Group 1 - Judging

okay recording is good oh [Music] okay [Music] stand by one second double checking our recording here we are recording on zoom so we're good but i double check the stream here all right i'll kick it off in uh less than a minute it's exciting you all right welcome everybody to uh our kickoff for hackerfest judging today is day one and uh i am kartik one of the co-founders of eve global and we're super excited for the last week of wrapping up hack fest so for those of you who don't know for the past four weeks we've had 470 hackers who've come in from over 50 countries and 19 different time zones workloading project that uses tools and technologies from the ethereum and the falcon ecosystem and we've seen an amazing set of people work on really interesting projects and got support from a lot of our amazing sponsors that have kind of come out and helped these teams succeed in terms of making their projects a reality and as a few days ago we've got up all the submissions for these teams and this week's we kick off what the projects are for everybody else to see and i'm super proud to announce that after the past four weeks of hard work from our teams we've had 132 projects being submitted by these 470 hackers so we're super excited to see what everybody has done and uh we kick off what that demos are going to look like uh today and uh i'm super excited for everybody who's going to be looking at this recording later on and all of our hackers do let's see what everybody else has done so i want to quickly go into logistics of how this call is going to go a lot of our teams are here ready to be uh to be judged and present what they've done and the logistics are going to be fairly straightforward we have about 13 teams today who are going to be demoing to our judges and each team will have four minutes for their demo and four minutes for a q a session from our judges and to kind of minimize any eb issues uh we've asked every team to record their demo uh so they will just uh play the demo and we'll go into the q a section after so we can keep everything on time and uh and uh for our judges and everybody else who's going to be watching this video a quick overview on how everything was set up for hack fest uh each team could have up to five members and a lot of the parts we're going to see were done in teams and some of them were also done individually and it was all also allowed and as part of the rules of this event all code that you're going to see as part of the demos was written during the course of the last four weeks during the hackathon and the only criteria was that the projects we're going to see today must incorporate technologies and tools from the protocol labs and the ethereum ecosystem so a lot of the projects that we have for this event are really creative uses of decentralized storage and smart contracts and the criteria for how the judging is going to go is uh is going to be as follows we have five categories that our judges are going to be rating all these projects um one in the categories are how technical a project is how original the project is uh the practical uh applications of the submission uh how easy it is to use whether it's a ui ux thing or as a developer tool and they general catch all that we like to call law factor that let's just sort of understand and highlight any outliers so before i move on to kicking off the first ever [Music] group for hockey fest i want to make sure that i kind of high level this is not a competition uh these are very much feedback sessions and as a result the hackers are here to learn we have a lot of newcomers that have come on uh over the past month to learn about what they can do with ethereum and fault coin uh their hackers are there to share their excitement they've been working super hard and tirelessly over the past few weeks we've been doing constant check-ins with them we've been helping them connect to mentors we've given them technical help on many things and we've done a lot of talks on uh what's actually possible in both of these ecosystems and today is the day they talk about what they've learned and kind of how they've applied those learnings into a project that they've been working on and the judges are here to give feedback to these teams the judges themselves are also super excited to uh to see what everybody's done and as a result this serves as a a window of opportunity for us to learn from what's possible in this ecosystem and also how we can actually improve and continue working on what we uh we would like to see more and and as like a very kind of uh thing i want to press on not everybody's trying to make them a business a lot of these are just fun projects that everybody is uh using and working on to show their creativity uh and there is no explicit goal for any of them to become a company or uh anything beyond uh showing what the applications are we want to make sure that they stay fun in an exploratory in nature so the schedule of how the next two hours are going to go is fairly straightforward we have 13 teams today these are the teams that are going to be going up and presenting and doing the hard job are on three judges for uh for today and and i want to kind of welcome them um and that's all from uh from falcone we have matthew from the topical domains and and canon from the ethereum foundation and uh both all three of them will uh we'll be kind of here listening to all of our our uh our hackers giving them feedback and uh and sharing their assignment with them so with that i want to kick off with our very first demo for hack fs and like to call out our next team and uh let's have them join the the chat and now please welcome watchdog now to kick off their demo hi i'm mauro and this is what i built for hackfest it's called watchtalk dao and it's an attempt to bring accountability into governments and public institutions you can't optimize a system that you don't closely measure so this dow can coordinate citizens around the world into gathering analyzing and acting on data to move governments into better representing our interests okay so the first prototype i built is focused on long-term storage of government data in a way that it can't be tampered with i come from argentina county with a long history of instability and corruption and the fact that our previous administration made a big push to open up its data does not mean the current one or next ones are not going to stop storing it providing it or even tamper with the data so what i did is just grab pretty much all the data sets that the government is exposing and just tutorial on file coin i have here the full data set and then i stored an index file which is just a json file with all the ipfs hashes and the file names that's important because you let's say in 20 years from now you're a researcher or an activist and you come upon this data set in particular here and you want to know that it wasn't temporary so you just need to grab the hash check that it's included on the on the index file it's there just check that this index file is actually the one stored on filecoin that's quite easy and then you're good to go like you just know that that wasn't tempered with it's focused on police accountability it's through obedience video evidence of police abuse of force in a long-term archive right now these videos are being stored on twitter and facebook and there's no reason to believe that they were holding for any particular amount of time so the cool thing about the archive is that it'll be built by the centralized network of activists around the world the way it works is let's say you want to start archiving videos what you need to do is you need to submit every a proposal using your twitter id and the reason should be a better reason than that one and that generates a boat you know dao i'll just take just one second i'll show in the meantime this is the aragon dal i built a little app that allows any doubt to curate their own list and this dao is using it to create a list of approved twitter accounts so once you once you become an archiver that's distance action confirmed so you should be able to see there so once you become an archiver what you can do is you can come here and if you see a relevant video that you want to archive you just need to mention or both bought with let's call watch dodge dao my computer is acting up and just hashtag archive and the bot will do everything it will archive it on both uh falcoin and ipfs i'm currently using flick storage for this and yeah once that's done you'll get a notification and [Music] and if you go to the archive here sorry that's the notification well i got on my phone it's not showing up on the screen yeah there and if you go to the end of the archive you should be able to see here there the archive video so that's pretty much it you can reach the site on both ens and that crypto and please feel free to propose an account as an archiver or yourself to join the dial thanks to filecoin and all the event organizers as this has been really fun to build bye can you hear me tomorrow any questions from the judges yeah i'll start um i i think i i missed it but what what's uh how do you how do you get people to um i guess uh store more right like is there an incentivized mechanism there where you want to further you know get people to actually store some of this because i feel like if it's on twitter right you're you put yourself at risk um and archiving some of these things and now you're uh the worst thing is to make yourself a an animated estate because i'm trying to you know make sure i can hold the police accountable hey and i understand um twitter is just one of the sources it's like the only one i did for this hackathon because there was no time but i could do a simple bot for reddit for example or anything to store data and then the the archivers could use pseudonymous accounts they don't need to use their main account um and yeah this is things that people are already doing like people are already archiving things on twitter it's just that for example one of the videos i archived um two weeks after it was posted on twitter it was no longer there so just in case you know we need this for long term is there uh i have a question is there um a way for people they can search through the archives i was because you know if you archive a bunch of stuff and then you want to come back like three years later um how would you how would you find the files i mean i'm sure there is i just didn't see the demo yeah right now there isn't again for a matter of time but that there