2016 June Meet Up
Ever wanted to know about, or work on, disaster and humanitarian response technology from a people-first perspective? Aspiration's #HumTechFest strives to build capacity for humanitarian aid and disaster response efforts through technology and community. These are participatory and community-driven convenings designed for field practitioners, media makers and storytellers, technology developers, information security practitioners, members of affected populations, researchers, and everyone in between. The agenda is co-developed with participants, facilitators, and partners in the time leading up and during the event.
Please check out our Participant Guidelines before the event.
8 folk gathered on
- Thursday June 16 from 18:00-21:00
- San Francisco Nonprofit Tech Center
- 2973 16th Street Suite 300
- San Francisco, CA 94103
for an informal meetup for digital responders.
We first did a quick overview of what questions and thoughts participants had.
- What tools do you use?
- HOT OSM - geo stuff it misses, takes too long.
- What organizations require humanitarian response?
- How to interface more with local organizations rather than international agencies
- What actions are taken by parties involved in humanitarian response?
- What is humanitarian response?
- Handling variability of RSVPs to the people who show up to help
- How do people communicate cross-sector?
- Bridging formal and informal groups?
- How can I help?
- Post-disaster ad-hoc networking
- Mapswipe microtasking, unskilled labor
- Tech-enabled prosthetic environments
- Who should we be talking to in the Bay Area?
- Ecosystem mapping -- what's going on out there?
- Who do you work with (affected people, partners, etc)?
- What do you do? (Aspiration, humanitarian response)
- Camp Roberts
We then did quick sticker-voting to discover where the most interest was, and gave roughly 20 minutes per discussion topic. Many flowed into each other.
OpenStreetMap is "built by a community of mappers that contribute and maintain data about roads, trails, cafés, railway stations, and much more, all over the world." The logistics for response to crises (both fast and slow) are often based on incomplete maps (out of date or never mapped to begin with). One way digital volunteers can contribute to the response is by identifying villages and other infrastructure on satellite imagery. This is often done at a map-a-thon. When MSF and OSM-er Ivan Gayton went to a map-a-thon, he saw that much of the time was spent on classifying squares (farmland, inhabited, wilderness) rather than identifying infrastructure (houses, roads, crops). This seemed like a waste of time. Alison joined us at the meetup to tell us about a technical intervention for this, called MapSwipe, to do some user testing.
She's looking to find the fastest and easiest way to do the interaction. Also wanted to talk about what project we could foresee using this for, or what sorts of imagery to include (such as from planes or drones).
MapSwipe users swipe one way on their smart phone screen for a map tile which includes infrastructure, and an opposite direction for those which don't include infrastructure. Users are also encouraged to tap in areas which seem to have infrastructure of some kind. The data is taken in aggregate (and cross referenced with other digital volunteers) to prioritize which map tiles more skilled volunteers should focus on sketching on.
Other tidbits from our conversation about purpose and usability:
- Can cache images when you don't have access.
- Portable MapMill.
- Imagery is donated by Bing.
- Can also load your own projects in.
- Reward systems for feedback in cute messages as you go.
- There's a slack that some folk contributing and interested are in! Alison will also send us updates later.
We took some time to review the Digital response ecosystem map and to get input from participants about entities, tools, data, etc. The information has been folded into the overview file.
How can I help?
We had a discussion about how newcomers to the space could engage. As about half of the participants were new to the space, it provided a great opportunity to understand questions (and what was already understood).
- I've worked with a ton of tools but not in this space. I use for my job or open source projects but not for this particular type of thing. But using the tool appropriately is a big question for me. What skills are usually useful or needed?
- Haven't had trouble finding groups.
- Asked to help at a Humanitarian OpenStreetMap Team (HOT) event because I do a lot with OSM. Was considered one of the experts there, but I was hesitant because the landscapes etc are so different from your own. You can feel like you're going to ruin some aid worker by sending them to a mound rather than a hut. Just having very little confidence despite having the background technical knowledge.
