The recent Disaster 2.0 Report published by the UN Foundation, OCHA and the Harvard Humanitarian Initiative (HHI) represents one of the most important policy documents to have been written in recent years. The report acknowledges in no uncertain terms that the humanitarian space is moving towards a more multi-polar system and that this represents an unprecedented opportunity for the future of disaster relief, albeit one that presents clear challenges.
We applaud and thank the authors of the report for bringing this to the attention of the policy community. We realize they were under a lot of pressure and had little time to produce this official policy document. Indeed, we understand that the authors and collaborating organizations were in a hurry to publish this report in time for the Dubai Conference to have maximum publicity. We also realize that innovation and learning happens a lot faster than reports can keep up with.
That said, we have a number of concerns about the report. The first is that while some of us were interviewed for this report, none of us were given the opportunity to provide feedback on what was actually written about our work and respective communities before the very public launch of the report. The drafts remained closely guarded and we were not allowed to discuss the report in any detail at the Humanitarian Action Summit. This hardly constitutes the dialogue that the report aimed to create.
Our second concern is the use of the term “Volunteer & Technical Communities.” This term, which becomes the framework for the entire report, is at times problematic since it suggests that the diverse communities involved in the disaster response in Haiti and beyond are similar enough to be discussed using one catch-all label across a variety of issues. The acronym V&TCs is used to make generalizations and as such should be treated accordingly, but generalized comments like “the VTC are struggling to get founds” clearly reveals the myopia of formal institutions when it comes to their reading and understanding of the new multi-polar space.
In addition to this there is a fundamental error in the way the term V&TC is used, highlighting some confusion on the different organizations and networks working in this space. CrisisMappers is a horizontal network of humanitarian practitioners, technologists, researchers and volunteers. Ushahidi is a 501c3 organization that describes itself as a non-profit technology company. OpenStreetMap is a volunteer project, while Sahana is a software company that creates a Free and Open Source Disaster Management system. CrisisCommons is a technical community of volunteers and so on. Placing all these actors in the same basket is not particularly appropriate since some of these organizations/networks are not volunteers.
Our third concern is the personal tone of the report. It does not appear to be objective, choosing to highlight certain issues or groups over others and selecting to promote a particular agenda and set of recommendations that has not been subject to consultation and feedback, and hence does not have buy in from the “new” actors in the multi-polar humanitarian system. We are certain that others in this space will have comments and opinions about the recommendations offered, which it would be important to discuss before moving ahead with any framework for action.
Our particular concern with the recommendations is that they propose creating new structures rather than leveraging existing ones. Why create a new Humanitarian Technology Forum rather than utilise the various existing ones (from RHoK to the CrisisCommons weekly calls)? Why set up a Humanitarian Innovation Lab rather than support and encourage existing groups to continue innovating? To over-organize this community would undermine the creativity that makes it so special.
Fourth, there is little to no credit given to Haitian Diaspora activities and no mention about local response activities such as Noula/Solutions efforts, no mention about the Cite Soleil mapping community. Given the fact that those actors are actually the first responders in major disasters, it is striking that the report does not address the 2.0 possibilities to improve the work with local communities as responders and partners of international humanitarians. On this note, there is no mention of the Ushahidi Haiti Project transition to local ownership, i.e, the Haitian software company Solutions.ht which is now using their own platform on joint projects with the Red Cross, IOM etc. Surely sustainability and local ownership of such projects is key to humanitarian response.
Fifth, the authors of the Disaster 2.0 report may want to explore why Mission 4636 worked well with the military but not with the UN. The report assumes that “interfacing with the UN” is the only measure of success. It is not. And if we did interface successfully with the military, why is the blame put solely on the so-called “V&TCs”?
We have dozens of additional concerns which we’ve listed below. There are also a number of errors both in how organizations are represented and how the history of involvement in Haiti is told. We have not listed these as they would reflect only the stories and organizations we know well. Everyone in this community should get a chance to correct and comment.
In our opinion, making cosmetic changes to the report won’t help. We believe the report, and particularly the recommendations, should be re-written and opened to the rest of the community in the process. That would be a real dialogue. This is why this blog post also appears as an editable Google Doc here. We invite all other communities who were written about in the Disaster 2.0 report to add their observations, especially if they weren’t given the opportunity to review a draft of the report before it was published.
- “Volunteer and technical communities (V&TCs) like OpenStreetMap, Ushahidi, Sahana and Crisis Mappers…”
Ushahidi is a non-profit technology company, not sure that it qualifies as a V&TC, same for Sahana.
