Data and Ethics
Teams going into the hill in nepal and there are teams supporting them. These teams are finding data about the villages but there are rebls up in the hills that might attack the NGOs.
Responders are at risk.
When you talk about Data and ethics you are talking about risk.
If you are taking data you have to worry about the
- people reporting their data
- reporting sources (eg voting monitors)
Humanitarian community working on coordination using the "3 Ws"
- Who
- What
- Where
NGO are getting attacked if their data falls into the wrong hands.
Question: When you are talking about collecting data are you talking about it being on the machines? In the field.
Katmandoo club collecting information doing mapping.
people in remote locations are sending SMS and data information.
I am here, I have this problem, this is what's around me.
The club is taking this information, categorizing it, and putting it on a map. Once an hour the Nepaliese army shows up and takes a list of needs.
Problem: people around the world are collecting information but some times that information doesn't reach the people running the "last mile".
Paper is often better. Technology can make you a target, it can fail, etc.
How do you trust people? What if technology gets stolen? What if bad actors inject false information?
How are maps prioritized?
there are 2 things going on. People are saying what their needs are and you know the 3 Ws of the relief organizations and coordination.
Hurricane Katrina prompted the development of tools used in Sandy, etc. Lessons learned have lead the US to be pretty good at relief.
Coordinating and standardizing
There can be a pain point between NGOs and Government. NGOs can push in, disrupt food and other areas, then bugger off.
Data is power, they don't want to share that power. It's also about sovereignty.
sometimes you can build protection into the data set. In order to get clean good data set you need to coordinate with many people and the social level filters bad actors.
Anything data you send out one must do risk analysis of data. Will this data get me or someone else hurt? Local knowledge is so very important.
The nature of disaster is that local capacity is overwhelmed. There needs to be systems in place to enable those local systems rather than coming in and taking over.
Scale:
- earth quake happens
- News moves
- NGO: sending crews
- Government: firing up the helecopters
- Mappers: start a map, then coordinate into one map
- Philippians has disaster hashtags already set up.
- Local people are the first to respond
Resilience is so important
Humanitarian Data Exchange (HDX) - CKan instance
search: UN OCHA HDX
Sudden Onset Disasters & Slow burn disaster
HXL data standards schema
Tableau software that takes in spreadsheets to visualize data
What are the biggest bottle necks?
- Rolodex of local connection curated by locals
- Accreditation of local NGOs, networks of trust