OK. Thanks. So today I'm going to talk about from geo4some to geo4all. I wasn't always this open source, open data person. And a lot of times people ask me,

how can I get a job like yours? And the answer is, I'm not really sure. I made it myself. I work on the executive director of the Humanitarian Open Street Map team. We use open source and open data to help communities prepare better for disaster, and also to help them respond. So first I'm going to talk just a little bit about my

journey to this open source and open data thing. And it's a little bit of probably preaching to the choir, because a lot of you are already in this world. But there's some, maybe you've come to Phosphogee for the first time to sort of see what this open source geo thing is about and how maybe you can start using those tools.

So on the left, before I became my open source long-haired hippie self, I was a .net programmer. I worked for a company that did a web GIS and sold data subscription services primarily to the US government.

So last weekend I was at the Map Action training and I was talking to some other people. And a lot of us have done this. What bad things have you done with maps? Not just lying with maps. What things were you like, well, I guess maybe I needed a job, I was young, needed money. The other person was talking about mapping endangered species

for development. So how many newts did you kill with your maps? That sort of thing. Mine was mapping for Washington DC lobbyists. I don't know if you've seen the movie Thank You for Smoking. This is a scene from it. So you have someone working for the tobacco company, someone working for alcohol, and someone working for guns.

So a typical Washington DC lobbyist scene. So we used to make maps that essentially would say things like, if you increase the alcohol tax or the tobacco tax over here, all these jobs are going to disappear. So you could take that map to a congressman and say, oh my god, what are you going to do

if you raise the tax on cigarettes or tobacco? And tracking membership of organizations and things like that. But essentially using maps so you can convince politicians to do whatever corporate interests we're interested in. Second one, this is a shot of Hurricane Katrina.

When right after the hurricane had hit New Orleans, I charge $100,000 worth of data to my boss's Amex card. I didn't know that would shut off an Amex card, but apparently it does. And then we sold that to the US government for some ungodly amount of money, I suspect.

I was too low down to know what that amount actually would be. I like to think that today that maybe buying a bunch of hotel data on an Amex card isn't really a profitable business model. So it was about 2009, I started to leave that job.

And I'd also started mapping in OpenStreetMap. And so as I was leaving the job, I talked about how I had been talking about OpenStreetMap for a while. And their attitude was, that'll never work. Even though I think in 2009, it was already working. And I went to work for GYQ where

there's more of a focus on open data and there was some open source projects as well. The first project I worked on was an open source geocoder. The biggest argument at the company I worked for before was, why can't you go find an open source geocoder so we can use it to make money but not contribute back? So it was rewarding to actually be

able to move forward into using software that other people could use. So I'm going to give a little background now switching a little bit more into the humanitarian OpenStreetMap team, which actually started as an idea, which I found out last week in 2005, when Mikel Marin first presented it at a JRC meeting as the idea of, you

could use OpenStreetMap for disaster planning. Hot and I became friends in 2010. That was when I first got it started. I had been mapping an OpenStreetMap for about a year before. And then the earthquake in Haiti happened.

Some of you have probably seen this screenshot before. And OpenStreetMap contributors who want to yell, oh, I've used the wrong attribution, these screenshots are actually before OpenStreetMap changed the license. But so here you see Port-au-Prince. And you see the before and after. In the first week, the OpenStreetMap community did this.

And what happened was the earthquake happened. And people in OpenStreetMap just started mapping, not really discussing it. And just thought, OK, we'll trace some old satellite imagery we have. And after about the first couple hours, people mostly on IRC started talking about it. And more and more people started mapping.

At the end of that month, 600 people had contributed to OpenStreetMap from all over the world to make a detailed map. And this was not the first time OpenStreetMap had been used by official organizations in a response. The first time had been a Map Action map in the Philippines in 2009.

But it was the first time it was really shown that it could be really important to have data quickly in a disaster. And this is one of the shots where, so this is January 22, a week and a half after the earthquake. This is a search and rescue team out of Fairfax County, Virginia in the United States.

