 Hello everyone. Welcome back or welcome if you weren't in the morning set. We're having a panel of our speakers from this morning about humanitarian open source software. I'm Kate Chapman. This is Chris Daley. I certainly have a series of questions for us to discuss but first since we skip questions in the morning does anyone have any questions for any of the speakers. Cheers. I keep hearing a lot about different kinds of health tech programs and open source programs for developing or whatever the polite term is these days countries. How do they all interact with each other. Some of them seem to overlap some of them clearly piggyback on top of each other. Are there any formalized systems or formalized ways in which all these kinds of projects work. It depends on the software. One of the things that's happened is there's a lot of hackathons for good days. I'm sure people have probably attended some of those and some of the more obvious problems people keep solving over and over and over again. That's led to some innovation but it also leads to a lot of duplication. There's a lot of software that it's difficult even to choose what to use but with a lot of systems by using standards and simply having open data when available that helps them interact. Sahana for example uses open street map. Ushahidi which is another well-known piece of software has an API and you can download data. It also uses open street map. There's some systems though that just don't work together that well. It sort of depends on the purpose. We also see sometimes you end up with local forks that work very very well in one particular local environment as well so it really depends on the software. I guess it's not really any different than most a lot of open source. People collaborate but then sometimes they fork and go their own way for a reason that maybe makes sense to the developers but not necessarily everyone else. It sometimes leads to surprising improvements but it really just depends on the individual project. Building what Kate was saying that's exactly one of the holes for the map foundation as well. Fortunately though technology is often used by some NGOs or other groups as a wedge to get funding also so they use that as a gateway to go to funders and say well we want to do this based on technology we want to do a small project and that leads to some conflict there with regards to can we bring them onto the platform and our argument is let us build the platform and the technology. You can build the front end so you still have a technology role but it'll freeze up funding and time for you to focus on your core remit which is helping the people in the field. I just emphasise open data sharing data as one of the things what Kate said. I think also with the hackathon thing I think people come along to hackathons and say hey here's a problem that hasn't been solved. I write some code to solve that problem and that code has already been written but it's that that's not the problem it's getting the code out there it's getting user feedback it's looking at a sustainable solution which I mean it's difficult to get that in a weekend hackathon with sort of spikes of surge effort. It's also difficult to get that in a single response. Any other questions? Okay great. Hi my name is Andrew Trigel and I'm in this here listening in this track because I'm interested in open source auto pilots. I'm the lead developer of an auto pilot called IjuPilot and I know it falls in the class of technology that you mentioned in the talks earlier about its complex technology it may not necessarily be field ready it might just get in the way sometimes the technology is too much for some of the operations you are but I'm interested in what would it are their use cases where open source auto pilot technology might be useful in humanitarian cases in the future and what do you want to be done by the open source auto pilot community to make ourselves more useful for humanitarian cases. So the humanitarian open street map team we want we want to use drones and UAVs we've done there's been some work not directly on our projects sort of the side of them using completely proprietary systems which goes against our mission and goals as an organization and honestly so far commercial companies have essentially bundled up an entire system and handed it over and and it's not to say that we're not used to setting up systems we just have never had anyone with a particular desire to make sure it's all bundled together so that we could go use it in a field test or funding so that we could go do that as well. I'm personally very interested I'm on the board of advisors of the humanitarian UAV network primarily advert to advocate for open data more than anything else because hot as a small organization we're not necessarily to fly UAVs in a crisis we might do it in preparedness work but we certainly want that information available. So for us it's maybe having a group or an individual interested in working on those problems and working with us we definitely have people within our volunteer group that are experienced with use of UAVs in development environments and have a lot of experience but they're prior to primarily using proprietary systems right now and so getting beyond that and there's other aspects of our technology that are also proprietary such as GPS that we haven't gotten away from either but just ways that we can approach that and also you know we really want to be able to have good ways that we can leave UAVs with communities so they're able to fix them and depend on them and all those things that everyone else was talking about with technology as well but we certainly are interested in it. I'd say that I've crashed RGPilot before as well I just thought I'd put that out. Literally got the UAV the basic use for among other type of which Kate's mentioned in the event of emergencies or even for preparation or resiliency before an emergency which I think there's a lot of untapped application for drones. Where I'm kind of excited about as a for unmanned cargo drones or for the ability to actually airlift in large quantities of supplies with a safety record that's even now I think probably just slightly better than the legion of drunk Russian pilots that we seem to employ to do a lot of that flying. No, I