Respond, Reuse, Recycle

1
Respond, Reuse, Recycle

• Technology response to Humanitarian Crises
• - learning how to crowdsource efficiently

2
Crisis Commons

• Local volunteering for global crisis management
  and disaster relief

• Global grassroots network
• of technology professionals, domain experts,
  translators and first responders
• collaborating
• to improve technology and practice
• for humanitarian crisis management and disaster
  relief

3
Humanitarian Technology

• Information gathering, coordination and sharing
  to assist in humanitarian crises
• Goals
• Get crisis responders and communities the
  information and help that they need before,
  during and after a crisis
• Outputs
• Maps. Mash-ups. Alerts. Information links
  (including translators and sneakerware).
  Expertise. More people found, fed, sheltered,
  connected, empowered etc.

4
Humanitarian Technology Communities

• Crowd Informers
• CrisisCommons
• Ushahidi / SwiftRiver
• Sahana
• OpenStreetMap
• Louisiana Bucket Brigade
• The Extraordinaries
• CrisisMappers.net

• NGO/Local Coordinators
• UNOCHA - reliefweb
• CDAC
• Diaspora

• Tool Developers
• RHOK
• Aid Information Challenge
• ICT4Peace
• Ushahidi

• OpenStreetMap
• Sahana
• CrisisCommons
• InSTEDD

5
Humanitarian Technology Communities

• Sustain OSM, Sahana, Ushahidi
• Communities built around specific software
• Tool information coordination during crises
• Continuous tool development
• Surge CrisisCommons
• Information support for specific crises, e.g.
  Haiti
• Build new tools for specific crises, e.g. Oil
  Reporter
• CrisisCamps
• Continuous (people, process, tool) development
  for future crises
• Innovate Hackathons
• RHOK competition to create the best crisis
  response software
• AIC creating audit trail for UK/UN/World Bank
  aid funding
• 1 or 2-day events

6
Community Roots

• Barcamp.org
• ad-hoc gathering born from the desire for people
  to share and learn in an open environment.
• intense event with discussions, demos and
  interaction from participants who are the main
  actors of the event
• Hackathon
• collaborative computer programming many people
  come together to hack on what they want to, how
  they want to - with little to no restrictions on
  direction or goal of the programming
• Crowdsourcing and citizen journalism
• Lots of people (communities) helping individually
  with a larger task
• Agile open-source, open-data development
• fast, accessible, efficient community coding

7
How it all started...

• 2004 onwards OpenStreetMap and other tools being
  used in US, UK...
• Late 2004 Sahana developed in Sri Lanka after
  Indian Ocean Tsunami. Then used in Pakistan,
  Philippines, Indonesia...
• 2008 Ushahidi developed in Kenya to map citizen
  journalist reports of violence after Kenyan
  elections. Then used in South Africa, DR Congo,
  Gaza, India, Pakistan
• June 2009 CrisisCommons founded in Washington DC
  after a tweetup by a group of technologists and
  communications professionals who wanted to use
  their skills to help prepare for and react to
  crisis situations both at home and around the
  world
• 2009 CDAC formed after a discussion in a bar...

8
How Haiti Changed Everything

• Late 2009
• First CrisisCamp spawns RHOK and Aid Information
  Challenge
• RHOK0 produces People Finder
• First Aid Information Challenge - overseas aid
  data starts to be available
• UN, CDAC, CrisisCommons etc all plan to develop
  information strategies and crisis response
  communities during 2010
• Jan 2010
• Haiti earthquake. Everyone just does it
• Massive and coordinated crowdsourced response -
  lives saved through tweets, texts and up-to-date
  maps
• Massive not-very-coordinated on-the-ground
  response
• July 2010 - Reflection and consolidation.
  Collecting lessons learnt and working out where
  to go from here.

