into circulation and the use for the shortest tokens will be that when somebody wants to create a network and launch their own decentralized network using shardas we ask that if it's a public network that they give some of the coins on their network back to the shortest token holders so if you're as sharp as token holders you know you might be invited to receive some like let's say decentralized reddit

coins or for you know for example if somebody was to launch something like a decentralized reddit with charges and and you know as as more and more applications get created and launched with shardis those shortest token holders would receive coins on various different networks and and the other aspect is that there may be some enterprise use to shardis where um you know some group of companies may want to use shardis behind a firewall and in that scenario what you can do is that you can buy the shardest tokens and submit it to our smart contract and it has a function on our smart

contract called purchase license and that will burn those tokens and issue the sender a license token and so that becomes like proof on the blockchain that you purchased a license for the charter software and now you can use that in a private setting that's yeah it's it's kind of a neat and unique i think i haven't seen any other project use a token for like software licensing yet but i think we're probably the first one interesting well and i mean something we haven't really uh talked about but uh come on where where's the the company uh based out of where is it like the first market that you're looking to to venture

into i mean of course it's very kind of decentralized distributed like like most other companies within the space are but do you have a specific target first to start off with well i mean we're we're based out of dallas although we're not anymore because since coveted everybody's just been working remotely and um i'm right now in tampa florida but um yeah so you know our target market really are you know applications that we probably haven't even seen created yet and so you know when you have something like shardis available and people can start thinking that hey i don't have to start from scratch if i want to build my own decentralized

network i can you know just take shardis and then create my application layer on top of that and this will be like scalable as much as you know the number of nodes that the network has that kind of starts to change people's way of thinking and even thinking of what's possible and so some of the applications you know i think we haven't really even seen yet we have seen some things like that like if you remember like steemit which was like a social network based around blogging and writing articles but based on a blockchain right but that software underneath that was like really really

tightly coupled with that application and so like nobody could really use that software to like say create a decentralized reddit or something else from it right so that's where you know with shardis we're kind of making the application layer very separate than the technology layer and and so with this kind of software available there's a lot of projects that i think who would find you know this useful for them to be able to launch large networks and networks that need a lot of throughput okay wonderful well great i mean truly actually inspiring as a as well i mean maybe we kind of goes off into maybe some more uh inspiring notes hopefully

uh whereby you can tell us a little bit we believe this space is heading the space the blockchain space as a whole yeah so i think we're in a really interesting time right now and we're very very early on and i and i really feel that uh the whole blockchain space may take like another 20 30 years to mature and i really see it kind of evolving like how the database feel did databases have been around since the 1970s but even today we still see like innovation in that space right like about 10 years ago we had uh no sql database today's and and then we've had like graph databases

and so forth so there's a lot of innovation that i think that's still ahead of us but and and in another perspective if you think about what current blockchain platforms are really supporting decentralized applications it's really things like ethereum and finance smart chain and polygon and you know new ones coming up like solana and algorithm and polka dot and things like that but i i i tend to look at these as the way that time sharing computers were in the early days of computers where because it was really hard to build a computer you just didn't build as many

computers and everybody just piggybacked off the computers that were there and wrote their applications to run on these time sharing systems but then as computers got more accessible and you know easier to build and anybody could have access to their own computer you really started to see a change in mentality of the applications and and what was possible and so i think that's what's going to happen in the blockchain space where once uh software like shardus is available and people see that hey you know what i don't have to start from you know ground zero and build my whole decentralized network from scratch and think about how to do the peer-to-peer communications how to do the consensus how to keep all these nodes synchronized

and how to deal with nodes joining and leaving all these things are taken care of for you then you just have to focus on building the application layer well now it becomes like hey you know any e-commerce website that i can think of you know like maybe even like a uber or airbnb or whatever now i can think of doing that and with a decentralized network right and so it kind of really opens up the doors to the possibilities and so that's kind of where i see blockchain technology going wow wonderful well well thank you so much hopefully it doesn't take 20 to 30 years to mature i'll definitely be too old by then to deal with the volatility

but it's a great vision thank you thank you so much and thank you for for joining us today omar i mean do you have any final words for our listeners today yeah i mean you can you know check out chardis.com and we every quarter we do a quarterly presentation to demonstrate uh where our software is at and and the capabilities and we've been doing this ever since the project started so you can join us for that which that'll be coming up on october 16th that's our next quarterly presentation we'll have things posted on the shortest website there's also a shortest telegram group that you can join and and several other social media uh

links there on the shortest website wonderful well make sure everyone to tune in to the quarterly update then and if you enjoyed this conversation feel free to check out more on our website cryptovalue.swiss where we host plenty of events educational material and even provide information on how you can join our growing community thanks again omar for joining us hey thanks for having me charles a pleasure stay tuned stay safe and until next time from the crypto valley in switzerland bye [Music]