 Hello and welcome Paul who is going to be speaking about the open food network. Okay. Hello Hello, everyone That's me. I'm pao pwedef. I work at co-op depths cooperative in Barcelona and spend my days working open food network and That's my Twitter account. So let's start with some vision So the goal that we have at open food network is that we want to help improve food systems. So currently big supermarket chains control the food market throughout the world Which has back implications for social and environment and also doesn't take into account the participants in the market Which are the food producers and consumers So our idea is that we want to help those small-scale food enterprises and distribute through short food supply chains and we understand that's a way to Bring resilience to systems and make it more competitive So what we found out is that those small-scale food enterprises they Share 90% of their problems. So all around the world people have the same problems trying to distribute those products to the local communities so we ended up with it's an international network of Local projects that try to help those small food-scale enterprises It started in 2012 in Australia and then suddenly started growing in France UK Norway Spain now in the States in Canada and many other countries and the last one Belgium so The idea is that if we all share these problems we could share the software and Then if we put all the money together we can have bigger impact because Those small local communities they cannot afford developing a software for all those problems and we can see the software and then knowledge behind that software as a common good that we need to protect and There's obviously politics behind it, and I'm gonna go just very briefly on it But we can feel at the global level we can see the software and that knowledge like a common good that we need to preserve but also those local communities like Every country that they showed They have the local community. They're people that want to help on that end And They have the local projects as a common as well, which means that they will protect that common and normally they organize themselves as Cooperatives and also at the lowest level these local communities are made of what we call food hubs which is those small enterprises that distribute food throughout those short food supply chains and Obviously there is connections between all of them and the needs of the food hubs Come up to the global level and the international network takes into account those needs and tries to solve them with the software so a little bit how we work in this project so the key point here is that we have a single money pot Which means that all those local projects Spain, Canada, USA, Australia they get funds from Public institutions sometimes from other ways crowdfunding campaigns and we put the money in a single pot and then from that pot we finance and we fund the development team and We try to reach that goal together So the global team it's more or less these people it's people from all those local projects and Yeah, we gather every year and we try to visit one of those local projects That was last summer in Barcelona But a year before we met in Australia and that's where we discuss what are the next priorities what we feel the project is going to be and How it needs to evolve taking into account those needs of the food hubs Then what we have is what we call the voting process so we have a discourse and there we share Coming from the gatherings We share what we think we need to implement next or maybe discuss or work on and then People on the global team vote at their votes and then at the end of this voting process we decide what's gonna be next Then we have what we call the pipeline So basically it is run by the core team and we put work to do on one end and then we have releases on the other one and What we organize that with four backlogs using Zenhub, which is an extension of github and Then we pick we order our priorities in each of them and then we pick the top ones and They fit the death ready column where the core team is going to pick them up and other contributors and Implement them then we have what we call the code review and merge process which is run by few people from the core team and Basically we have to step approval promises So you need the approval of your food request by two people from the core team and then a tester is gonna pick it up give her approval and then The tester moves the issue to the ready-to-go and then one of those core team members merges the food request And as you can see I think we are a very welcoming community and we really encourage people joining the project and Then that feeds this sort of the weekly releases. We don't have a set a schedule for them But that's mostly when we do it and we put a lot of effort on it Because we think that's the way to breach the gap between the local communities and the food hubs, which are not technical people We explain them what we built what we improved what we fixed and we follow the change look Categories and obviously we name our releases after food So let's get into detail about tech So basically what we build is a software, which is a distributed marketplace for those small-scale food enterprises It looks like this on the front side So these the experience for one of those eaters from a food hub anywhere in the world and it's a basic e-commerce experience But it has a back office built with free But what we added in top of a spree is the layer of enterprises, which means that you can have multiple shops in an instance and This is how it looks so we have free commerce what we call the network on top Which allows to connect those food hubs between each other and then we have this custom shop from that I showed Which is not a sprees code So then the stock is playing all Ruby on rails with Postgres and AngularJS for this custom shop form So nothing really fancy, but it works So one thing that we need to take into account is that all of an instances what we call all of an instances It's one of those local projects. They have an isolated server So there is this database per country region and there is no connection between them So each of those instances is offered a sass By those local projects, which in turn are affiliates of the open food network And the architecture is very simple the infrastructure Very very simple single machine VPS with a unicorn and possibly sequel Because the scale that we have and since we split that into countries. It's very low traffic We can survive with that so far And then we organize that with unscable scripts We have provisioning and deployment there and we also share that repo So it's the code, but also the configuration management That makes it really scalable and easy to share between countries But then obviously there are many problems into this the first one is dependency management That's something that because of the history of the project. It wasn't really taken into account the people that started they were rather junior people back in 2012 and It wasn't something that was taken into account and what we face is that our spree version is from 2012 And that's a huge problem because obviously rails version depends on it Ruby version depends on it And that's why we have rails 3 2 Ruby 2 1 which are both abandoned and It's also a problem with angular because now angola is running on 7 and we are angular 1 5 so we have a huge problem there and We've been working more than a year on the spree upgrade to migrate to the next version the major one But it's a huge challenge to be up to date Another problem is that Many instances can find tech people. That's particularly important for Canada and the people leading the project there They weren't technical people. So it's hard for them to reach to developers and The way to solve this we think it's also Communalizing develop so we started sharing the code and then we share the answerable scripts We need to share as well the way we organize the DevOps because these people we cannot pretend them to find tech people and That's particularly interesting for Belgium so they started four months ago and there is no tech people because we also share the duty of Deploying and provisioning their servers and that's lowering a lot the barrier of entry for new instances obviously we would like to have faster delivery and It takes time to test releases and we want to automate that as much as possible and then the biggest one which is what we call the single instance and Now the project is getting traction and more countries are joining But again as happened with Belgium We don't want people to invest money on infrastructure or technical stuff That they can't we want them to invest the money on the project itself and then we will organize together globally and we'll share the duties and Make it easier for new countries to join and also share the expenses of running an instance That's what's gonna happen now with Portugal because they are joining the Catalan and Spanish instance and We'll see how it goes. It's gonna be a huge challenge, but a huge game-changer and So we need more hands as you can see and I really encourage you to join the network and Please go visit the ripple and help us out. Thank you very much A little bit of time for questions. So if somebody has questions Yeah, we are somehow and that's because the way it started in Australia It was a solution for them and then UK came and they wanted the same but they were The teams were isolated themselves. They had technical people So we are trying to overcome that and that's we are looking forward with this with this join effort with Portugal because they might lead to a European instance and They might help and resolve this centralized Everyone can join the team, you know what I mean the project We had discussions couple years ago about blockchain and all that stuff and we think that we're struggling already with this Very simple stack. So we want to keep it simple manageable and easy for people to join and Then when the project grows, maybe we will think of these technologies. Yeah Any more questions? So that's a problem that we want to solve and it's going to change a lot the project The thing is that we first want to have data On our databases and then together we'll decide analyzing the data What can we do but we are already discussing with people actually here in Belgium? That investigate more these logistics part, but that's a key point for all those small food food haps. Yeah, definitely