might have to, that's the, I'd say whether that's going to be up to that same sort of level of He's trying to build a, he wants to send 10,000 drones into Syria from the medication so he's got quite a lot of effort in trying to help him out on the autopilot side to do the appropriate software but he's interested in things like avoiding air defense systems. Yeah right. We think they're from a critical challenge. In some ways I think he's just crazy what he's trying to do. In some ways I admire the effort and things, it's an incredible effort. But what I'm interested in getting feedback is sort of what things, which direction should we push the autopilot development to be more useful in these types of applications in the future. And that's one area that Mark is working on for the dropping medications but there may be other areas we just haven't thought of that we could perhaps develop. One of the areas, I mean which is, I'm not entirely sure for drones but it's one that can also be filled with balloons or kites is, and there's a site overlap with the military but as humanitarian workers continue to operate in insecure and I'd like to say continue to have been for, but it's like area denial is not so much, but to be able to be able to rapidly survey areas and have I guess an understanding or an operating picture of what is going on in a given area can add considerably to the safety and security of NGO staff that are operating in environments like Syria where there's high walllessness and where kidnapping is probably one of the most serious threats that people face there. That having that sort of eye in the sky I think is an area that I think a lot of NGOs would probably be very hesitant to start moving areas so much because of the military implications but I think that considering the growing insecurity that organisations face that we need to consider these types of tools. So it's interesting we're talking about this over lunch and with fear of being a bit of a Luddite I'm sort of a bit like oh no, not drones. And I think I've got a couple of things that I think cautions around and I've had this conversation with others is around drones in this context and I think one of the first ones is is this the right solution or is this just the cool solution and I think it's being really needing to clarify that. I mean I'm sure there are use cases but making sure that this is the appropriate solution not just I mean don't get me wrong drones are freaking awesome but are they the right tool here and it's a very sexy solution but some of the problems are really not that sexy and not that interesting and some of those are more easier to solve when you get more value out of it. The other one is a sort of more of a sort of philosophical thing and I think it really it's important that the drones empower communities and that drones aren't used as a sort of a way to impose humanitarian relief over people. The drones you can just have humanitarian relief without even having to talk to the affected communities. And I think that's a really we should be in the work that we should doing we should be looking at how we empower communities not just fly things over and drop things on and one of the challenges there is in a lot of the affected communities especially in conflicts drones do fly things and drop bombs on them so yeah so drones are scary so and it's you move into that whole crossover between humanitarianism and the military and there's intersex I mean a lot of military's assets are used in humanitarianism there's a crossover there so there's a lot of a lot of cautions I mean but how do you make drones less scary and how do you make sure in doing that you're not making people vulnerable to drones that are scary so just some thoughts. One thing I was going to mention that I think is important with the open source use of drones though is that you can build them with the community in a way that you can't with the most proprietary solutions in a way that if there's ownership of them and they've been built they can be less scary I think. And I think one of the things there is where is that data going from those drones I mean the data if it's owned by the community that they're looking at their community it's not some white person from afar looking at their community which I think is somewhat disempowering. Any other questions? I had a question I guess for Noll or really about Noll's talk. Yeah like obviously you're critical of over land grabs and stuff like that but it still seems to me like you know even going in with the best intentions there's still a possibility that by demarcating people's land like this and bringing in all your technology for measuring it that you can still be imposing a system of private property in areas where it's not necessarily yet dominant. Because obviously there's concerns that particularly with organisations like the World Bank in the wake of disasters reforms can be imposed on people that aren't necessarily wanted. So without wanting to come at it from too critical a perspective I was wondering kind of how you'd fit those concerns into your framework. Now that's a great point and it probably didn't come through strongly enough in the talk. We're not about trying to impose private property models what we want is something that's flexible enough to actually work with the communities to look at how they're handling tenure as well so it's that social tenure relationship not necessarily the traditional formal land tenure system that you'll see in developed economy like in New Zealand or Australia so it's giving the NGOs or community groups or communities themselves the ability to go out and map their own lands could necessarily be just community boundary not individual plots so they can start managing their own resources better as well and it's through that that they can also monitor the use of their own lands also. Just follow on Michael's talk on how the open source model is a bit different from the humanitarian missions where there are a lot of users who are not actually contributing back to the code base say myself I've been in deployment of a software called GeoNetwork where it's a kind of a map repository kind of thing. There are lots of diverse focuses for different groups. Is there any efforts in like how developers actually