Sahana in Haiti: More rigour, less marketing needed

The Sahana Free and Open Source Disaster Management System in Haiti by Chamindra de Silva and Mark Prustalis appears in ICTD Case Study 2: ICT for Disaster Risk Reduction published UN-APCICT/ESCAP. Though more than half of the essay is a generic description of Sahana, Section 6 onwards deals with the deployment of Sahana in Haiti earlier this year and is worth reading.

One of the most enduring memories I have of the virtual relief efforts in the two weeks after the earthquake in Haiti was reading emails on various groups and websites by Mark Prutsalis asking, nay begging at times, for vital information on hospital locations to be made public. Mark’s crowd-sourcing geo-location of this vital infrastructure is in my mind one of the best examples of how a global community can be galvanised to help an urgent humanitarian need. Precisely because of this, I wish the essay dealt more with lessons learnt and identified in Haiti and less with marketing Sahana as a platform. For example, Mark’s message on 23rd January 2010 on this Yahoo! group dealing with the exercise of geo-locating hospitals is well worth reading in full. As Mark notes,

In the past 24 hours, my call for volunteers for this effort was answered in an overwhelming fashion. We started with 100 names of hospitals in Haiti that we knew existed, but did not have coordinates for – latitude and longitude – such that we could plot them on maps. For some, we had street addresses; others, maybe only the municipality in which it was located. With little instruction other than to think creatively, we have now completed this task. At this hour, 3 remain… and I’m confident that someone will be able to track those down as well.

Though the essay deals with Sahana’s success in mainstreaming the Emergency Data Exchange Language – Hospital Availability Exchange (EDXL- HAVE) standard to meet the type of medical reporting that was necessary in Haiti (the operational status of a hospital or health facility, its bed availability and resource inventory etc), it does not go into details on how this information was found, cleaned and geo-referenced, which as Mark himself points out in January, was a significant global volunteer effort. In addition to Sahana’s own platform and a number of other web and mobile based platforms, I posted the resulting data to the ICT4Peace Foundation wiki on Haiti. This was genuinely useful information, and a cogent example of going outside the UN to crowd-source actionable information.

However, Sahana’s commendable efforts notwithstanding, it took just shy of two weeks to get this information online. This is unacceptable. Access to data repositories and data redundancy were significant challenges on the ground, but there was sadly a sense that even vital data was available – especially within the UN and Minustah – it was not always easily or immediately shared.

In a similar vein, Erik Hersman from Ushahidi on 16th January said the lack of a database of organizations on the ground was a huge impediment to relief efforts, calling for Minustah to release this information. Nicolas Chavent from the Humanitarian OpenStreetMap Team on the CrisisMappers Google Group on 22nd January requested a comprehensive Minustah data model and data dictionaries to aid mapping efforts. The ICRC simply did not play nice with any other relief agency within the UN and especially with those outside. Tim Schwartz noted on 21st January, also on the CrisisMappers Google Group, the degree of manual labour involved in getting information out of the ICRC missing persons database and into the PFIF format used by Google and others. He averred,

I implore any of you that have any connection at the Red Cross to try and try again to get us in contact with them. It is disappointing that we have the two largest systems out there not able to talk to one another, but that is just what the ICRC’s site is: lacking communication in both the programming sense and the human relations sense.

Emphasis mine. Erik Hersman had echoed this frustration with ICRC, and also pointed to CNN’s obduracy in this regard a couple of days earlier on Ushahidi’s Haiti Situation Room.

These and a number of other processual and technical problems are flagged in several documents that interrogate the use of ICTs in Haiti. Haiti and Beyond: Getting it Right in Crisis Information Management that I co-authored for the ICT4Peace Foundation was one of the first to eschew the hype over virtual relief efforts and flag serious, enduring concerns over the quality, sustainability, effectiveness and efficient of ICTs. Haiti earthquake: Breaking New Ground in the Information Landscape by the Humanitarian Information Unit (HIU) of the US State Department is another excellent critical look at the relief efforts using ICTs, and what more needs to be done.