will be yeah again my country 40 years ago had a military coup and a bunch of people disappeared and right now the the current democratic governments are looking for the evidence and well back then there was any but we can't really expect twitter to store everything for 40 years reliably so this would be one of those use cases i think it's a great idea uh which country by the way i'm just curious where you're from argentina oh argentina okay sorry yeah cool yeah uh just a couple things more uh feedback than anything else i think uh i think you present a very compelling use case certainly i think the point about uh just because people are making uh you know because if if there's an emphasis on uh trying to acquire and preserve this information today that doesn't mean it'll stick around forever we know that from history and so i find that uh i think that's a that's a very good point um i want to echo uh the idea about uh having some way to search and kind of filter uh uh archive information i think that'll that combined with uh adding more sources to this stream not just twitter right now um both of which i'm sure are already on your radar are uh kind of probably your highest priority next steps um to to make this usable and impactful uh but i think it's a great idea i think uh i think it's a good example of taking something that's very clearly aligned with kind of uh the the decentralized web and uh decentralized storage missions and um making it happen uh effectively so good job thanks yeah i just want to say i really like what it is that you're building here so my mind is kind of swimming with how this could work because i'm just imagining like i think twitter is actually fine like it's probably enough to get a lot of it because we see videos in the us right now about police violence and and it's uh i don't know i think it's a great idea and it's super uh super straightforward um i think just gotta get more people to use it another question i had is um how right now all the data is being stored on fleek uh what would you i know you obviously didn't have time to do this but what what's your ideas long term for how to ensure that the data stays up you know if if fleek wasn't there for instance um have you thought through how maybe that would work yeah the first thing i built was the police archive and use flick because i didn't know the tech that much that's why the second thing i built which is the open golf data i use just like a lotus note directly and if i were to build this again what i'm gonna do uh i think i really like the whole textile uh stack i think that's really good so maybe have a couple different oracles like different institutions that hold the same data and just use um like the power gates and and sort of thing directly not rely on flick and would there be a way for the users to actually pay because the way that i see it is like if i see something i want to store am i paying for it is somebody else paying for it um what how do you think it should be paid for the the initial way i thought about it it's just kind of a non-profit the good thing is that all this data shouldn't be that much data even if you store all the let's say the police case all the police abuse cases in the world that's probably all the videos in a year are going to be like one minute of the videos that get uploaded to youtube thankfully there are not that many videos probably so it shouldn't be and storage is is fairly cheap hopefully it will get cheaper with falcon so it shouldn't be that much money um so yeah maybe at the beginning as a non-profit and maybe afterwards the dow doesn't need to always be a non-profit maybe it could do some like for-profit things kind of dangerous there but yeah we should we would have to see great okay thank you so much watchdog dao um we're gonna move on to the next presentation um thanks again um okay up next we have uh codigo hub um so i believe that you're all in the call now and you're sharing your screen so please start your presentation and then we'll move into q and a after that hi everyone we are team codico and we build codigo the iot portal for firmware distribution management collaboration with a team of five people and we built it from scratch the idea was conceptualized at the university of edinburgh and we reshaped it into codigo hub a shop window for users to manage their iot devices while incentivizing developers to contribute firmware for the billions of iot devices currently connected to the internet it's basically a secure firmware distribution service for the iot in cadigo hub users access firmware and incentivize developers to contribute while a reputation system ensures trust between them and the validation system enforces policy rules but let me show you welcome to coding i see a map with all my devices and their active status plus i see a summary of my activity on the hub such as number of downloads uploads and bound its claim i have a new device running on the background called humidity sensor so let's add that the hub automatically picks up the active stylus of the device using mqtt tip and assume that my new device has a wireless module that i'm looking for firmware i can search the repository and i find something relevant i can check the source code i can check the firewall the firmware details and i can also check the community method such as number of likes number of downloads reviews i can also check the profile of the developer i see that the community tests them they have some contributions the number of likes so i'm happy with them i can either download the firmware or my pc but i can also just directly send it to the device over there the firmware will be fetched from filecoin and push to the device over mqtt that's nice i'm happy with the firmware so let's leave a like i can also leave a comment or i can donate to the developer however assume that i didn't find the the firmware i was looking for i can create a bounty and set an amount of ether on it to draw the attention from the from the developer community on the hub the bounty will be registered with the bounty smart contract the bounty has been registered and now i'm waiting for a developer to notice now i'm back as a developer and i want to make some money on the hubble let's check out this bounty i can check out the description the device and i can see that it's open and for 0.01 s so let's try to steal it the bounty smart contract has validation rules that will allow me to do that so let's actually make a contribution we upload both the binary and the source code to make collaboration easier to increase transparency and allow users to directly deploy firmware on their devices the new firmware will be registered with the smart contract for firmware considering and will also be kept on filecoin my contribution has been registered so now i can go take the bounty honestly the bank is for 0.1 so i should see my balance increase by 0.1 i have received the bounty and i can check my historical earnings on the platform by a duration or by a firmware finally the hub is all about building up your profile as a developer by making good quality contributions such that users will like your like and download your contributions and you will build up your community score and finally this way you will make more on the platform and your ending your earnings will grow adiva hub is a reactor deployed on flick we use three box user profiles ethereum smart contracts for tracking bound is used in firmware and file coin for our storage needs finally we use mgtt to communicate with the devices this is our tech stack finally we would like to implement the graph and servable domains to visualize user history in the network and to link crypto domains with user profiles also we'd like to develop sticker validation using different metrics and chanyak a research project validating the link between source code and binaries thank you for your time please check it out and give us any questions i think that's us let me stop sharing nice um quick question to start um what advantages are you getting from uh like is the decentralized aspect of it whether that's ethereum or blockchain primarily to provide a marketplace for the bounties and you know the usual single point of failure the absence of single point failure advantage or is there more i think that's mainly uh to be honest with you um yeah do you have any examples of some firmware that's been uploaded we've made some example contributions for arduino codes for a wireless firmware for a wireless module for example and it's we started this project on the hackathon so all the contributions is what we've made and then who would be like the who do you think you would target to use this so like you said arduino do you think it's people doing um like small at-home uh toy devices i'm curious so there's a couple of um people we'd hoped to get quite well a number of different stakeholders ultimately would initially be people who are making toy devices at home but manufacturers are something we're also interested in get involved in the process because we're looking to you know have a distribution service that can work for a lot of devices so we need that kind of scale and support from them as well also and not only not only single users but for example uh companies or people who have deployments of devices and they want to be running a global single firmware on all their devices that would be a good use case for them because they could manage all their devices and also deploy make sure that they know what they deploy and deploy a single firmware to all their devices easily just a comment one of the first things i thought was man it would be great to have a place where i'm storing all the versions of firmware because every now and then you're looking for something that's like really old you can't find and so like when i saw you guys presenting i was i immediately thought that that's what this was for i was like oh um so i just thought i'd throw that out there because when they they update it and then my thing doesn't work and then i gotta go find the older one um so i could actually see that also being able to archive the firmware for forever is also interesting uh yeah kind of related to that i'm just wondering what are uh what are your uh essentially future improvements and enhancements like obviously there's work to be done in terms of getting it out there to more people of both creating varieties and developing firmware but do you consider the product itself to be essentially done or what more would you like to do on it so from our very