You can tell he didn't really think his picture would get used in presentations all over the world, since it's a little blurry. But so this is where we started to see people were actually using the data on the ground. But then you also had more official, larger organizations in the Emergency Operations Center,

like the Pan American Health Organization, the World Bank, Map Action maps. People on the ground were asking for OpenStreetMap, which was really the first time. What's funny, though, is where I come in with this. So this is the DC Roller Girls. I don't know if you're familiar with roller derby.

It's a sport which involves people on roller skates checking each other, similar to hockey. And so in 2010, HOT, the International Organization on Migration, wanted to write HOT a check. HOT had been an idea for five years, but we weren't anyone you could write a check to. But I had been on the board of directors of the DC Roller Girls

when they obtained nonprofit status. So I said, heck, I'm not a lawyer, but I can fill out nonprofit paperwork. And so then we became incorporated, because we had been traveling to Haiti since the earthquake and training people, but usually in partnership with other groups. At one point, Map Action helped us.

OpenGL let some of us subcontract, because we were just sort of a gang of open source and open data enthusiasts that wanted to help. So now I'm mostly gonna talk about what HOT's been doing since then. I like to think of this as, oh, the places your software will go,

because without people developing open source tools, we wouldn't be able to do what we do. We do a lot of workshops, and so we train a lot of software, and we're able, because it's free and open source, to do two things, simply give it to them for free, and two, if there's problems or bugs, fix them quickly or add features.

One thing is, your software doesn't just have to come. We do a lot of, we have a lot of volunteer work. We sometimes have contract positions. We always need technical people to help. I've been talking about the idea

of a hot vacation for a while. So for example, I'm primarily based out of Jakarta, and we've had volunteers come. And Jakarta's not the amazing vacation space you would think it was, but nearby there's beautiful beaches, other things. People travel to Bali on holiday all the time, and it's the sort of thing where you could combine volunteering

with going and seeing a beautiful country, and there's a lot of situations like that as well. So after the initial work in Haiti, we continue to work there today. Our last trip was around May of this year.

This is a team here of both HOT and Community OpenStreetMap Haiti in a project in Sanmak recently this year. In case you're wondering where your software is, it's in those black boxes that say OSM on them. So what we do is we create essentially OpenStreetMap kits.

So you get a computer, GPSs, all the cables to hook it together, a printer and scanner. We're slowly phasing out the scanner because mobile phones can do the same thing now. And you have a mapping kit in a box. And then we put it in a waterproof Pelican case with locks so it can be secure. And find partners that can keep it

and allow people to have access to the technology so that when we go give a workshop, at the end, they're not like, oh, I learned all these cool things, but I don't have a computer. I don't have a GPS. I don't have access to the internet. So to facilitate that.

Here's another typical mapping in Haiti by motorbike, GPS up in Cap-Haitien, same thing. We continue to map in Haiti both for earthquake recovery, but in this case of Cap-Haitien, it's to teach youth how to use technology so maybe they can get a better job someday.

And also to map a commune, a province in the area, so that people have that base map data so you can make planning decisions, economic decisions, see where businesses are, and measure resources. Just some more of our team from down in Cap-Haitien.

So after, it was about a year and a half after we had been working in Haiti, the question came about through AusAid and the World Bank, what could you do if you mapped ahead of time? So it was amazing that the OpenStreetMap community did all of this mapping

after the earthquake in Haiti, but there was no reason that data couldn't have existed before and then been updated. So we began a pilot in Indonesia, and the idea was let's start mapping together ahead of time. One of the groups we worked with was community mapping,

to do community mapping was community strengthening groups in eastern Indonesia. There was another AusAid program there, so AusAid through the Australia Facility for Disaster Risk Reduction was helping out, it was funding us to do a pilot to see about using OpenStreetMap. But there was this other group

that were doing community mapping. The map you see here was made in Karel Draw, so there's nothing geographic about it. And what it's saying is, it's mapping the resources of the community, as well as every single house in that community, essentially who's poor, who's rich, who's middle income,

and how do they relate to those resources? And each village decides what it means to be poor. So you can imagine if the village has a discussion and decide to change slightly what it means to be poor, you have to go in and recolor everything manually in Karel Draw. Think about what you can do with GIS to not do that. And backing up this data,

there are Excel spreadsheets of detailed data, so it is data driven, but the actual map making was not. So we started working with those groups to see if digital tools could make it easier. And one of the things that was funny was we were really sort of worried about printing or how people would use the tools. And a big part of it was sort of printing maps,

taping them together, taking them to a printer where you could get a large format map. But then I started getting pictures like this from community centers. So the local organizations we were working with had just essentially picked up a projector and were going around and showing QGIS and showing the Java OpenStreetMap editor

to the villages they were working with and just having the discussion on the wall like this. And we tried a variety of other things within this first year. We had a university contest where whoever could map the most buildings got a scholarship to Phosphor G. This was in Denver.