There is immense potential in the emergent coalition of Sahana, InSTEDD and Ushahidi outside of the traditional UN Cluster approach and platforms like UN OCHA’s OneResponse. Every single one of these actors contributed significantly to the underlying technical architecture that allowed standards based information generation, exchange and archival in Haiti. InSTEDD for example has an excellent write up on this here, and Ushahidi’s writing on this score are innumerable, and just a Google search away. With due respect to their work however, as Chamindra’s and Mark’s essay in this compelling tome demonstrates, the marketing of a single platform – necessary perhaps for fund / profile raising – in contradistinction to others negates and risks undermining the value of the collective, which is greater than its constituent members. It also risks glossing over vital and enduring concerns about the use of ICTs in relief work related to, inter alia, challenges faced actual use cases, the hugely instructive nature of project failures that are often hidden or cast away, lessons identified and learnt, issues of local ownership, stakeholder interaction, language, accessibility, gendered concerns, community participation and sustainability.

For example, though the essay goes into some detail about the potential of Sahana’s Shelter Registry (SR) module, it also notes that it is not used at all in Haiti. The authors don’t ascertain why it is not used for what is clearly a vital need on the ground in Haiti with tens of thousands of IDPs, and what, if any alternative ICT platforms are used for this purpose. Is non-use to be interpreted as a rejection of the SR’s functionality, and by extension, Sahana’s usefulness in long-term recovery coordination and collaboration efforts? Is it that actors on the ground don’t know about its potential capabilities? Are there other, better more effective systems out there that Sahana can incorporate features from? These are questions not asked, but should have been.

Sahana’s work is a source of pride as a Sri Lankan, and their technical innovations are no less significant than say those by Ushahidi. Sahana has also matured as a platform, from what it was post-Nargis in 2008 to its implementation and work in Haiti this year. And yet, devoting more than half an essay in a vital publication to the mere marketing of the platform, sans any anchor to real world use cases, suggests Sahana – even after 5 years of existence – is, needlessly may I add, insecure over how it is perceived and used.

I really hope they follow up with a more rigorous essay expanding on Sections 7 and 8 in particular.

InSTEDD’s response in Haiti

I may have been one of the first to blog about InSTEDD when it launched in early 2008. While other crisis information management providers in the aftermath of Haiti’s devastating earthquake in January this year widely promoted their work and efforts on the web, InSTEDD’s work, though captured in the Haiti earthquake wiki I created and curated for the ICT4Peace Foundation, remained comparatively obscured.

I’m glad that Eric Rasmussen, CEO of InSTEDD, wrote in to the Foundation with a comprehensive account of InSTEDD’s technology deployment in Haiti. As he notes in response to the ICT4Peace Foundation’s two-pager Haiti and beyond: Getting it right in Crisis Information Management,

InSTEDD was one of the first teams on the ground after the Haiti earthquake. Nico di Tada and I arrived that first Friday, the 15th, and established a foundation for information flow that has since received more than 90,000 SMS text messages from the Haitian population, and broadcast more than 600,000 SMS messages to our 23,000 (free) SMS subscribers in Port au Prince. We developed the backbone on which Ushahidi, Crowdflower, Open Street Maps, Samasource, the Haitian Red Cross and a dozen other agencies conducted business. We were the core substrate on which everything else in SMS happened those first few weeks but, because we’re quite technical and rather quiet, very few people know.

A comprehensive account of InSTEDD’s work in Haiti was also attached to Eric’s email and can be downloaded from here. It is compelling reading, and commendable work.

My association with Eric and the founding members of InSTEDD goes back to before the Boxing Day Asian Tsunami in 2004. I am not uncritical of InSTEDD’s work, and the significant concerns articulated in the ICT4Peace Foundation’s paper on the use of ICTs in Haiti hold as true for them as they do for other crisis information management platforms, services and tools. What I appreciate in Eric and several others at InSTEDD is a sensitivity to the vexed nature of designing and deploying ICTs to support humanitarian aid in places like Haiti, often grossly oversimplified and under-valued by a technocratic hubris that shines through in some of the content I have read on the web.

ICT and Protection: Can Information and Communication Technology Enhance Humanitarian Action?