verification is one of the big tasks that we're kind of looking to do next and to build that into the system so that's not currently being implemented at the moment but we've got a few ideas of how to do that based on existing systems and existing solutions so we'd look to incorporate and there are tools to be able to handle that additionally we would also like apart from software verification to build a static verification verification rules for example bounties to make sure that they got to for a good purpose nobody a developer can just steal it by not committing nonsense firmware or something using a firmware reputation metrics user reputation metrics and [Music] yeah and that note is the bounty currently uh confirmed by the creator of the bounty just saying yes this firmware is what i wanted or is there more at the moment it's confirmed only on the blockchain so if there is a firmware registered for the for the device type being requested after the after the boundary has been registered then that makes it eligible but we would like to improve from that awesome well thank you so much uh team codigo uh for your awesome presentation uh we're gonna move on to the next team uh who are now in the chat um this is uh project koyo so i'll please invite you to share your screen and start your video presentation hello my name is yalda and i'm going to talk about a new platform that started up in hack fs called koyo koyo is a decentralized network for the future of collaboration and ownership it's the brainchild of myself and hackathon partner peter phillips right now we have a software consultancy together called the koyo collective we've both worked in the dao space for over two years and through our experience we have come up with a few goals that we are ultimately aiming to achieve in the long term we need collaboration and governance tools that eventually take all dimensions of privacy into consideration and which are ultimately community owned and governed and within these tools we need to enable safe spaces for members of a community so dissenting opinions are not penalized we decided that the goal of enabling safe spaces would be a good place to start at haccfs one of the major features we have developed is the ability to create an organization with one person one vote proposals where votes are anonymous we use semaphore for zero knowledge anonymous voting three box for profiles messages and threads ipfs for decentralized storage and hedgehog as a client-side ethereum wallet so now i'm going to walk through a demo that was recorded in our local environment we do have a version of the product live on coio.network although we're still testing it and fine-tuning it right now we are chewy and chewy is going to create a new thread so he populates the title and the message and what this is doing is using three box threads in the back end to generate this discussion and now what i'm going to do is switch over to another account cap so cat's going to add a reply to this thread that chewie just created and then let's flip back to chewy chewy sees cat clicks on cat sees cat's profile and clicks the follow icon and then whenever we foot back to cat cat gets a notification that says that chewie has started following them now we're going to go back to being chewy and chewy's adding cat to a new organization and now he's going to generate a voting key that's going to prove that he's permitted to vote while keeping his identity hidden now that he's done that he can create a proposal this is going to create the voting object using the sum before contract and now you see the active vote there so we're going to flip back into cat and cat is going to vote yes and after it's confirmed we're going to see that the quorum changes from 50 to 100 percent we have a lot more ideas that we're looking forward to incorporating into koyo and we will continue developing after hack-a-fest thank you for your time and attention great thank you kodio um questions comments from the judges and you can disable your video share now if you want quick question to start if you don't mind uh what is that for can you tell me a bit more about it i've never heard of it before sorry what is what for sam before the the voting platform that i think you used oh um it's a it's a protocol uh for i guess it's like they call it a zero knowledge gadget uh peter do you want to talk more about that or yeah so it's um it's doing serial knowledge proofs on uh on ethereum um so it's it's basically you have um all these identity commitments that get added to a merkle tree and um and then on the front end you're you'll do a proof that um that proves that you're a member of the group without actually um without showing what your identity is um so it's able to is you're able to show your member without revealing your identity when you vote gotcha so you folks use that yep to essentially prove that you should be allowed to cast a ballot yeah exactly so yeah sorry yeah and i think for like the demo maybe like the whole story wasn't there for why you might want to have this anonymous voting because for this use case he also had three box profiles and it was two people and they're creating this organization but we think of it like in a you have a larger organization maybe you have shared resources and someone within the organization they are having a proposal to fund themselves and you might not get people participate in votes because they don't want to like hurt the other person's feeling by voting no so with anonymous voting you can vote no on for example funding proposals and you won't hurt people's feelings and people won't build up resentment that oh you voted no on my proposal so whenever you create a proposal i'm going to vote no for you um i think i'm part of like moloch now so sometimes i think of this i'm like i feel like people aren't saying anything or voting because they don't want to hurt people's feelings so we're really excited to use an experiment with some before so this is this is super interesting and i guess my question thrown back to you just in what you just said is do you think this would help to encourage molecule i guess members or other dows to engage deeper and maybe not even just voting but also trying to get more proposals in don't really i i think it will right now um some afford doesn't work for token weighted decision making um i believe that there are some enhancements that can be made to the protocol to to enable that but i feel like it would it would really help a lot um and it opens up you know different use cases like uh like local um like more like if you want to like build a city and you want like people want to like vote on policy decisions just so you can have your privacy on uh votes in general so i think that there are a lot of different use cases but i think this would be a really good one for dows in general too thank you cool yeah i guess i'll just broadly say uh i like the idea i think there's uh lots of scopes there uh and i think you guys have some some cool directions to go in um also a much less important note i really like your aesthetic i like the colors in the video i like your logo a lot it's all a good look thank you yeah i'm feeling this cool 80s vibe [Laughter] nice good job okay uh if there's no other questions uh from the judges i think we'll move on to the next project uh thank you team coyote it's a great presentation thank you okay so the next project to come up uh is uh team certi i'm gonna play the video um just to avoid any technical issues here so let me just queue that up hi um we are building sky storage on file coin next we are of two people rajat and sunny though one is front and one is black and engineer next so we are we are making video player thing in which we are using textile powergate to interact with filecoin and ipfs and we are using it plus ipfs concept to save the video file details uh in the hrm and we are using flick for host the ipfs website okay so next so initially you need to register and then you can add file and then you can access file so so next next we will see the demo okay so this is our demo sky storage here a new new user needs to register click here to register once during the registration filecoin will uh falcon powergate will generate the token and falcon address for the metamask address once it is registered you can see the file coin address and metamask address all the details in the right side sorry for the leg yeah successful register successful register you can see in the right side my thomas address icon address and balance okay now we will add the uh add file currently that there is no any video user can change the storage config of the video file like hold storage or cold storage where the user want to store okay file coin uh if you enable this file coin that then enable will be true and for file coin okay now we will add the okay yeah so we are using small dot uh mp3 for video name okay and once it is a mac storage deal after after uploading the video in falcon the metamouse transaction happen for this saving the video details in the uh ethereum smart contract okay so we are using each plus ipfs concept once it is upload you can see over here a video and and below the video you can see ipfs and falcon at both storage this video is stored okay so you you can play this video over here so we are giving video functionality also here okay now i all i have another account also where i have already uploaded video and i prefer some file coin in account eight i'll show you okay let me refresh it okay once it is refreshed i already uploaded some videos like three to four videos in ips and falcon already yeah you can see so so so this was your project we and this project we have hosted in flick hosting service so we are we are using flick also yeah after demo time we will go to the next slide sorry yeah so how it's made we are using slide slide dot host and the text help target to interact with file coin and ipfs and we and user can also change the storage config for video file before the upload and we are using ethereum uh smart contact to save the details and click dot co for uh uh hosting the website and next uh next.