And just working with other groups already doing mapping. The reason you don't see anyone who's won a mapping contest this year was in analyzing it. So people mapped a lot for the contest but never mapped anything again. So didn't really work as far as an incentive structure for that. So this year we've been working on a program,

Scenario Development for Contingency Planning. And what that means is you develop a scenario of how a disaster might happen and then you plan for how as a disaster management agency you'll react. How much food you need, how many hygiene kits, where can you evacuate people? And so what we're doing in partnership with AIFDR

is we provide training in those tools so that disaster managers can actually make decisions using software. The main software we use is Innosafe. So it's impact modeling software. And what we do is, so the OpenStreetMap data can serve as sort of the base infrastructure data

and then you can have a scientific model that says, for example, this type of tsunami happened and you can say how many houses will be flooded or in the case of an earthquake, what percentage of houses will be destroyed. So we've been doing workshops in six provinces in Indonesia to teach this.

And as I said, it's OpenStreetMap, Quantum GIS, and Innosafe. We've also worked, experimented a little bit with other tools like using PostGIS for some of the more advanced students. But the main goal is to be able to say, okay, what would happen in a disaster, have a report, says okay, this is how we need to plan,

and then develop a contingency plan. QGIS gets around a lot. I don't know if you recognize the gentleman in the foreground, but that's the president of Indonesia. He's getting a demonstration of how Innosafe and Quantum GIS can help make better decisions

about flooding in Jakarta. So all the places your software goes, it goes in front of the leader of the fourth most populous country in the world. Pretty awesome, huh? So in Indonesia, we had a new experience happen this year. Sorry the picture's so bad.

When you're Instagramming things, heading home from the airport during a flood, doesn't occur to you those pictures might be useful later. So in January this year, Jakarta had one of the worst floods it had had in five years. My team was actually not in the city at the time. We were out giving a workshop in one of those other places. But we came back to it and took a bus home.

It took some of us a long time because there was more than a meter of flooding in a lot of places. What was interesting from an open data perspective, however, was that for the first time, OpenStreetMap was used in showing people

where the flooding had actually happened. So this is a map from the disaster management agency showing what areas are flooded. And this was published on their website. There was also a web map, but one of the biggest problems was people wanted information so badly as soon as they put it up, the server would fall over

because there was that much demand for information. And it's one of the first times they've been able to provide it fairly well. And overall in the program, the thing I'm most proud of this year is we mapped over a million buildings since the beginning. And this is through training over 500 people

about OpenStreetMap. This is our Facebook badge for it. It says, I contributed to mapping a million buildings in OpenStreetMap. We do a lot of social media outreach. So it's really been exciting what has happened in Jakarta. We're also releasing all of the...

The data and the software being open is important, but having good manuals are also extremely important so people can get started. So we've been working on a curriculum with a specialist and it's teaching beginner and intermediate

and trainer level of these tools, trainer level skills for these tools. And it's gonna be submitted this year to the Indonesian government and with a pending approval will actually be an official training program of the government. I think it will be the first time, at least for OpenStreetMap that a national government

will have an official curriculum for this. Wouldn't be complete without pictures of the team. So nine people work for Hot in Indonesia. We share an office with Wikimedia Indonesia. What's core to that is a partnership of people who are committed to open data and open knowledge.

Our office has only been in existence since mid June. And this was our first barbecue a couple of weeks ago of getting everyone together, inviting friends and partners with similar interests to come join us. So we don't just work in Haiti and Indonesia.

We've also been supporting volunteers through the Eurosha project. And what this program does is it supports young European volunteers for six months in countries in Africa to help with information management of local partners. A big part of that ended up being OpenStreetMap but also Sahana was taught,

which is a crisis information management system. And so Hot provided the technical assistance in the field so some of our staff went for two weeks twice, once to set the volunteers up and once to check in and see if they needed technical assistance.