- “As organizations, some V&TCs are struggling to attain financial sustainability, especially when asked to respond to successions of major disasters.”
Which V&TC? This is one of the main issues we have with the report. The term V&TC is being used as a catch-all for technology companies, volunteer groups, etc that have notable differences. One should be careful about generalizing like this because it may give the wrong impression about specific groups. For example, financial sustainability is not the core concern of volunteer groups like the Standby Task Force for example.
- “During the response to the 2010 earthquake in Haiti, the volume and velocity of data began to overwhelm this approach, helped by a new dynamic: the rise of the cell phone.”
The rise of the cell phone is not a new dynamic. Cell phones are mobile communication technologies that have been around for a long time now. The new dynamic was not cell phones, the new dynamic was a process not a technology, i.e., the process of crowdsourcing micro-needs assessments by leveraging the ubiquity of cell phones in Haiti. These phones were around during previous crises in Haiti, but no one took the initiative to use them for disaster response.
- “Most importantly, however, there has not been a mechanism for coordinating the collaboration between these two groups and no formal channel for these groups to engage in a dialogue about the underlying problems of information management.”
This is true for Haiti, but not for Colombia and beyond. This is why we shared the full report from the OCHA Colombia earthquake simulation exercise with the authors of the Disaster 2.0 report during the first week of January 2011, i.e., in the hopes that our findings and recommendations would be factored into the Disaster 2.0 report. Our assessment of the Colombia deployment details the mechanism that the Task Force is using to coordinate and collaborate with formal humanitarian organizations. This mechanism was revised several times during the past 6 months and was used when the head of the Information Services Section at OCHA requested activation of the Standby Task Force for Libya.
- “Yet without a formal interface for information exchange with the humanitarian system, or appropriate data standards, this new data-added to the raging river of information that aid workers faced as they struggled to build the relief effort from the ground up.”
Yes, this was true for Haiti, but not for Libya. An interface was further developed by the Task Force in collaboration with OCHA in the months leading up to the Libya deployment. At the very least, the report should include a footnote to this effect.
In addition, OCHA reached out to the various volunteer and/or technology communities in response to the Japan earthquake, e.g., to Crisis Commons about explicit data standards and dataset collection. In both these situations a common Skype chat was the most effective and fast interface for information exchange that allowed almost real time activation and dialogue to go on.
- “As the volunteer and technical communities continue to engage with humanitarian crises they will increasingly add to the information overload problem.”
This implies that in Haiti we were part of the problem, but that is not true. We were not overloading the humanitarians simply because the fact of the matter is that the vast majority were simply not aware of our efforts. Those responders who did collaborate with us identified ways for us to interact with them so we could be most helpful to their operations. In relation to traditional responders, once we established personal contacts on the ground and introduced the system, we quickly identified ways to actually decrease the overload problem.
In Libya, however, the problem was exactly the contrary: lack of information. The overload of information is one problem, but far from the only one. In the Libya situation (and then Japan), CrisisCommons and the Standby Task Force were asked to provide information that the humanitarian community didn’t have or couldn’t collect because limited capacity.
- “The affected population become mobile-enabled.”
They had been mobile-enabled before, it’s just that the humanitarian community never sought to secure a short code and draw on crowdsourced SMS in order to communicate with disaster affected communities.
- “As a result, both the information that V&TCs submitted and the faster methods for information management that V&TCs used and shared only exacerbated the field staff’s sense of information overload.”
As far as the Ushahidi Haiti Project goes, we never submitted information actively or directly in an unsolicited manner. The situation had more to do about part of the humanitarian community simply not knowing about the information source or how to use it. Again, as previously noted, the quote above tries to paint new actors as part of the problem, which was not in fact the case.
- “The affected population became mobile-enabled. At the time of the earthquake, the state of cellular connectivity in Haiti was such that tens of thousands of citizens could, and did, directly communicate their needs to a domestic and international audience, and those citizens expected that a few thousand international responders would be able to monitor their hundreds of thousands of requests for aid and at least deal with aggregate trends, if not individual requests. Some of these increased expectations were created by the V&TCs, who were responding to messages sent over social media.”
There are two fundamental mistakes in this paragraph. One is that affected population didn’t become mobile-enabled. As explained in the following lines the local population was already enabled to use mobile phones. The second one is that V&TCs were not responding to messages directly via social media, but rather only to SMS.
- “During the Haiti response, two new data inflows were added to the system: one from the V&TCs and one from the affected community of Haitians.”