9
What makes a suitable Crisis?

• Issues
• Too little information Haiti maps
• Too much information Tweak the Tweet
• Infrastructure
• Local infrastructure is overwhelmed People
  Finder
• Some information channels exist SMS, USBs to
  Haiti
• Stages
• Mitigation landslide predictor
• Preparedness OSM worldwide
• Response Ushahidi
• Recovery Haiti Amps Network
• Sustainability CDAC, Karl and Carels project

10
CrisisCommons.org
11
Haiti Earthquake

• Earthquake January 12th 2010
• Response within hours CrisisCamps around the
  world
• OpenStreetMap, Ushahidi, Sahana, CrisisCommons,
  NGOs, Haitian diaspora, Haitians working
  together

12
What does a CrisisCamp do?

• Connects peoples skills time to improve crisis
  information tools and responses
• This supports
• Crisis affected communities
• Organisations in the field (international NGOs,
  local organisations)
• Crisis communities (Ushahidi, OpenStreetMap,
  Sahana etc)
• Organisations in the space (mapping, telecomms
  etc)
• A CrisisCamp links people who want to help with
  places that they can

13
Haiti CrisisCamps
14
CrisisCamp London
15
Handling too little information Maps
16
Handling too little information

• Telecommunications team
• Kept comms going to/from/in Haiti
• Ushahidi
• Connected people in need with emergency
  responders, via the US team
• Biggest moment lots. Guiding responders to
  people who would otherwise have died.
• We Have, We Need
• "Craigslist" of self-identified needs and
  requests by non-profits assisting in Haiti relief
  operations
• Built in days
• Biggest moment getting generator fuel to a
  hospital 20 minutes after they tweeted for help
• Haiti Hospital Capacity Finder
• Listed free beds in field hospitals

17
Handling too much information

• People Finder
• Provided a single place to look whos missing,
  whos looking, how many
• Input from databases, SMS, tweets, info handed to
  NGOs
• Information for Radio Broadcasts
• Searching for and organising news about Haiti
• Tweak the Tweet
• Adding tweet codes to reduce the information
  overload
• Feed into Ushahidi and Sahana

18
Moving from them and us

• Empowering anyone with a phone to report and
  request information
• Haiti project 4636 - SMS to volunteer to Ushahidi
  link
• Breaking the language barrier
• Language projects and Haitian Diaspora
• Connecting translators to locals, coordinators,
  responders
• Reconnecting local information infrastructure
• Information for Radio Broadcast
• Karl and Carels Project
• Connecting low-bandwidth users to global
  information sources
• Low-bandwidth ReliefWeb projects
• Low-bandwidth Ushahidi
• Low-bandwidth CDAC

19
Other Crisis Responses since January

• Chile Earthquake
• CrisisCommons Chile team responded
• CrisisCommons Argentina and Columbia helped
• China Earthquake
• Chinese Diaspora responded with camps and
  translation
• US Oil Spill
• Louisiana Bucket Brigade used Ushahidi instance
• US team developed Oil Reporter app
• Icelandic Ash Cloud
• UK team started news and twitter watches
• Response watches that didnt turn into crise
  responses
• Earthquakes, floods, tsunamis

20
Preparing for Future Crises

• Expert support to crisis coordinators
• CDAC website reviews and prototypes
• UNOCHA Reliefweb reviews and low-bandwidth
  prototypes
• Preparing information for crisis-prone areas
• Populating CrisisWiki with information useful for
  crisis responders
• Populating OpenStreetMap
• Cross-Community development support
• SahanaPy software development
• Ushahidi software development
• London expertise provision User Experience and
  low-bandwidth software

21
CrisisCommons Lessons Learnt

• Have your infrastructure ready. In the beginning
  was organised chaos 30 camps, 8 countries, 5
  languages, 2000 campers, 10000 translators, one
  project list and one country in serious trouble.
• Do what you can with the resources you have.
  Camps picked projects from the list - which
  emptied quickly. Not all of these projects were
  completed. Or started.
• Check if its been done before. PeopleFinder
  replaced 8 main databases. People redid map
  sections because the updated areas werent tagged
  (now fixed).
• Coordinate. Real-time coordination is important
  but neglected. Its difficult across timezones
  and languages we needed a dedicated operations
  centre but didnt know what it was.
• Load up on leaders. Our bottleneck was project
  managers. The virtual camp was difficult to
  maintain without a dedicated leader.
• Timezones confuse almost everyone. A simple
  what time is it in spreadsheet saves a lot of
  pain and missed-by-an-hour meetings.