And so this was in the Central African Republic, Kenya, Cameroon, and I'm blanking on one, sorry. And this is actually one of the initial training workshops. There's some people in this room who,

this took place in France who came and helped from this community. So it's great we get to use your software but also people come and help us train it as well and then it goes all over the world. This is a recent project in Dhaka,

which is through the OpenCities project. So the OpenCities project is an initiative through the World Bank to map 100 cities in Asia that are prone to disaster. And so the idea is if you have this detailed information you can better plan.

And it's amazing to me that Hot has moved to so many countries. From the last slide you can tell that I probably should have them written on my hand or something. And so the recent OpenCities project, person that we were involved in was in Bangladesh

and Sri Lanka, once again teaching people how to collect information. And this is working with government and universities and civil society groups. A lot of times, you need a big community to build a map. So you need to do outreach to whoever's interested, whoever can benefit from that data.

We've also been working in Senegal for a while. We had a volunteer community mobilizer take a break from university and spend six months down there teaching. And in that case it's just, who wants to learn about OpenStreetMap? We provide some equipment and we partner with coworking facilities and hackerspaces

and see who we can train. And take very, very detailed information. I'm sure many of you use OpenStreetMap. And where we're standing today is very detailed. But there's also plenty of other places in the world

where it may be the only option for a detailed map. I would think in Nottingham there's probably government, there are government options. But other places, there aren't. So sort of like I had to make my job myself, we make our maps ourselves. Something else that we've been doing

in partnership with the American Red Cross in Ghana is asking volunteers from OpenStreetMap to essentially just digitize every building in an area. And these are volunteers who can help from the comfort of their own home. And we have a piece of software called a tasking manager

where you can check out an area to work on. What was awesome about this is then Robert Banach, who I assume is somewhere around here, went down to Gulu and actually worked with people to add more detailed information that you can't get from a satellite. And so that was used to make a fire risk map.

The specific reason you need a fire risk map in this case is these are the huts that people were digitizing. What was interesting is if you've only looked at satellite imagery of your area and you see, and like me, and you're from Europe or North America,

you may not realize you're looking at someone's home and the potential for it to burn when it's made out of materials like this. Another great friend of HOT and OpenStreetMap is Map Action. So in the Philippines, there was a typhoon in December of last year with heavy flooding.

And what's awesome there is that the OpenStreetMap community has been doing response to flooding and typhoons on their own since before HOT was really more than an idea. And so that base data goes into maps used by response on the ground that Map Action make.

For your explanation, if you don't know what Map Action is you should definitely visit their booth and talk to them. But essentially it's an elite GIS team that travels to disaster sites and they provide GIS support. So a lot of that is making maps. And in this case, in partnership with UN OCHA,

which is the Office for Coordination of Humanitarian Affairs at the United Nations. Sometimes we actually don't do things OpenStreetMap related, despite the name. So Hurricane Sandy hit New York City this year.

Anyway, the thing that we ended up doing is we forked a program called MapMill, which the public laboratory uses for sorting imagery. So they take balloon imagery and then you have to sort through the thousand pictures you took to find one that's good.

What we did instead is the Civil Air Patrol, which is the United States Air Force auxiliary, actually goes and takes pictures of every major disaster in the United States. They have over 500 Cessnas on standby with volunteers. But they used to just go on a DVD to the incident commander and maybe they got used, maybe they didn't. The photos were geo-located and fairly high resolution,

just a regular camera. And myself and Skylar Earl had forked this application at an exercise earlier in the year to see if we can make that information useful. And so what this does is it matches things to the national grid, the pictures of the national grid,

and then we ask people to rate the damage. Is there no damage? Is there some damage? Is there a lot? And then that goes into an average. So people get a grid that says, okay, this area has heavy damage. So it's just an estimate, but then someone can go on the ground and check things out. What was amazing was 6,000 people sorted images.

First time we, sort of the typical, the first, in the initial response, we put a server up, immediately it fell over. Put a much bigger server up, we were okay. But it allowed easier damage assessment on the ground and helping find people, find out where people needed help.