Yes and it was only thanks to these informal actors that the affected communities were able to get their voices out. The Ushahidi Haiti Project and Mission 4636 were the main tools that allowed affected communities to have a real voice during the crisis.
- “The information managers in Haiti had to confront two such ﬁre hoses: one from an emerging V&TC and one from the affected population. These are the subjects of the next two sections.”
No, they didn’t, many were not aware that this information existed (unlike Libya). It was our team drinking from a fire hose, trying to make sense of the incoming flow of information from hundreds of different sources. We were tackling the fire hose head on and filtering & curating the information to provide more structured and actionable data on a live map.
- “4636 Alliance is a partnership of FrontlineSMS, the Thompson-Reuters Foundation Emergency Information Service, InSTEDD, and Internews to offer an SMS shortcode and associated services to Haitian citizens. It was afﬁliated with the Communicating with Disaster Affected Communities initiative.”
First of all, the official name is “Mission 4636”. This list of members is incorrect, the correct one can be found online easily.
- “Ushahidi’s director of strategic partnerships mobilized approximately 200 students at Tufts University Fletcher School of Diplomacy to monitor and geolocate reports from Twitter, Facebook and the 4636 alliance on the Ushahidi platform.”
We monitored over 300 difference sources including Twitter, Facebook, YouTube, Flickr, radio, mainstream news, television, and official UN/Humanitarian reports, in addition to mapping the most urgent life and death text messages from Mission 4636. We never limited our monitoring and geolocation of reports to Twitter, Facebook and 4636 text messages.
- “Based on new code written during the first hours after the quake, the modified Ushahidi platform provided a mechanism for Patrick’s classmates to begin collating social media posts and plotting each message on a map.”
New code was not written during the first hours, all we did was customize the platform which didn’t require new code to be written. New code was written during the second week to add new features deemed necessary to support the live crisis mapping operations.
- “2. Pinpointing the location of a plea for help took lots of time and had to be right.”
That’s a false generalization, we would say that 20% took “a lot of time” but the vast majority were geo-located well under an hour (if not 20 minutes) when we processed them, especially when the OpenStreetMap of Haiti became available. Furthermore, the Haitian Diaspora were doing the majority of the geolocation themselves with an average turn-around time of 10 minutes. That said, we did at times have a backlog of text messages.
- “Working separately from Ushahidi, several NGOs partnered with the U.S. State Department to launch a single SMS number for aid requests in Haiti. These partners, who included Rob Munro of Stanford University, FrontlineSMS, InSTEDD, Energy for Opportunity, and the Thompson Reuters Foundation, worked with cell phone carriers in Haiti to set up a resource called an SMS shortcode.”
This is false. Josh Nesbit from Medic:Mobile (formerly FrontlineSMS:Medic) sent a Tweet that he was specifically looking for an SMS gateway for http://haiti.ushahidi.com. Someone (in Cameroon) saw his Tweet and put him in touch with Digicel. Within an hour, Josh was on the phone with a contact at Digicel and the latter agreed to provide a short code on the spot. Later, the State Department provided an official letter to Voila-Comcel so that this telco could also provide the 4636 short code on their network because they were at first hesitant. Josh then reached out to Robert Munro at Stanford. Erik Hersman and Brian Herbert from Ushahidi were part of the early conversations on 4636.
- “Messages began to flow into 4636 almost immediately.”
There is no mention in this paragraph or report that it was thanks to Brian Herbert of Ushahidi that we had a new (built over night) platform that allowed volunteers to log in and translate & geo-locate text messages coming from the 4636 short code. These would then be pushed to the backend of the Ushahidi-Haiti platform for review, geo-location verification and publishing. It was only 2 weeks later that Crowdflower and Samasource entered the picture with their micro-tasking platform and training.
- “But the newer V&TCs do not yet have the resources to mobilize large numbers of volunteers, nor do they have the capacity to send small teams to the field for every disaster.”
The Standby Task Force response with OCHA in Libya makes this statement false. The Task Force has almost 500 trained volunteers from over 50 different countries.
Also, noting that we do not have the capacity to send small teams to the field assumes that this is part of the mission of all these V&TCs which is definitely not the case. (This again explains why the use of the catch-all term V&TCs is a disservice). Much of the novelty and power of the new communities in this space is precisely the fact that a lot of the work can be outsourced and done remotely. There is nothing in the Mission of the Task Force about our deploying to the field. Indeed, as our byline reads, we are an online volunteer community for live mapping. So writing that we do not have the capacity to send small teams to the field reveals that the authors of the report don’t understand who we are and what we do.