- Question about any volunteer effort is "is what I'm doing having any impact at all?"
- Dealing with self doubt
- Even if your project impacted the overall response, how do you know? "327 people liked your road"
- Accountability. What organizations are we working for?
- Rating of stars for nonprofits for donations, what about the nonoffical nonprofits? How do you get a sense for an alignment for what you're trying to do?
- Rate for use of your funds, rate on use of data. There's an EFF project tracking EULAs, could we model off of that?
- Talked about Response Badges, context guides, and accountability
- Have a background reader on infrastructure on Nepal or something, so people getting involved could learn at a deeper level. It can be hard to just click on stuff and have no context. Take the fact books and convert them into responder friendly formats. Lots of shapes for mapping, the forms of buildings and roads that are specific to certain places. Also what look like buildings but aren't. A primer on that could be useful, feel better when tagging things.
- NGOs operating on the ground likely have these fact books for regions. An adaptation for region-specific digital responses could be interesting.
- related project on this wiki: Frontline communities holding responders accountable
- Coming in to help when you don't any context is intimidating.
- A lot of that stuff might exist outside of this context. Design or architecture or urban planning, the beginning of the studio has a design project somewhere where no one has personal context. that's to get people out of their own local. But there's a primer on urban strategy which could be adapted to something like this. But that's well hiding.
It's people, not necessarily institutional support. But the institutions had to make space for it.
Example: Google Crisis Response would try something out and assess whether or not it worked. One thing was taking maps from a bunch of different places and then combining them. National Weather Service overlayed with imagery that only google had, with something crowdsourced like Ushahidi. Large, formal datasets with fast changing datasets in the hopes that would help people synthesize the data to make better needs assessments. Audience for the map was: agencies, less formal organizations and responders. We would build all these maps and pull together information, but then weren't closing the loop (on resolving discrepancies and also letting people know if the data was useful or not). Gov is working behind really strict firewalls so we can't reach out. But we can get ONE site cleared through the firewall, so a coordinating one like Google Maps is way easier.
Gov thinking about how to support groups talking to each other. Markup languages etc that are more interchangeable. HXL is one example.
Google Crisis Map, while a site hosted at google.. the code is 100% open source. You could use it yourself. not just Google APIs, also OSM. Can point at any tile source. The project doesn't contain any data, it just links and displays the data.
Do you get data that smaller groups wouldn't? I don't think we ever used anything restricted. Ask them to make it available publicly.
First time I saw the Sandy Crisis Map was at a rebuild by design. Big government grant which was a design competition for hurricane projects around NYC. A sea wall around Manhattan, a marsh. One linked back to that map.
Side of self care. Especially in crisis you are already altruistic to a point you'll run yourself ragged. You're not around the people who take care with you, tell you to sleep.
Clinical psychology about prosthetic environments. How a long term care giver establishes such a close relationship with that person that they do radical translation and joint embodiment.
Those who are remotely deployed are not taking care with themselves. Concept of an ecosystem where these devices can plug into each other. Catching companies at their first big hurdle helps out.
People who feel compelled to stay online and fix right now. A thing taht is useful about this is the only way you usually have to reach out is chat. Social expectations make it hard to ask. Expect to respond and engage in conversation. Interact with a different method.
Thank goodness for emojis.
Lots of teardown online. Electrical devices, super easy to reverse engineer.
- Alison - Still new to humanitarian crisis response. The whole wormhole. Seems intimidating but pressing on it shows it's super warm.
- Logan - List of open source projects, tasks I can start on. I have tech skills.
- Matt would like to know more about how to get people involved. Let them know they can get involved whether they have technical skills or not. Making it accessible. Don't bog someone down with humdrum work when they can be coding etc. A progression ladder, skill-specific tasking.
- Scotty - appreciative of people coming in. I'm so jaded.
- Nemesis - exciting, excited to know this exists and that you exist
- Nikki - similar sentiments, would do again
- Ping - this was fun, didn't know what to expect, some serendipitous things happened.
- Willow - was taking notes.