And all this effort that goes into these flying planes and taking images was done, was actually used in a way where people could easily interpret it. Another project we did unrelated to, typically we're in developing countries or the global south as some people call it,

but we've worked in the United States and we also were able to travel with OpenStreetMap Japan to the tsunami affected areas as part of State of the Map, which is in Tokyo last year. Here we are in Kamashi and we were able just to have a mapping party with local groups because complete cities and complete villages are gone.

And as things are built back, you can slowly map the change. Because right now there's entire strip malls that are made of trailers, it's amazing. Because I had never been to a disaster area in a country with resources, essentially. And so HOT, in conjunction with OpenStreetMap Japan,

was able to go just do some mapping, essentially. If any of this inspires you, you can always get involved with HOT, but I also wanted to mention the Digital Humanitarian Network, which is sort of an umbrella group

that brings together these digital volunteer groups like HOT, the Standby Task Force, DataKind, Translators Without Borders, all these different groups doing volunteering and helps coordinate it. And what it does is if you're an official organization and you see all these volunteer groups, who do you talk to? So it gives an actual specific place

that people can go to talk. And I'm a coordinator for it, there's four of us, and we filter that and help figure out what groups can actually help solve whatever the problem may be. So we're not all about training. We do do some other things.

I wanted to mention Learn OSM. So this is the, I think, the first large OpenStreetMap manual available online. It's currently in eight languages. And it's just designed so people can get started. A lot of OpenStreetMap information's in a wiki, and most of us probably have some experience

finding something in a wiki and it's out of date or not difficult to find. And sure, you can go fix that, but if you're just getting started, you just wanna be able to go through the steps. So in partnership with Mapbox, who designed the site for us, we've made this available online.

All the materials are in GitHub, so it's possible to fork it, submit a patch, translate it into another language. What's been sort of amazing about this is when we first released it, it was only in English and Indonesian. And apparently if you release things in English in a language that people perceive that not that many people speak, though they do,

you're like, well, why isn't it in my language? And so people just started translating immediately. So like, well, it's in Indonesian, why isn't it in Portuguese or Japanese or German, et cetera. One of our other tools is, allows people to export data out of OpenStreetMap.

We continue, it's a Ruby on Rails project, and we continue to add items to it. Our newest feature that I'm excited about is it actually allows translation of data. So we talk about translation of software and manuals, but if the data's in English, for example, and it's just difficult to understand. So the OpenStreetMap data is all in English.

That's just how the base tagging system is in English, it's just the rule of sort of the community. So what this lets you do is build a lookup table, so then you could export your data and get it in another language. Because it's easy to enter OpenStreetMap data in your language, essentially, through translation of the tools.

But getting it back out in English, a lot of people have no idea what's going on at that point. This is the tasking manager that I referred to earlier. So it's a relatively simple idea, but for us it's been really important. It allows people to pick a grid to go digitize or work on, and then check it back in.

During the earthquake in Haiti, we told people, just kind of map somewhere. Just find a blank spot. But if you're new, or there's just a lot going on, it can be difficult. So that eliminates the problem. Currently, we're still working in Sudan, where there's heavy flooding.

If you know how to edit an OpenStreetMap, you can always log in. It's tasks.hotosm.org. And take a square and digitize it. I also wanted to give a little mention. We have a technical working group we meet every other Monday on IRC.

Our next meeting is this Monday. So if any of this stuff sounds interesting, and you think you can help from a technical perspective, join in, see what we're working on, and if you have ideas of how we might help, you might help. These are relatively informal, well, IRC meetings.

Finally, I want to give one more shout out to joining the gang, coming on a HOT vacation, really volunteering or working for us. So this is Joseph Reeves, one of HOT's board members and volunteer, came with us to help train people in Indonesia last year, and obviously made some friends.

So if any of this sounds good to you, and you're interested, we'd love to have some help. And I personally think it's a lot of fun. I often get accused of my Facebook page looking like I'm on vacation all the time.

Really, it's like, there's a lot of cool places in the world that need help. So thank you very much. I hope you learned a little bit more about what HOT's about, and the paths from proprietary and closed software to open data

and open software, and not mapping for evil. Thank you. Okay, thank you. We'll switch straight across.