- “The CrisisMappers community is forming a team called the Standby Task Force, composed of experts who are willing to train and deploy to emergencies to provide broad support for imagery, mapping and crowdsourcing.”
The CrisisMappers community did not form the Standby Volunteer Task Force. The Task Force was launched during the 2010 Crisis Mapping Conference but many were critical of the way that we proposed to move forward. That is why the title of our online volunteer community is “Standby Volunteer Task Force” and not “CrisisMappers Standby Volunteer Task Force.” The mission of this online volunteer community is to provide live mapping support to organizations that request help. We do not deploy to emergencies but provide remote live mapping support. The Task Force is composed of hundreds of volunteers, many of whom are skilled professionals and students in various fields. See our Volunteers Profiles section for more details.
- “[Ushahidi] morphed from a blog to an SMS-based platform to facilitate submissions of reports from the field.”
The Ushahidi platform was not originally a blog and is definitely not limited to SMS. Information can be sent to the platform for mapping via: webform, email, SMS, smart phone apps, Facebook, Twitter, Check-In’s, voicemail, etc.
- “When some V&TCs did not have as strong showing a in their support in the 2010 floods in Pakistan many veteran humanitarians were puzzled.”
It’s important to note that the Task Force didn’t exist at the time, but even so, core volunteers from Ushahidi-Haiti and Ushahidi-Chile were directly engaged in the PakReport.org crisis mapping efforts. In addition, Andrej Verity of OCHA got solid support from requests he sent to the Crisis Mappers Network. Again, it’s important not to suggest that all V&TCs are alike or respond in similar ways.
- “Some newer V&TCs have begun debates between internal factions, often between technologists who wish to provide platforms for general use and crisis responders who wish to support the application of these platforms to specific humanitarian operations. Some have debates between factions who believe one technology or method is being advocated above others. Others worry that the organizations will need to make fundamental changes to their mission and organizational design to chase the resources necessary for transforming a volunteer organization into an institution that performs a critical role during a disaster. These debates are ongoing. There is also a real possibility that competitive dynamics—if taken too far—may start to damage brands which the V&TCs have begun to build within the international humanitarian system.”
This sounds like a very judgmental personal opinion. We have no problem with it if it is supported by real examples supporting this statement otherwise it sounds like a very broad generalization that cannot be used to support or start any constructive dialogue about it. With regard to brand being built, this is also somehow unclear and should be explained in details or deleted.
- “V&TCs are instead accustomed to shaping malleable architectures of software code.”
It is very problematic to use V&TCs as a catch-all and then to generalize across different groups like this statement. At the Standby Task Force, 90% of volunteers are non-techies, non-software developers, they are crisis mappers.
- “Many V&TCs have little experience with working on megadisasters.”
In this statement and throughout the document there is a confusion between groups / networks and the individual volunteers that compose them. One of the core strengths of the V&TC groups / networks is precisely that some of their members are experienced field responders. They have experience in disasters, and they share this knowledge in a collaborative way with other volunteers. We have seen this in recent Task Force deployments.
- “Competition and Collaboration. In the V&TCs, competition has emerged as a disruptive dynamic. Communities that need each other’s strengths are being force to compete for money in the donor pool, and some are ﬁnding the need to compete in areas of overlap or are being asked to tackle issues where other V&TCs are already working. Some competition is good; but coordination of effort needs to be put in place so that scarce resources are not wasted on work which is already complete or better done by other organizations. also ensure technology is in the hands of practitioners before the disaster.”
The competition issue should be addressed more accurately and constructively. OpenStreetMap and the Task Force are not competing, for example, they are collaborating as partners. That is why the Task Force represents a coordination effort with CrisisCommons, OSM, Sahana, etc. The Task Force is not looking for funding, so the donor pool comment is inaccurate as a generalization.
In addition to the notes above we would also like to add one more issue that we found very surprising in the report: Crisis Commons, which is one of the main and most active V&TCs group, and which was incredibly active in the Haiti response as well as CrisisCamps, are hardly mentioned in the report. Crisis Commons posted a list of all the projects they have worked on and have a detailed descriptions of their activities during the Haiti Crisis.
In conclusion, we want to reiterate our opening comments. This Disaster 2.0 report is an important policy contribution. It establishes that so-called V&TCs have a positive role to play in the more multi-polar humanitarian system that is evolving.
Thank you for reading.
Contributors: Patrick, Jaro, Rob, Anahi and Helena