js and material url for uh building the ui okay next uh what do what do in future we we want to make encryption in a video player so we can securely share this video and we also taking part in apollo so we will also do more hack in that that event next so this is the uh flick hosting domain so here you can see our website also okay yeah so thanks thank you very much and if you have any question we love to hear okay thank you bye-bye great thank you team circuit so i'll turn over the judges for any questions okay i have a question who's the user are you trying to get like youtube people to publish their videos on ipfs or what do you think is the use case for this yeah so currently we are we want to make like video players so uh like we want to make encryption video player where share the video or like and uh and can only video can play by the user who uploaded like uh he he needs to add the password and once the password is done so it will decrypt it and uh taken that they can see their video uh so you're imagining kind of like sub stack for video on the decentralized web do you know what sub stack is no uh i didn't know okay yeah so um so as a content producer would i upload a video on here and then i could maybe sell it to people so they could have access to that video and then they would need to get a token in order to see it i'm just trying to understand the use yeah so yeah we want to make like this also like we can we can store our private video or any video which we want to uh store in the file corner ipfs as an encryption model and so and also we can uh share the video which you want to share uh and quickly on like our our favorite okay and then i'm gonna steal this question what are some of the next things you would want to build like what are the other features you think are missing that you'd want to build for the product yeah um next thing yeah uh currently is uh like we want to uh move this power gate to a textile bucket so using buckets and i think flick flick space demand is also providing our authentication encryption so we we want to move on to that using apollo events okay thanks one question do you know uh are there uh other similar things that already exist today do you know if uh kind of did you like explore the space for i guess ipf has hosted video players before you started on this project i'm responding yeah um maybe some existing project uh like but we didn't see encryption like that so uh so so we have explored the uh sheet of the uh hack face like of 100 ideas in which that we we get this problem statement so we started to on developing on this okay uh there's no for the questions we'll move on to the next team uh thank you team for the great presentation uh and we'll call up the next team uh which is uh decentramall um i'm gonna play your video for you as requested so let me just queue that up and we'll get started hi we're tmd centro mall and we're building an unstoppable shopping destination this is our team we are solving a problem that current digital real estate holders are facing which is that their current real estate is hard to monetize and it's hard for them to build on top of it and and or rent it and operate the space so we're making it easy for them to rent and operate it we're building on top of textile ipfs league and others we want you to own part of the centralized web make it easy for you to buy and sell the space we also want you to take ownership by governing the mall uploading your curated product catalog and incentivize shoppers to keep shopping we'll jump right into the demo so this is the mall and then we click on anything and we go here uh we will have your own space showing up on the landing page if you go to your profile this is the space that you have bought it's available for rent so you can just rent space you can rent it out yourself or you can rent out others here we're gonna add a new clothing store and just like that you can add a new storefront and link it back to your online shop here we can see that the space is rented and links back to the store [Music] you can also search for different stores or categories so you search for nike choose nike and research for that these things are clickable too [Music] and it's also really easy to buy the prices determined by the bonding curve and here you have bought your own store [Music] today's ens name awesome thanks to central i'll turn over the judges to ask any questions great music obviously that that presentation had the most prep of anything we've seen so far um i am worried that i maybe didn't get it though uh what was what is i guess the decentral part of the central mall and why all right i'll i'll take this one um so a lot of people get got really excited about uh buying digital real estate and then only to find out like how difficult it is to build it you have to hire an external agency to build like a 3d space and you know just to be able to display your digital art or maybe uh try to sell your your physical products like socks or t-shirts or shoes and uh you know i come i come from e-commerce as well and i'd like to be able to display my products in a decentralized fashion basically on an unstoppable uh a place and and just be able to like really build it as easy as using like woocommerce on wordpress or something like that if i can also own the marketplace by having some say in the governance of the mall itself then so much better yep i find that i found that last bit persuasive uh i guess my concern would be uh you know i think econ obviously you have more for background than i do but i guess i worry that e-commerce already has challenges with people you know wanting to go to actual stores and they're like do i want to do this online at all um jumping from that to you know getting them onto the the still fledgling b where like might be a child it might be yeah that limits you but i think like corona kind of fixed that a little bit you know not a lot of people want to go to stores anymore so a lot of people shopping online but i think also a bigger issue is like if i want to put my products on on a marketplace like i have no say what's on the front page of that marketplace like it's up to them to decide like what they're going gonna put on um whereas like if you own them all uh you know you could kind of decide you know if the mall gets a lot of traffic you could kind of make those calls right like who gets on the front page or like how it's all distributed and like how to kind of so i'm i'm a fan uh the i'm curious about how the governance works this reminds me of the million dollar homepage uh and then it's like crosstalk the million dollar homepage plus decentral land like if i was gonna pick kind of how you're doing it here i do like the idea that i could buy my way to the front page so um yeah but i'm interested about governance so who's the mall owner like who's the person who's gonna make sure the traffic comes to the site because i understand like is it the way i get is people like buying spots on the homepage to advertise they're good right and and so they can just do that directly on the smart contract um who's incentivized to make sure that people get to that page i'm curious about the governance aspect sure like i think you know just like any um you know like whenever you buy a house and a compound or or you know if you buy space and like a on a retail street like there's there should be some community kind of setting and like making those decisions and wanting to hire like an external agency to do the advertising and so forth i mean ideally anybody who's getting spaces to actually do some some retail work right so you'd want to do part of the promotion yourself to kind of advertise your mo your space on the mall but as a mall itself i think the governance you know i mean we're still early stages it's just a fun experiment we had tons of fun working on it and i think the idea would be to you know everybody to be able to call the shot and say okay well we need to hire this you know this particular uh seo guy or you know this particular um you know advertising person to help us get more traffic so that's that's the that would be the end phase of like everybody has control over those questions yeah yeah so just a comment it's like it feels like the thing that you're actually experimenting here with is like governance for online stores um and i mean i think that's interesting obviously there's a lot more to it but i think it was a great job for the hackathon thank you for giving me something to think on any other questions from the judges or even questions from the hackers to the judges uh about you know what to do next to the project i i would say you need to put this up on uh not crypto so please reach out to someone on the unsoftball team so we get you decentralmall.crypto so we can have this up here so we can uh show it off to our community awesome thanks but now do you have any questions uh questions for the judges at all um not nothing that i can think of i was i wasn't expecting that um well thank you for uh your words uh yeah we'll consider about the crypto actually because we are using flick for now thanks guys great okay thanks so much team essentials from all um so we're gonna bring up the next team uh that is uh himsheet uh so himsheet i'll invite you to turn on your videos share your screen and begin your presentation [Music] let's start by creating a song book i'll call this one hymns now i'm going to add some songs and i'll just paste the text in here and then let's add a lot more songs these songs are stored in an indexeddb this makes it really easy for us to do a full text search next i'm going to share this songbook with the device i'm going to use as the controller when we go out to worship and yes that qr code was an ipfs hash next i'm going to create a channel and a channel is a group of people who see the same content at the same time and you've guessed it this is ipfs pub sub and i want to share this channel with the people i'm hoping will come and worship with me so this time i'm going to use the url i'm going to copy and paste it into an email and send it out some friends and say let's go up and worship on colton hill there may be some bystanders who want to join in so i'm going to print out a poster as well and once they scan this it'll drop them straight into the lyrics so the idea of him sheet is that one phone or a handful of phones can control what a larger number of phones see on their screen full screen so this is me grabbing every device i can find and trying to get them to display the same lyrics at the same time the laptop in the middle is controlling what the other screens see but it's not just laptops and phones that have browsers you're starting to find browsers embedded in tvs and also in smart projectors and so this is a cheap battery power projector and it's got android inside of it and that means it's got chrome and so it means i can use him sheet with that what makes this especially interesting is