Representing the ICT4Peace Foundation, I will be part of a panel organised by the Humanitarian Law and Policy Forum at Harvard University looking at how, if at all, ICTs have strengthened humanitarian aid. The guiding questions of this web based seminar echo concerns, challenges and opportunities I have repeatedly raised on this blog for years. Some of the more substantive posts, which have been quoted widely have been,

As noted on the website describing the web based seminar,

This Live Seminar will examine questions and challenges pertaining to the development, use and effects of Information and Communications Technology (ICT) in humanitarian activities. Increasingly used by humanitarian professionals in situations of emergency, armed conflicts and disasters, ICT has emerged as a component of effective and sustainable delivery of humanitarian relief. Yet ICT remains relatively under-theorized and utilized differently across contexts. Against the background of the increased use of ICT in humanitarian activities, this Live Seminar will address the following questions:

  • How have technological innovations including crisis mapping, early warning, and crisis informatics shaped the roles and responsibilities of humanitarian professionals?
  • In what ways has ICT affected the selection, collection, and dissemination of conflict-related information?
  • What metrics are available to discern the scope and significance of ICT’s effects on coordinating humanitarian aid delivery?
  • How has ICT transformed institutions providing humanitarian relief?

These questions will be examined through critical inquiry into recent innovations in ICT, their application, and their (potential) consequences for humanitarian professionals.

Naz Modirzadeh (Senior Associate at the Program on Humanitarian Policy and Conflict Research) and Claude Bruderlein (Director of the Program) hosted the discussion.


  • Sanjana Hattotuwa, ICT4Peace Foundation
  • Salem Avan, United Nations
  • Olivier J. Cottray, iMMAP
  • Mark Dalton, ReliefWeb
  • Mike Hartnett, Global Relief Technologies, Inc.
  • Patrick Meier, Harvard Humanitarian Initiative

Haiti and Beyond: Getting it Right in Crisis Information Management


The ICT4Peace Foundation released a new briefing today, co-authored by me, critically looking at the response to the devastating Haitian earthquake in early January. We chart in brief the significant progress made in the use of ICTs in humanitarian aid, and also how much more needs to be done in order to sustain and systematise new developments, platforms and trends.

Read it in full here. As the report avers,

Haiti is seen by many as a turning point in the use of ICTs in disaster response, and rightfully so. However, vital lessons for humanitarian aid and first response clearly identified in the Asian Boxing Day tsunami response remain unheeded, along with points regarding aid work and the use of ICTs enumerated in the UN OCHA +5 symposium report, of which the ICT4Peace Foundation was a key partner. Disaster-affected communities remain largely passive recipients of information, having to deal with, amidst significant trauma, competing information on aid delivery and services. Beyond the hype, the majority of those affected by the Haitian earthquake were off the radar of ICTs. Compounding this, as early as April 2010, Haiti is receding from international media and global attention, yet significant long-term humanitarian challenges on the ground persist. It is unclear how the ICTs first deployed in the country will be sustained over the long term, and in particular international crowd-sourced platforms relying on volunteers. Significant problems of coordination, collaboration and aid delivery dogged the disaster response effort. The Head of UN OCHA, Sir John Holmes, in a strongly worded email in February expressed his frustration over the UN’s aid effort in Haiti, noting that “only a few clusters have fully dedicated cluster coordinators, information-management focal points and technical support capacity” and adding that the disjointed effort is casting doubts on the UN’s ability to effectively provide relief. Beyond the UN, significant concerns were raised over the coordination and collaboration between civil and military actors, and the international community as a whole.

Much more can and must be done to strengthen disaster preparedness and crisis information management. There are no longer excuses for ill-preparedness or haphazard aid response. We already know much of what needs to be done and going forward requires requisite funding coupled with political will of the UN system and international community. Some key ideas and suggestions in this regard are,

  • The accelerated development and population of easily accessible datasets with essential information shared across UN and other aid agencies, to help identify, prepare for and mitigate disasters.
  • Developing ICTs that work better in, and are more resilient to austere, traumatic environments.
  • Significantly improving interoperability across all systems between UN agencies and other key platforms outside, including UN OneResponse, Ushahidi, Sahana and InSTEDD’s Emergency Information Service.
  • Using endogenous technologies, help communities develop their own capacities and capabilities for disaster early warning, prevention and resilience, is vital.
  • Greater cooperation between governments and NGOs, based on standard operating procedures governing information sharing to help aid work.
  • Global and local business, as we have seen in Haiti, also has a key role to play in generating and sustaining financial inflows and strengthening aid. They need to be partners in crisis information management.
  • The development of a comprehensive crisis information management preparedness and assessment tool box, including appraisal mechanisms, especially in and for disaster prone regions and countries.