22
CrisisCommons Lessons Learnt

• Not all projects made it. Common causes were
• No end user buy-in. You can build it, but they
  wont come if you dont involve them. Especially
  true for local communities.
• No team, or no team buy-in. Leadership matters,
  and projects need both people and management.
• Short-term team. Its difficult to sustain
  long-term development. When the adrenalin wears
  off, people drift away.
• Implementation too slow. Be agile build
  something, build it fast, get it used, get
  feedback, repeat.
• Many projects did make it
• Partnership is key sponsors, users, developers,
  communities
• Good business analysis and systems engineering
  are key
• Ownership and leadership are key
• In a crisis, speed matters more than beauty

23
RHOK
24
Using those Lessons Learnt RHOK

• RHOK reused CrisisCommons experience in
  coordinating and running world wide projects,
  camps and experts
• Projects
• Prepared in advance did up-front business
  analysis on real-life NGO and local problems
• Reused CrisisCommons project experts, e.g. UAVs
  and Haiti Amps Network
• Reused Haiti connections to provide subject
  matter experts
• Infrastructure
• Reused CrisisCamp organisers for RHOKs in Sydney,
  Washington etc
• Reused CrisisCommons structure for RHOK wikisite
  built in 1 day
• Ran continuous operations centre
• watching RHOK information feeds
• Linking RHOK0 to RHOK1 teams (PeopleFinder),
  country to country (Turquilt, People Finder,
  wikis) and team to team
• Used a timezone spreadsheet!

25
RHOK to CrisisCamp

• Tools for Haiti
• PeopleFinder tool was built in RHOK0
• Help with aerial imaging problems
• not enough high-res data for OpenStreetMaps
• OilSpill data explosion
• Turquilt project UAV video mosaicing solution
• Help with CrisisCamp Projects
• Nairobi effort and expertise on Haiti Amps
  Network
• CrisisWiki interface improvements and many
  others
• Tool innovation
• Landslide prediction software

26
What still needs to be done?

• Tools
• Big gaps in NGO coordination and situation
  awareness
• Preparation
• Handle UK crisis information vulnerabilities
• OpenStreetMaps for crisis-prone areas
• CrisisWiki entries for everywhere
• Organisation
• How to efficiently build and maintain the tools
  needed in future crises
• Without stifling innovation and the OpenSource
  spirit
• How to keep this local but global

27
What Next for the Communities?

• OpenStreetMap, Ushahidi, Sahana etc
• Well established technology communities
• Already global
• Already building strong links e.g. to
  GrassrootsMapping and UAVs
• GrassrootsMapping, Louisiana Bucket Brigade
• Good models, but no UK equivalent - can we build
  one?
• CrisisCommons CrisisCongress 15th July 2010
• Idea CrisisCommons as surge coordinator
• Idea CrisisCamps to prepare people for local
  crises
• Idea continue VirtualCrisisCamp 3 hour tasks
• Idea RHOK as ideas generator for CrisisCommons
• Choice continue or stop London CrisisCamps

28
Volunteer Skills Needed

• Programming
• Telecommunications
• Mapping
• User Experience
• Communications PR
• Translation

• Local knowledge
• Relief work experience
• IT project management
• Facilitation and admin
• Enthusiasm
• Making tea!

29
How to get involved

• Give some free hours
• http//www.crisiswiki.org/
• http//www.theextraordinaries.org/
• http//wiki.openstreetmap.org/
• http//wiki.crisiscommons.org/
• AIC, RHOK, CrisisCamp events
• Help develop tools
• Ushahidi http//www.ushahidi.com/join
• Sahana http//wiki.sahanafoundation.org/doku.php
• OpenStreetMap http//wiki.openstreetmap.org/wiki/
  Develop
• CrisisCommons http//wiki.crisiscommons.org/
• Others http//wiki.crisiscommons.org/wiki/Other_C
  risis_Relief_Communities
• Get help and tools
• http//cdac-haiti.org/en/ - information response
  community
• http//wiki.crisiscommons.org/wiki/Crisis_response
  _tools

30
The End

• Points to take away
• Its not us and them anymore, its us and us
• You can help - or hinder - from anywhere. Your
  choice
• Getting the right information to the right people
  at the right time saves lives
• Overwhelming people with information doesnt
• Sometimes your tech skills can help people youll
  never meet, immediately and in ways you couldnt
  imagine
• Sometimes it takes longer, but its no less
  valuable
• Thank you for listening
• Any questions?