that i can run this from a wi-fi hotspot on my phone and so this means i can take this whole setup out on the street for him singing in the winter when it's dark in the evenings this is me just leaving going up to carlton hill ready okay cool [Music] how allows an unlimited number of groups to meet together and these groups should be able to be really large and it does so at zero cost and this zero cost thing is is really important because the last thing you want to be doing is is turning this into a business you know jesus he went into the temple courts and he saw that worship had become an industry there that people were selling animals for sacrifices and they're charging you know more than these things costed and so he got together a whip made of cords and he drove those people out i don't want to be whipped by jesus why do i think this app is interesting well there's many things now that are preventing christians from meeting together to worship indoors firstly there's covert another reason is that we're starting to see council culture in the uk and this is when large christian gatherings are planned there are twitter campaigns against them and we're no longer able to meet to worship in large public venues and in some countries churches are being demolished by authorities now this is something which has happened in the past and it's always brought christians out into the open spaces great thank you turn over the judges for any questions that's fantastic i absolutely love that idea um it's extraordinarily creative um i think i think you make a very compelling case for how it could be used i love the idea that uh ipfs slash the d web can be the uh internet equivalent of jesus cleansing the temple um and i love that you used how great that art um all i watch is that i was around carlton hill that day uh because the i'll i want us to sing that song in public with people um huge plus for creativity great presentation um and obviously i think you have a you have a specific use case in mind but the general idea i think uh can be applied very broadly i'm sure the the rough idea of you know people needing to view the same information for some kind of uh group activity applies to uh to various scenes but certainly uh this is uh probably gonna be more effective than you know the usual um and now we'll all open uh this worship book and turn to this page and you're you're flipping through like this gets this gets the job done very well i think very creative and i like the idea a lot great job thanks i i i thought that was really fun um that was that was super cool uh quick question and maybe i missed this there is there any ethereum in there no no there's not at the moment um the one place that i might put it in is as a um a url shortener um because we're using long uh hash ipfs hashes in lots of places and so actually having a url shortener would be really nice to make it easier to share on twitter and things like that because we're picky piggybacking on current social media which we kind of need for adoption awesome do you have a is there a qr code right behind you where where yeah there's one on the wall um and something i didn't mention in the video is that there's also a social share button so if i click that on here it brings up share on twitter share on whatsapp share on facebook so it's using some kind of nice new web apis there's also a full screen a weight clock api which allows you to keep the screen open which um it's quite nice too very cool awesome a couple more questions how'd you come up with the idea and also how did you do all of that in two weeks because that's a lot of work so there's um a disclaimer at the bottom of the video um so if you click on the video you'll you'll see that um i created the uh the slide viewer the song viewer um about six years ago it's an npm module and so that i just dragged into the project um and uh i just say read the read the disclaimer in the video it tells you exactly what i have done but i have done a lot in this time i've built the entire ui confirming from scratch and you know all this kind of stuff so i was doing with svelt which is a new kind of ui framework great okay if there's no other questions um then we're actually gonna uh take a five-minute break now to make sure the judges have a minute to write down their thoughts about all the teams they've seen so far uh you know use the washroom if they need to um so thank you uh team kim sheet we'll start a five minute break now um and the next team after that is going to be unchained index uh so unchained index uh just uh let me we'll just stay on the line and we'll get back in a few minutes so we'll continue our programming after that thanks foreign okay and we are gonna get started as soon as all the judges are back and the team's ready okay unchained index i believe we've got you on um so please go ahead and start your video um oh sorry ken are you back just want to make sure before we get started yeah sorry i i thought i'd put my camera back on but i'm here no problem okay untrained index uh please take it away we're going to talk about our project the unchained index the web3 native api my name is jay rush i'm going to introduce my partner miriam zandy who's going to walk us through the demo thanks jay prior to the hack we had an application called true blocks that ran as a local first desktop only application we call it a user centric account explorer trubox is 100 private and decentralized but until now it required the end user to run infrastructure locally trueblocks builds an index of appearances of addresses as it scrapes each etherium block periodically it creates a snapshot of the index which we call a chunk and an associated bloom filter which is an order of magnitude smaller than the chunk the api scans the index and informs the ui of which transaction details to give to the user web 2 of course the infrastructure is on a server leaving the user interface in a browser server applications are not private are not censorship resistant they are not decentralized and they are not trustless they are not web 3 native and the problem is this query right here for our submission we wanted to see if ipfs could help trublox retain the benefits of web 3 without requiring the user to run infrastructure we left the api on the user's machine and moved the indexer and scraper to a server we added three new pieces of software a publisher a consumer and a smart contract called the unchained index now each time a chunk was created the publisher pins it to ipfs and then redundantly pins it to pinata with each pinning the publisher updates and manifest a hash of that manifest is then posted to the smart contract the manifest contains ipfs hashes pointing to each pinned item in addition it includes pointers to documents describing the data finally it points to a git commit of the source code that created the index this allows users to verify the data they choose making the data as trustless as possible note that the publishing process does not deliver data directly to users as a result as the number of users increases the cost of publishing does not increase on the user side each time the user queries the consumer reads the manifest from the smart contract and downloads any new bloom filters which are used to determine if any chunks need to be downloaded the process then retrieves those relevant chunks from ipfs a notable aspect of this pattern is that the api and therefore all queries remain entirely on the user's machine making the application completely private another nice piece is that each user carries their own weight small users pin a small amount of the index perhaps as small as tens of megabytes whereas larger and larger users pin a larger portion and popular contracts like uniswap or compound which appear in nearly every block would pin nearly every chunk to us this seems fair so now for a short demo this is the publisher running on the server machine and actively scraping ethereum mainnet building the indexes here in red it says it pinned the index chunk for this ipfs hash and here it pinned the bloom for a different ipfs hash and finally it pins the manifest and post it to the smart contract address i'm going to switch now over to the user side of the equation this is the trueblocks account explorer desktop application in this window we're showing the index chunks of which there are none but there will be soon here we're running the api which the ui connects with i'm going to double click here and the api has gone out and starting to download chunks of the index from ipfs you can see it starts to fill in here now it's starting to download some chunks and as soon as the chunks are downloaded it will hit the ethereum node to bring the transaction records into this application and here's all the transaction records in summary the unchained index is a naturally started incrementally and deterministically produced index of every appearance of every address on the ethereum mainnet it preserves user privacy and is reproducible unlike traditional old-fashioned web 2.0 apis and blockchain explorers let's do this let's build web3 native apis awesome thank you unchained index uh we'll turn it over the judges for any questions sure i'll take a stab um so isn't this first of all isn't this still going to be pretty slow i'm just curious about how what the speed is for doing um any look up against this and then uh at the end you say what is a web 3 you know like uh we need to have web3 native apis can you just expand on that you know maybe just from a high level theoretical theoretical perspective um why why i should care so is it slow and then what do you mean by web3 native api uh sure so um first of all it's surprisingly fast but i think that has to do with pinata being in the mix and we also use the ipfs gateway so there is a sort of web 2 aspect to that part eventually we hope that as the data gets distributed ipfs would be more likely to find a copy of the data that it's looking for on other users machines so we pin those chunks that are downloaded to each individual user on the end user's machine as a default so in the hopes that the index gets broadly distributed making it more and more likely to be found by ipfs just straight up without the aid of pinata because we want to encourage that distribution to happen naturally we want that distribution to happen without the user actually having to do anything um the was that both questions sorry uh yeah and then so uh so web3 native api that would basically it would only really work for append only uh type databases then