talk to those users and seeing what they need so like is there any effort in convincing the users to contribute their useful customization back to the code base themselves? Yes but I'll have to dig deeper. So I mean one of the things we've done in Sahana which I think is really good is we have different templates so everything I showed you was different templates and all of those templates live inside the code base which is sometimes challenging working for big organizations which don't understand open source and we're like we want to make everything we do for you open source and sometimes we just do that and we haven't had to big forgiveness at all. I think really the contribution that users make is sort of the requirements we do encourage users to contribute to our Wiki. I think it is a big challenge is to create a community which really motivates and encourages technical people to get engaged and to also and to have those technical discussions that you need to have in open source projects but also enables users to participate and I think one of the challenges we face is I think we're very technically dominated and so we need to create and one of the things I'm hoping to work towards is create forums where users can come and participate and not feel technically intimidated and looking at different modes for that is mailing lists is the right way to do that, our in person meetings conference calls a better approach to that. But I think we've got all of these different templates in our core trunk code and what we're wanting to do this year is to look at what are the similarities between those and talk to different user groups and say you're doing something like this, you're doing something like 90% the same, can we just make these things the same and improve standardisation. Does that answer your question? For the work that I do with geo network we are facing kind of similar thing where there's so much customisation going on for each of the deployment that basically what they do is they just grab our base system and modify it into oblivion and then if we go to the next version they need to do all of them again because it's so much hoping modified. It helps that we have a small team so it's usually the same team that's doing that where our team is committed to the trunk so I mean we still have our base and then what exists in the templates so we're trying to invest as much in the trunk as possible instead of in customisations and it's also it's a funding challenge because you often get funding for deployments but not to maintain the trunk. We're looking for funding at the moment if anyone has any. I was going to mention I'm part of a project called InnoSafe which does impact modelling. Your scientific hazard model like your tsunami or earthquake and you combine that with base infrastructure data such as population information or roads or buildings and it started primarily in Indonesia and Cartosa is an open source software development company that Tim Sutton is involved in and he's been doing as every time he comes to Indonesia he partners with the local Python meetup group. The Python meetup group I would say only about half the people that attend so a lot of people want to learn. It's not a language taught in universities for the most part in Indonesia so he divides them into groups, people who don't have an ID installed, people who know a little bit of Python and people who are ready to work on InnoSafe because usually there's a couple other developers with them as well so the goal is to help develop and foster that software development and he does hire people from the world so it's also potentially a job opportunity and so part of that has been trying to get everyone to contribute to core essentially and part of it also is InnoSafe is very small there's a couple specific funders and the expectation is that you're contributing features to the core software. One of the things we're struggling with is as we grow we still manage to do that as well. If you look at GeoNode which started in a similar way to InnoSafe there's actually quite a few organizations involved now and while there are customizations the organizations seem to take turns funding things so instead of everyone writing a proposal to add a specific feature one group says okay we're going to work on this feature to avoid some duplication. Okay, I'm going to ask one then. So we all talked about, touched a little bit about being involved what's the best way to get involved with your individual projects and for you Chris if people thought the humanitarian road work seemed cool what's the best way to get started with that? I don't know, my personal story is a bit weird. I sometimes tell people I kind of fell into it that I didn't have anything else better to do but that's kind of glossing over. If people are interested in the field of humanitarian logistics it's very very broad field. It does have quite a lot of IT as I said I worked as a GIS operator and web developer before I went to work for Doctors of Outborders as a logistician I might have skipped over that I was in the Army as well for a few years which taught me how to maintain batteries and drive tanks which are not necessarily skills that are in high demand but certainly if anybody is interested in it I strongly encourage people to simply inquire with the agency that they find the work that they do the most attractive. I certainly know that organizations like Doctors of Outborders, Adelaide and other parts of Australasia but if picking up the phone and giving them a call is the way to find out how best to get the process started but certainly volunteering and being more engaged I guess with those organizations beforehand as an asset so best to find out from them what they're looking for at the time. I think Red Cross also do recruitment. Most of the major NGOs do and that's basically people ask me this question all the time and unfortunately my answer tends to be to pick up the phone and ask them what they're looking for. Well as a start up we've got a couple of different streams coming down the pipe I suppose. We're obviously going to be hiring ourselves internally over the coming months but also because we are going to be piggybacking and utilizing what others are doing I'd encourage people to look at the likes of