Social Media in Haiti provides critical information on Haiti’s need

OCHA’s ReliefWeb has (a very rough) transcript of my recent podcast with IRIN on the use of technology in Haiti’s January 2010 earthquake relief effort.

Read it here. Read about the podcast and listen to the original version here.

A pertinent excerpt from the interview:

TUNBRIDGE: Did you see what is going on high-tech world here is unprecedented. Did you think it is going change forever the way we do respond to disasters?

HATTOTUWA: The way we respond to disasters it will always be the same. It will require sweat. It will require physical effort and it will require political will. These three key ingredients are the backbone of any post disaster relief effort and these will not change even as we move into the future. What I think highly demonstrates quite clearly is that technology is the fourth element. We have seen an unprecedented effort with regards to technology deployment to find out very rapidly after the earthquake what the needs on the ground were, where the most urgent cases of aid were in Port-au-Prince and the rest of the country. As well as technology deployments to generate financial aid for the relief efforts. I think Goggle set up a dedicated page very quickly after the earthquake. Apple allows you to donate through Itunes. And unprecedented sum of money that the American Red Cross – the ICRC have got through mobile donations. I think at last count 10 to 12 million US dollars. This has been unprecedented in relief efforts for disasters of this nature and certainly yes it will change the way the world responds to disasters. Because these are now technologies that as we have seen with Haiti can be extremely quickly deployed and we have seen with platforms like USHAHIDI, SHAHANA, INSTEED that is a global community behind these efforts to shape it to translate the systems into Creole and French, to fine tune the systems to the needs on the ground in Haiti. I think this has been unprecedented in the way they respond to the disasters and certainly sets the parameters of what we can expect in the future after disasters such as this.

Haiti and the perennial challenge of information lock-in

Erik Hersman from Ushahidi makes the following pertinent observation, amongst others, in a recent blog post on the Ushahidi Situation Room for Haiti.

“Decision-makers on the ground still do not have access to accurate, real-time data. That may be because of firewalls, lack of bandwidth, people are unaware these resources exist, the command structure of an org does not allow people to use open sources, or the decision makers do not want that data.”

My response to Erik was based on the immensely frustrating phenomenon of locking in vital, life-saving information into formats not easily integrated with other system, accessed, downloadable or mashable. I said,

“One example is the hugely valuable master contact list in Haiti published yesterday by OCHA and available on the OneResponse website as a ZIP download containing an Excel 2007 format spreadsheet with multiple tabs. Far more simpler would have been to just upload this information to the web for people to access and search? In fact, what I did was to save each tab in that huge spreadsheet as a separate file, upload it to Google Docs, publish them as webpages and link to them on the ICT4Peace Foundation wiki… Simple, effective, efficient.”

OCHA’s master contact list that I refer to comes as a ZIP file, which contains a single, very large, Excel 2007 format spreadsheet. I make that point because I know a good many people in Sri Lanka who have older versions of Office / Excel, have not installed updates and thus cannot by default open this file. So while the file was great for offline use for those who could open it and among other uses, for emailing around and uploading to various communities of practice, it struck me as rather odd that this information wasn’t more easily accessible online. So with a minimum of fuss, I created the following:

  1. Primary Contacts in Haiti
  2. Cluster Leads
  3. IM Focal Points
  4. OSOCC – MINUSTAH Base / OCHA – UNDAC team list

Erik’s response to my comment is even more pertinent. It’s a small example of many others I have observed during the first two weeks of the crisis information management response to the Haiti earthquake that I’ll write about in more detail in the coming months.

IRIN podcast on use of technology in January 2010 Haiti earthquake aid effort


UN OCHA’s IRIN news service has a podcast up that features some of my observations on the use of technology / ICTs in Haiti’s January 2010 earthquake aid efforts.

As I was speaking, I kept thinking how much had changed from the response to Cyclone Nargis just two years ago (albeit dealing with a very different political regime), and yet how much more needed to be done to fully address key points I made way back in 2006 on the use of technology / ICTs in humanitarian aid.

The direct link to the MP3 is here.

The podcast also features the extremely insightful observations of Paul Currion, really one of the best in this field. Related to what he says is this recent blog post of his.