right because if you updated it then you know obviously everyone's if they're storing it locally they're not going to be you know there's nobody really update that i'm just kind of curious what do you think the use case is for web3 native api so so that's why we we mentioned uh deterministic and incrementally produced because once we create a chunk then that chunk gets a frozen solid ipfs content addressable hash and we can publish that once to a smart contract as part of the manifest and then that piece of information is available to every one of our users applications or anyone's users applications forever so that you never have to republish where this data is so there does have to be someone pinning this data and i think that the community should purposefully decide to be the people that pin this data which includes a large user base and there does have to be kind of a backup organization that pins this data but once it's pinned and once it's available it's available forever in the same location and i hope that people recognize that we can choose as a community to do this on purpose let's say uh what would be like a first use case it's okay if it's contrived i'm just like what type of community do you think maybe this would benefit i think it benefits almost all the communities in web 3 because the way the thing's shaking out now is we're basically building a system that has gatekeepers around it web 2 companies basically who are trying to sell access to this data sell access to not only the node data but also an index to the node data and i think that if these gatekeepers are successful in monetizing the data or the indexes to the data that we're going to end up in exactly the same place we are now with uh gatekeepers and i want to make this data i i always felt that this data had to be completely permissionless or it will get captured eventually and that's not a good outcome quick comment and a quick question i think we're at time um yeah we can do one we can do one question and we'll then we'll move on sounds good uh yeah quick comment i think i like it i in general i like things that are like doing this handshake between um executable code and decentralized storage and i think uh this is nice for that reason uh quick question what what happens in true block completely independent of hack fs i guess um when there's a chain new york yeah we we stay away from the front of the chain by six minutes and then because we're creating these chunks we wait until there's uh 500 000 records in the chunk so that the chunks are about two and a half three hours behind the chain and we think six minutes uh we so there's a distance between the chunk and the front of the chain so we we don't worry we we reconsider the front of the chain until it gets six minutes old and then we stop reconsidering the front of the chain but we don't produce the chunk until about two and a half hours okay so only a reorg that touches over two and a half hours of history would mess things up for you seems kind of unlikely and and the uh the end users that's just the nature of the of a blockchain and there's really kind of nothing you can do about that other than just inform the end users of that yeah that's the unsolvable problem sounds good yes great okay thank you so much unchained index uh we're gonna call up the the next team uh to bring in their presentation uh so that is uh finfile uh i think john we've got you on already so if you wanna share your presentation we can get started uh john we don't hear your video uh if there is supposed to be audio i think maybe you haven't clicked the share audio button oh boy you'll have to stop sharing and then you'll just click the share screen video again and then make sure to click the dialog option in the lower left happens all the time don't worry about it there we go hi everyone welcome to the demo of our hack fs app fin file pinfall is a python flask app that allows the user to input a stock or top 100 crypto symbol and the app will generate a five-year weekly chart and an excel file with the same price data users can send this data to filecoin via textile io's hub where they can retrieve the data using a url generated by the hub let's take a look at the app in action let's generate a chart of amd advanced micro devices the app generates a chart we can now download a png image of the chart as well as an excel file with the same price data let's push these files to filecoin first we need to initiate a hub bucket a demo we do not want to encrypt and it generates three urls we're only concerned about the second one let's push our data to filecoin you want to push these two changes now we want to generate those bucket links we're only concerned about the second link and there we go our two files are now on ipfs under that bucket id let's inspect one of the files awesome our chart of advanced micro devices is now on file coin let's take a look at what we have planned for the future users can use a chart layover if they want to compare two similar assets in the example we have ethereum and bitcoin and the price action over the last five years let's take a look at the flow of files in the fin file ecosystem users can push data via the hub bucket where external contributors can push data to our financial data store users pay filecoin to finfile or external contributors whenever they access their data next steps we need to hook up a grpc client to automate file transfer to filecoin add new apis for more granular data and metrics and consider ways to integrate content from other contributors we need to improve the ui aesthetics and functionality thanks for watching you can find our repo at the link below great thank you finn file i'll turn over the judges for questions i'll jump in with a question um what what benefits does this provide over kind of you know uh some of these services that already out there like like finn vis or whatnot okay um after speaking to an interested party if you think about the traditional finance industry they have things like bloomberg machine wall street journal feed so they can get that and get that information pretty readily now i would focus it on cryptocurrency uh cryptocurrencies and offer uh other types of data that they can't get from some place like coin market cap or crypto compare um such thing as uh hash volume uh how many apps are on the network very specific information that could probably only be found um on our with our app on our on our on our site interesting uh why why store it on a on file corner ipfs um typically from my experience in the uh in the securities industry when you're looking for historical data say data that that that's 10 15 years old very granular data uh for example for technical or fundamental analysis it's very very difficult to find it um and especially if you would consider with with cryptocurrencies there may be some data that that's very specific that could really be not be found anywhere else and it would be nice to create a repository where institutional investors or individual investors who do when their own homework could access our files on filecoin okay thanks that makes sense that's recurring theme is storing data for a very long time that makes sense is there is there any thought about so this is maybe this is a little further out but uh or maybe crazy is there is there a thought behind kind of you know because there's longitudinal data uh and you know with incentive structure you can actually start to think more about the reputation score of technical analysis uh maybe i think a lot of traders uh trade on ta and and so if you have that you can start to build a permutation over kind of how all that uh works over time who's more successful uh kind of all the uh sometimes it just looks like you know randomly drawing lines everywhere uh maybe someone's a better uh line drawer than someone else and then maybe you know having prediction markets over who they think is the most successful at you know doing ta that might be cool is this something you thought about or something you're interested in uh i'm ver i'm very interested in that um i haven't thought about that because uh i completely agree with you um anybody can draw lines and make them look like hey wow that looks cool it makes sense but but all right but but it means absolutely nothing so to have like a reputation score uh and and to score like the data that was filed like all right this one has like a like a 95 95 percentile success rate compared to the other one who does pretty lines and maybe their success rates only 45 percent that would that would really be great because it would differentiate the quality of the data being stored and no no i'm sorry god no no i please please i i was just going to say it would be interesting to have that also thrown on a prediction market because if you stake on someone else's ta and you actually think you know i fully believe that this is going to work you know that unlocks capital for the person and i think at some levels there's some confirmation bias there so if everyone's following the same trends it sort of moves the market a little bit and i don't you know maybe this is going into some weird territory but you know there there might be something there with prediction markets and actually staking with the person whose ta you you agree with and you you know you draw the same lines with and you kind of all pull your money together to you know throw into amd on a i don't know maybe like a two-week call on amd right wow very very interesting hadn't thought about that that just uh that just creates another aspect to it john i had one more for you which is what what things had you thought about that you want to add to this project if you're going to keep working on it okay um the the data feed that i'm using is yahoo finance because it's it's a python flask app and it kind of was a good fit um but i i think that to make the quality of the feed better i was thinking of uh introducing uh chain link oracles and uh and getting feeds from um like a chain different chain like adapters there's some really cool stuff out there and then another one is there uh is there a way for other people upload their own data file so that maybe there's a way to socially share or is it just like you just want to pull it and then i'm curious if there's a social component like i can i can add to the data set or not um well i'm trying to try and make kind of like an ecosystem where people absolutely could absolutely could add to it and then you know ensuring that it that it is truly decentralized cool okay we're