Ushahidi and OpenStreetMap and the communities of developers around that and how they're contributing to the core because they're the sorts of models that we're going to follow as well as some of the platforms that are out there. So there are a lot of open source tools that we'll look to leverage we're not going to try and reinvent the wheel by any stretch and if people come on board with our community and they have those skills then it brings great value to what we want to do and help expedite our journey to get rolling quicker as well. So right now, have you got me? Let's just come and have a chat with me. We've got a lot of information online about how to contribute, how to get involved, which you can look through but because we're here in person I can give you a much more rich and customized answer if any of you are interested. I mean, broadly speaking we do try and be an open source community so there's bug lists that people can get involved in and instructions on how to install our software. We've actually got a position open in Bangkok where anyone wants to head over to Thailand and work on Sahana over there at the moment. So yeah, there's information on our Wiki but I can have a much more detailed conversation with any of you one-on-one if you'd like. Take, did you want to? Sorry, just one comment. One thing is that I've just talked to Kate before and I found that because I actually lecture in GIS here in the University of Auckland and over time I saw a lot of students who are interested in humanitarian aspects of GIS but they've never been able to find a way to get into it from Auckland and they don't know who to talk to and when they just send an email to say a general email address to say the Red Cross they would say like we will come back to you once we know some information and they never come back. So yeah, I would say that it would be good if you guys could actually go to say universities or lectures for GIS or geography or something that's related and talk to those students and that way you would be able to get a lot of people who are interested. One thing, one challenge that at least I face in the Sahana community is there's lots of interest but what we really want is commitment and we do spend a lot of time talking to people who are interested and introducing the project and I think that's an important thing that we need to do is what I really look for is someone who says found Sahana, struggled through the documentation actually it's not that hard, we've done a lot of work to improve it and someone who's really invested a lot I mean my own path into humanitarian work after the tsunami I tried emailing no response there. I don't know if I necessarily encouraged that especially in the changing face of humanitarian action and it's a lot of the major disasters, major security hazards as well but I think just showing that commitment, putting that personal commitment and makes you more attractive to those organisations and also picking up the phone as Chris said is more effective in emailing sometimes I did want to add a little addendum to that an interesting point which is that organisations the big organisations can be very vertical in the types of profiles of people that they're looking for and very field focused in the sense of either logistician, nurse nutritional advisor, they're very clear in their very vertical roles and some of them are what the organisation was founded on doing some of them are I guess to say horizontal so out of the box sort of things like GIS although it's not new technology the application of it in humanitarian circles is still very new and there's very very few or very small number of organisations that actually have their own GIS capacity in house and it doesn't surprise me that if not just GIS but a whole range of other more horizontal fields you can email the info at address on the website or even talk to somebody necessarily in the call centre you're going to get a check and somebody will call you back and that might not happen but I would highly encourage, like I found it quite interesting to come more into contact with the GIS resources at Doctors Without Borders an organisation of which I've worked for, I'm a member of, I've been a board member of through mapping with hot then I did through my own organisation, it seemed to be far more broadly publicised outside of the organisation than inside and I would imagine that this is very similar with other tech projects that if you're wanting to get involved in humanitarian tech then maybe the sideways approach of volunteering and being committed to a tech project that's in collaboration with the big organisations would be a far more direct way of getting in touch with the people in those organisations that are trying to marshal the resources, the tech resources rather than just the info or phone number or email address. Twer would probably be able to engage with those people as well. And one thing I was going to say I think also with the software projects, sometimes it's easier to get in so Hot for example, we're a small organisation with 15 staff and I had an intern lived with me for six months in Jakarta and it was just because she seemed really committed and really interested and I had an extra bedroom and a larger organisation isn't going to be able to do something that casual probably. But the way to get involved with us is first by mapping we get a lot of emails of people who want to go off and do field work somewhere and we have more people than we have projects at the moment or that sort of thing. But we pull from our volunteers when we make those decisions as well. So that's a great way for people to get involved. And as I mentioned to you, within OpenStreetMap there's an initiative called Teach OSM which is aimed to get universities better involved and Hot is involved with that but it's also the broader OpenStreetMap community. Sahana and OpenStreetMap also participate in the Google Summer of Code so it's a good way for students to get involved with paid internships. Hot has two interns with the GNOME outreach program as well at the moment and I'm hoping we'll do that again as well. Because those are I think it's also important, we talk a lot about volunteers but it's also important that people need to eat and they need to pay their