out of time uh joe one things maybe think of is you should check out a project called numeroy um i'll send you the link in chat but it kind of has some of the just the ideas you were just chatting about here and you might just find it interesting okay awesome thank you so much yeah thank you so much um so thank you finn file we're gonna call up the next team uh that is a mobility marketplace i believe you're on the call already so please go ahead share your video and we'll get started hello friends this is darkhorse mobility marketplace what's the idea to provide a decentralized platform that connects data creators in this case mobility users with data consumers advertisers why it's about time we compensate people for the data that they create and the ads that they see this project kind of draws inspiration from products such as broad a brave browser it compensates users for seeing ads um and the ocean protocol which is a general purpose data sharing protocol built on ethereum we have three kinds of user groups here micro mobility users which are the data creators and the data consumers which are data scientists and ad agencies and some of the tech we have here is ipfs ethereum fleet hosting storage textile thread db lib p2p metamask infura and ens integration as a micro mobility user i'll be interacting with the react native app the react native app supplies the data that makes the marketplace tick so basically we'll be delivering the data that will be able to purchase by the advertisers and that will be posted to threaddb as well as interacting with the smart contract and opting into the data rewards program in return for receiving ads the user can earn and later withdraw ethereum rewards we'll begin with the react native app where users will push their profile to threaddb as well as initialize themselves for the rewards program each user will have will simulate a trip here and each user will be able to push their own collection for advertisers to purchase the rights to that data and once the trip is completed it will be served in ad through ethereum and ethereum smart contracts and ipfs which will serve them a reward which they can then claim so here an advertiser can view the data sets that are available for purchase so for example here we have a data set from milan and in particular we have one for the year 2020 month is the 7th and then the day is the 10th so here i can view some more metadata on that particular data set and if i'm interested i can purchase it and this will bring up the metamask transaction i'm going to confirm it okay the transaction was successful i'm going to go over here to the data sets view where i can see the actual data sets that i've purchased and then i can actually download them to my machine so another thing an advertiser can do is uh create a campaign and these campaigns are served to users once they finish trips so here i'm going to go through the process of creating one i'm going to give it an organization ufc i'm going to say it's going to be ufc 255 and i have a budget of 0.5 this this just pulls from the active users we have in the rewards program it's read from ethereum and here is just a regular algorithm that we have for detecting how many users we can reach here we're just going to set the asset this is stored in ipfs and the hash is stored in the ethernet smart contract and a little note here is that any unspent ethereum from the from the campaign will be refunded when the campaign expires we will also speed this part of the video up great so the transaction went through and now if i go to campaigns i'll be able to see my active campaign here's the asset here's the details the budget when it expires which is 15 days from now and how many impressions we had on the ad another feature that we have is the feed where we can see all the activity for example we can see all the campaigns that have been added and here is our most recent campaign we can see when a user received rewards and here for example we see the campaign that the user saw in the react native app thank you it was darkhorse great thanks guys i'll turn it over to the judges for any questions not a question but no cool idea overall i think it was well executed uh i'm impressed that you got charles barkley as a user um what are your immediate next steps um where do you want to take this definitely want to introduce some kind of encryption layer so as of now uh we need to have more privacy um data stored on fleeq uh there's no process in the back end to pull the data from thread db and to fleek that would require some kind of go process we didn't get around to that and honestly we just want to improve the algorithm for distributing the rewards as of now it's pretty basic uh but obviously we know the advertising world is anything but basic so yeah that would be some of the next steps yeah do you feel like that's the core of your product like serving the ads uh yeah protecting me and figuring out how you'd be uh rewarding yeah i would say that's the core of what we worked on and then just the concept of how we're gonna serve it um especially based on like in the microability world how they're going to be served it based on where they are and serving relevant ads yeah it's all about compensating users for their data and for their like for you could say for their usage of the product and either whether it be them contributing data to a data set or them uh actually being shown ads um i guess both ways they should be earning money right yep seems reasonable so what are the current i guess um what are the current constraints of the the i guess ad tech world now that stops this from being a i guess a reality sooner well yeah so from from somebody that we spoke to recently apparently there's a huge like bidding system when it comes to advertising and that bidding system happens within nanoseconds most of the time so you know where where is that going to live in this current web 3 architecture for example right is it going to have to be some kind of go process uh does that remove from the trustlessness does that remove from the decentralized nature of the architecture um that's the first one right off the bat how do you think it should be done what's the best way for that to actually happen so right now right now the the bidding process is is really fast and this is your just anecdotal stuff that i've spoken to a couple friends who will work at teams um it's it's pretty insane how crazy that the the tradition like i guess the traditional world is for for advertising is but for i can imagine in like the decentralized world if we can have you know actively bidding on ads for like uh i don't know maybe like a 30 minute lock up um maybe in like crypto voxels right or just giant billboards and stuff uh how do you imagine something like that would work i mean the first thing that comes to mind is maybe there's some kind of some kind of pool where all the ethereum is from all the active budgets and then maybe there's some processes in between that are maybe managed by the advertisers maybe maybe through like a mobile app whatever they can increase their bid and then there's just state channels between the advertiser mobile app and the pool of ads that are going to be shown within the next for example 30 minutes like you said uh but eventually we'll definitely need some kind of layer two technology because you're not gonna do that on on mainnet right is it something you guys are interested in moving forward with definitely yeah so what's next what's like the immediate next step after this hackathon immediate next step is improving the reward algorithm and adding the encryption layer for sure because that has to be that has to be in there very cool awesome guys yeah no i think i i like it i think it's well executed from from start to finish and good scope good job appreciate it follow us thank you thank you awesome thanks guys uh thank you mobility marketplace uh we're gonna call up the next team uh decentralized docker hub uh and there's just two more teams so there's the centralized docker hub and then one more after that uh and that'll be the end of the session so you got the video up already so please take it away hackathon i've worked on decentralized docker using decentralized docker hub can easily push and pull docker images from ipfs and filecoin it is powered by powergate i decided to work on decentralized stockrab because i believe that we need an alternative to centralized docker relying on a single centralized source for docker images is risky because if an image gets deleted then it will affect all your dependencies in the past there have been cases where a particular popular package from npm got deleted and all the packages which relied on it were affected hence i believe the decentralized docker hub is an effective decentralized solution it is also a cost effective solution because it depends on ipfs and filecoin where ipfs is a hot storage layer and filecoin is a permanent cold storage layer now let me quickly show you a demo for decentralized docker the first step is to install docker and docker compose the next step you run power gate lotus and ipfs containers these are the steps i already have it running over here in the next step you create a ffs instance from a new tab and note down the token from the output this is the token i have already exported this as an environment variable so i'll skip this step in next step you determine the docker image that you want to push to ipfs and file kind you can either create a new docker image locally or fetch it from another source in the next step you will clone the decentralized talker app repo and then check out to the decentralized talker rep branch i've already done that in the next step we'll push an image to ipfs and filecoin first let me push the alpine image as you can see that first it got added to the hot storage layer that is ipfs and the next step it skipped filecoin because it is already print in the next step i'll try to push the ipfs image the first step it got added to ipfs and then it skipped coin because it is already print and the next step will pull the docker image from ipfs and filecoin using the cid from the previous step as you can see that it successfully pulled the ipfs image similarly you can also pull the docker images using the ens domains firstly you need to link the ipfs eid to your ens domain and then you can pull it using this command it also supports the test net for the enh domain in this way we pull the alpine image using the ens domain alpine dot docker dot eat and finally you can create a container from the image that was pulled from ipfs and filecoin as you can see that the container got created successfully this concludes the demo for decentralized docker thanks great thank you uh we'll turn over the judges for uh comments feedback and questions good solid idea definitely i think you identified a very specific uh use case and sounds like it would work uh one immediate question how uh like where are these things being shared so say if i if i'm pushing something onto your thing how do i let people know that i've pushed this new image and how do people find out what uh where to get it from maybe uh yeah maybe people can uh create their own uh who's your own page and if you ipfs and they can link it to an ens domain and share that you know that they are hosting these uh docker images and also like say for example uh if i'm hosting certain images then i can link it to a certain uh ens domain and use sub domains for that and i can advertise that thing right but but okay that's that but people uh people essentially already have to know who i am uh yeah okay and uh people can host their own uh pages on ipfs you know which links the ipvs hash over there sure all right i agree it's it's a needed idea i do think there is some work to be done around discovery uh because it's basically gonna be impossible for anyone to find this unless they know to look for you uh but i really i mean i love it because it's super simple uh and it's it's works and it's needed so i think good job on like it's hard to find a use case that makes sense and i think this was a good good find thanks okay there's no other questions or feedback uh we'll move on to the next and final team uh in group one uh thank you to centralized docker hub um thanks so we're gonna call up uh and just get on the call newsbuff so veronica whenever you're ready you can start your presentation and we'll go from there hi all this is my submission for hack affairs 2020 news buff is for citizen reporters who are on the ground when news happens this underserved community and their stories are vital for change and accountability in society in the current mainstream media the 24-hour news cycles ensures any attempt at large-scale change only gets covered when it can be portrayed in a dividing and fear-mongering manner this can easily be seen with the ongoing black lives matter protest in the u.s where the federal government has sent unmarked officers to subdue protesters through violence intimidation and dangerous tactics and despite these newsworthy geneva convention violating tactics mainstream media has failed to inform the public and only covered the protests when there was polarization to be made through violence some major outlets that were meant to be informing the public have become nothing more than clickbait farms sensationalists and polarizing forces through the insatiable force of capitalism other outlets have had to adapt to serve niche communities rather than the general public resulting in them taking sides in a debate they were meant to be agnostic in their coverage of yeah so those outlets have never been lower and yet we don't actually see the video is there supposed to be video along with the audio right now oh yes there definitely is okay sorry okay hold on like the audio desktop right now i kept i kept waiting for the video to start because i thought it was just like an audio intro okay no no hold on let me try reshare uh that's odd there definitely should be uh visuals i think you shared the desktop uh instead of the specific window where the video is playing oh i see it okay can you see now there we go okay i love it can i just press play for where we were uh that's fine if that's okay with you yeah i'm told that okay the need for high quality verified information has never been higher this vacuum created a space for social media to replace traditional news in consumers information diets while convenient for the consumer this has resulted in vast opportunities for misinformation and propaganda to fester social media platforms are designed to be addictive and controversial creating an echo chamber in the size of which we have not faced before the lack of curation and accountability on the side of social media has only made things worse the real effects of which we are only just beginning to feel we need a canonical source of ground level information we need a purpose-built platform that will not incentivize virality over integrity newsbuff will be this platform through the use of decentralized storage we are able to create a censorship-resistant platform newspuff makes use of decentralized storage and hosting as well as ethereum smart contracts to create an immutable storage of information we take advantage of the latest web development frameworks to make a responsive and modern web application this is an overview of how the system is built a non-functional token with 3box is used to identify a user within the system ipfs through textile is used as a storage layer for content and dye is used as the currency for tipping the entire system is composed and viewed in javascript with the use of a typescript helper to integrate with textile the live site is then hosted through fleeq now let's go through what this looks like with a demo you connect your wallet it checks if you have a three box account because i already do it log me in automatically if you hadn't had one it'll make you one here's the home page full of lots of information about us which i recommend reading you can then browse posts you can also tip creators for their hard work you yourself can also create a piece of content when you create content you will have to sign a transaction in order to add this content to your token you can then view the content you created after hack affairs news buff plans on a v2 which will be mobile first and include a dow for community curation of content as well as some machine learning models for initial curation of content ideally we'd also like to raise funding expand the team and launch version 2 in november and for a quick intro hi my name is veronica i've been a full-time blockchain developer for the last two years and as of june this year i left my job to create news buff if you'd like to know more please don't hesitate to reach out or join the telegram group thank you so much for your time and your attention great thank you i will turn over the judges for comments feedback and questions yeah uh great idea i think uh you it was was not difficult for you to convince us of the need of something like uh your product but but you do a good job of it and uh i just want to say shout out for uh you know making this your full-time mission that's very cool um what are your uh i guess uh how do you plan on getting people to actually start because it looks like it's close enough to be used today i'm sure there's improvements you'd like to make uh what are your plans on actually getting people to start posting content and using your service uh great question um so my plan is to kind of uh target people who are in desperate need of this like censorship resistant platform so specifically the protests in the us and in hong kong um but i don't think it's necessarily at a usable state for that group of users just yet um as a lot of you know web3 technologies kind of adds quite a bit of friction so i definitely want to make a second version where i've reduced a lot of those friction points before actually trying to target my end goal for users uh and how do you plan on reducing those friction points so mostly just because obviously this was a hackathon and i i sort of hacked so it was it was it was a lot just to just to do it just with me um so it's mostly just like optimizations kind of bundling transactions that kind of stuff just reducing the the general friction and then like finding a really nice wallet to integrate with that kind of simplifies it on their side in terms of key management and that kind of stuff okay oh yeah i had a couple questions so um i'm trying to figure out like what the what the part about this is um that's like difficult and so like if you want to do news the hard part is actually curation here at least that's one that's one of the hard pieces so um and you mentioned a dao uh it doesn't look like you you have that here um so you know anyone can just take a picture and then post it up on ipfs and then claim that it's news but like the actual having like a some sort of like scroll that i actually know is news um how does uh newsbuff solve that problem by the way great name uh but i'm just curious but i'm just curious how it's going to be news and not just cat pictures um because no for sure so so there's kind of two levels of curation the first one being a machine learning algorithm that's run client side on your content before it actually gets into the system as to say uh and that does like initial curation making sure it's not hate speech making sure it's like relatively you know kind of fits the news bubble of like words and stuff that gets used often and used obviously it's not the be-all and end-all um algorithms tend to you know have blind spots and that kind of stuff um but the second layer is actually like a human curation dao that's financially incentivized um through oh a one minute warning um through through like a paywall dash ads uh kind of model going on yeah so just real quick i've actually seen um a lot of these uh similar projects and um so some of the things that i've seen people use to attack this problem is they'll have like their friends because they'll actually use the social graph that's one way i've seen that happen in order to curate news um yeah so i'm i'm very keen on these types of projects uh but i do think that the hardest part is actually the curation uh and and not the posting so i want i will be following along for news buff and great branding thanks um no 100 uh the the curation of content is actually by far the most difficult part of this um how do you prevent like russia ddosing something and that kind of stuff so sorry i know my time's up sorry okay uh well thank you so much muspuff uh that is the last project we're gonna see today in group one uh thank you to all the teams uh after so many weeks it's really exciting to see what you've all built and we're looking forward to seeing all of the teams as well over the next coming days um so we'll i think uh stop the stream stop the recording there uh and close us off thanks very much all our judges as well uh and obviously we'll be in touch after we call with any details thanks everybody