bills and all of those sorts of things and so having both types of opportunities for people as well. Does the panel have any questions for each other or the audience? What would you... I can ask another question. What do you think your the challenge that first comes up when someone says what are your challenges within your organization or what you're trying to your mission and what you're trying to accomplish? Well we have obvious challenges as a start up and we have many of them but in terms of our sectoral challenges I guess we're working in a sphere where there is long held traditions within government that we're trying to break down as well so we talked about ineffective land administration systems that exist right now or ones that don't exist when we try and get into governments there's a lot of rent seeking that goes on as I mentioned with the corruption angle but there's a lot of protection or self-interest that goes on about protecting their own jobs. One of the places I worked on mid-2000s was Zambia and I went in there with the good intentions of helping streamline transaction processes within the ministry of lands and putting a new land information system in. Well I found out that the land information management staff who were running an open source system and they were espousing the virtues of that but at the same time they were identifying the vacant land parcels around the city outskirts and either selling those off or putting their own families names down. A couple of those got caught out and sent away but you can't underestimate the people challenges that you're going to face. As we've heard tech is just a piece of it. If it's 10% or if it's 20% a lot more comes down to what is happening around in that sector and it comes back also to how do you get into the sector. Yes, focusing on the technical aspects of what we do individually as organizations or collectively in terms of developing solutions but thinking beyond that. From my point of view what are some of the greater challenges of crowdsourcing data and putting it up on the web. Not from a technical standpoint but what are the security aspects, the transparency aspects, how do I mitigate or deal with unintended circumstances. So those are certainly some of the big questions that we're facing. It seems that getting the finance part was easier than what those are actually going to be and I think building that community around what we're doing as a platform and having people understand who are not necessarily in that sector. I think the diagram that showed the split of contributors and users is going to be quite important to us because that's really what's going to affect us. We'll have a lot of people that I think will contribute to the code and the platform but they won't be the end users. So how do we bridge that gap and build capacity within the local users as well so we get more crossover. I would say that's one of our big challenges is diverse groups of stakeholders. You've got your contributors your documentation contributors, your design contributors how do you get all of them talking together and then you've got very different users who I don't think I mentioned in my talk, when I first started working for an NGO in Indonesia there were huge cultural differences and it wasn't between me as a Westerner and the local Indonesians, it was more between me coming from a very sort of technology focus to working for an NGO and I sort of say it's coming from a world of black and white to a world of shades of grey. It's different mentalities and things work in different ways so it's bridging those different stakeholder groups. I mean it's looking at how do you engage them all in a way and empower them all in a way so you don't have sort of your developers who feel that they're just being sort of told what to do by sort of expert users who don't really understand so it's how to bring those all together. The other thing we face at the moment I'm sort of as I said I'm relatively new in the role. Sahana's been around for a long time but it's sort of as a foundation the foundation hasn't really been very active in the past sort of five years and so my challenge is really to invigorate the foundation and look for funding to help support that so that's one of the challenges I face. I mean I think just from the broader adoption of or integration of tech into humanitarian I think that the single biggest thing I've seen at that end when addressing the much greatly outside circle of the end user as opposed to the contributor is language both concrete language that people speak and also the meta language of computer interfaces. A great disparity in the people that we employ in people who either speak both languages come to have a much clearer advantage over other people in their communities in terms of their access to that employment and how we bridge that gaps first there's that computer interfaces are becoming better touchscreen technology is becoming more pervasive so people are becoming much more accustomed to it but I think there's a bigger challenge of that as much as English may be the global language not everybody speaks at a function at a level that's functional. So I think that that's a challenge for all developers to consider in both the user experience and also how they make it internationalized and that's a word. I know for us one of the big problems is long-term sustainability of the types of projects we work on. A lot of times people want to fund us to map the city as quickly as possible but then of course the data gets out of date. You can't build a community of any sort in two or three months. Our most successful project to date has been in Indonesia and I lived there for three years. I just moved away six months ago so we've been there for almost four years now and we have a local staff in a local office versus some of the other places we've worked just for a couple weeks. It's really a community. One or two people might be really interested but they don't really have enough support so when we're doing paid programs and paid work it's thinking about how it accomplishes our broader goals rather than a very short-term exercise. Any other questions from the group? Sorry if this has been asked before just came in. In terms of hardship getting people to create a community or something like that, have you considered tapping into an existing community to do the maps, say for example maybe the local scout, so girl guides as part of their, because they do some sort of mapping, tracking thing that they might actually use that as a learning tool or something. We do. I find to find that local community and who that community is can vary from country to country. We've worked with scouts before and one of the key things for working with them was simply having a couple scout leaders that were really interested. Sometimes we work with civil society groups as well and one of the main things that we've discovered works is finding groups that need maps but maybe are doing them in really inefficient ways and then saying hey we'll help you do this mapping better. They were going to make maps anyway so we showed them a way that might be more effective but it does seem to really, it just really varies country to country. For example, in Indonesia we share an office with Wikimedia Indonesia and we have a very close partnership. Our Indonesian staff actually technically work for Wikimedia Indonesia but then in the Philippines there's actually a much stronger OpenStreetMap community than Wikimedia community so we work closely with the OpenStreetMap community and it really varies country to country. We definitely look for those communities but sometimes it takes some time to find them and figure out who they are. Thanks. If you could go back, I don't know, 10, 20, 30 years, what are the big, I don't know, can you name like a single design decision you would have changed about some of the systems you're encountering? This isn't really a design decision but we've talked a lot about naming for example so the Humanitarian OpenStreetMap team. Humanitarian response is very different than international development and so in the first couple years we essentially wanted to do disaster preparedness which is international development but convincing people that we did international development with the word humanitarian in our name was difficult. So I don't know what I would call this instead, it's too late now but thinking about those sorts of things as far as decisions it's difficult with what we do because we are on top of OpenStreetMap and OpenStreetMap made some very deliberate non-traditional mapping decisions 10 years ago which I think actually hold true as being good decisions but they make it difficult to hook into traditional mapping. For example the OpenStreetMap tagging system uses key value pairs instead of a traditional tabular database structure which does not fit well into traditional systems. So what we've ended up doing is building tools around them but in building those tools around them I guess the design decision was we should have thought a lot more about the interface and who we worked with to build those interfaces because we have a couple where they do amazing things but no one can figure out how to use them so that human computer interaction I don't think we've given that enough thought all the time. I would say from limited software tools that I've used over the last 15 years I think it's probably a common thing that developers would know but there's a tendency to design for the systems that people will have next year or the year after next rather than the one that they've been working with for the last three years and why is it taking half an hour for this access database to just show one single drop down menu. So I think that's probably something that everybody who's a developer is familiar with but I can't emphasise it enough in developing for humanitarian tech. It's got to work on the systems from three or four years ago not for the ones that are going to be released next month. And I don't mean to offend any developers in the room but I guess the biggest design decision that we would have changed from the previous company is that we had developers designing an interface. I think that holds true these days with the focus on UXUI and getting that upfront, that experience. Developers have a very different mindset and it's a completely different language to the UXUI working with the end users and we didn't pay enough attention to that and I guess we would have been much more modular in the design that we deployed as well rather than hard coding everything into a large enterprise system which made customisation for that much more difficult. Of course we could charge more for it but it gave us a hell of a lot of time trying to work with distributed teams on development as well. I would say build less design more. I think we're often too eager to code a solution without really knowing what the problem we're trying to solve is and to really confirm that that's a problem. To investigate the problem in all angles, figure out who the stakeholders are, who's going to use the solution. One other thing I would say is maybe I think we maybe have over structured our data model a bit. I think we could have potentially used a simpler data model which gave more flexibility to the users in terms of what they put in different tables and instead of trying to guess what the data they're going to be managing is make up a bunch of field names and then find out that actually when they start using it half that is blank and they're using it for something else. More unstructured data and then responding to what the users use it for. I worked on a, reminding me of a design decision I made once that was bad was attempting to replace Excel unsuccessfully. I did. Part of that is just being careful in figuring out how you're going to replace things. For example, we're working on a flood reporting application that hopefully will be used by the Jakarta government but I have no expectation for everyone to be using that for reporting this year. We're hoping to replace some factors in some SMS and some phone calls this year. Just some. It'll take years for adoption but getting anyone to use it is the first goal. So that's actually it for us on time. If you have any other questions feel free to approach all of us. I know that myself Noel and Michael will be around this week. Chris is just attending today but he'll also be at the Penguin dinner if you have any questions. Thank you for attending the first open source humanitarian mini conf at LCA.