The Sahana Singularity

Two reports came out last week that relate in varying degrees to the work of the Sahana Software Foundation. These reports help lay out who we are, what we have accomplished, and where we are going. They are important reading for our community.

Volunteer Technical Communities: Open Development

The first comes from the World Bank’s GFDRR Labs entitled  “Volunteer Technical Communities: Open Development”. The GFDRR (Global Facility for Disaster Risk Reduction) Labs is a small group within the World Bank, whose mission aims to “leverage public-private partnerships, open data and open source technology to develop practical tools to assist in decision making. To source innovative solutions, the Labs engage a global network of collaborators and thought leaders, including grassroots and Volunteer Technology Communities (VTCs).” Great stuff. I look forward to finding projects we can work on together.

While the work of the Sahana Software Foundation was not featured in this report, which was based on research completed way back when in October 2010, it did have the foresight to acknowledge that and gives us a nice shout out:

“This list is not meant to be comprehensive. Many other communities made significant contributions to operations during the past year, including Sahana…”

Sahana was also mentioned in the section on Open Street Maps, referring to our collaborative work to geo-locate hospital status information during the Haiti earthquake response, and with Crisis Commons as one of the projects that CrisisCamp volunteers supported. Such mentions are important references for us or any organization. Being left out of such a report, whether intentionally or by oversight, can have a damning impact on future charitable funding and project opportunities. In that context, I appreciate the mentions made in this report  and I think that this community has gotten much better at being sensitive to being inclusive and positive as much as possible.

Disaster RELIEF 2.0: The Future of Information Sharing in Humanitarian Emergencies

The second is the long-awaited and eagerly anticipated report from the UN Foundation and Vodaphone Foundation Technology Partnership, Harvard Humanitarian Initiative and UN Office for the Coordination of Humanitarian Affairs (OCHA) entitled Disaster Relief 2.0: The Future of Information Sharing in Humanitarian Emergencies. The authors and researchers of this report are well-known to the Sahana community and know us well too, and in my opinion they did a great and balanced job at seeing through a lot of the bluster into the effective approaches to providing information management support to humanitarian operations coming from the Haiti experience. And fortunately, a lot of what the Sahana Software Foundation has been doing and advocating is really at the forefront of their findings and recommendations. This is very heartening.

What is somewhat disheartening is the waves of criticism now inundating the authors and researchers of the report that started interestingly enough on the Standby Task Force’s blog.  Gisli Olafsson’s well considered reply is important to track here  as well.  Most of the subsequent discussion relates to who was and was not mentioned in the report, and also the nature of what the term “V&TC” really applies to and whether only volunteer organizations have any value. I honestly see no need to enter myself into those debates.

A fair assessment of the report is that it is a fair assessment. Like all reports, it is imperfect.   It may overgeneralize, it may omit mention of some important actors, it may even over-prescribe a solution…. these are all reasonable comments and to be expected.  But the findings and lessons learned are rock solid and as even the SBTF blog begins:

“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.”

All those who contributed to the report should justifiably be proud of their work. I hope that the ongoing discourse can be positive and join the call of the authors that the report provides “a useful starting point for discussions, and that they spark considerable debate.” (For that last part, mission accomplished!)

Anyway, there are four main points I want to pull out and share with the Sahana community about this report and what it says about us and what it means to us:

1) It is a clear call for financial support for our work

Let’s begin right at the beginning in the Executive Summary. All you need to know about how the report values our work is right there on the first page (well, the second page actually… make that page 9 of the final report with all the beautiful graphics and excerpts). Let me give you an excerpt:

“[N]ew partners are offering faster, more effective means of analyzing an ever-increasing volume and velocity of data…. Volunteer and technical communities (V&TCs) like… Sahana… approach problems in ways that challenge the status quo. As organizations, some….are struggling to attain financial sustainability, especially when asked to respond to successions of major disasters.”

That’s really important stuff for us. While the quote mentioned other humanitarian organizations and groups such as OpenStreetMap, Ushahidi and the Crisis Mappers, the excerpt is not taking anything out of context here. Please read the full report. It deserves it.

This summation of their findings is nothing less than a call for funding the Sahana Software Foundation’s work in order to make it sustainable and effective.

2) The importance of being a part of the system

The report highlights the importance of contributions coming from organizations that are trusted by the relief and response community. In identifying the challenges in dealing with the gaps between the capabilities available and the system’s ability to absorb that assistance during the Haiti response, the authors noted:

“…. the humanitarian system had no formal protocols for communicating with these volunteer and technical communities (V&TCs)…. their institution’s policies and procedures were never designed to incorporate data from outside of their networks…. the rate of investment in humanitarian information management over a complex global network is failing to keep pace with new technological realities. The humanitarian system could use…. the evolution of systems for widely sharing data during the response to a disaster to make faster, better decisions in emergencies….

This report sounds an alarm bell. If decision makers wish to access to (near) real-time assessments of complex emergencies, they will need to figure out how to process information flows from many more thousands of individuals than the current system can handle. We need to fundamentally rethink how the humanitarian system manages information in light of the increasingly complex system of networks and data flows.” (pp 11-12)

So clearly, the challenge has been issued to the traditional humanitarian response community, represented most symbolically here by OCHA. It is clear that they have already responded to this challenge. OCHA and other humanitarian organizations are actively engaging humanitarian technology organizations in dialogue about how to move forward with these relationships and the Standby Task Force for one has already shown how this can be done with its response to a request from OCHA to assist with the humanitarian response to the civil conflict in Libya. But the challenge is not just for humanitarian organizations, as the report notes:

“As the Volunteer and Technical Communities continue to engage with humanitarian cries they will increasingly add to the information overload problem. Unless they can become a part of the solution.” (p. 18)

So how do we become part of the solution? I can tell you we are on the right path and the next couple of points address that in more detail.

3) Call for Professionalization

This is also an issue which I have been espousing for some time now and is at the heart of the strategic planning initiative for the Sahana Software Foundation that I initiated at the beginning of this year. The need to professionalize and be professional about what we do. Part of that approach is detailed in my article on Developing a Service Industry to support Sahana software that was published in the Open Source Business Resource last December. The other part is very internal. If we want Sahana software to be successful, we must be successful as an organization even if and (maybe especially) while operating on a volunteer basis. And to be successful as an organization, we have to act as one.

The Disaster 2.0 report’s call for professionalization includes aspects which we have been advocating for Sahana and other HFOSS organizations, including:

• Humanitarian Principles and Code of Conduct – Chamindra de Silva and Paul Currion wrote the original HFOSS Code of Conduct which we have been working on refining for much of the past year through the humanitarian-foss working group. I consider the current draft pretty close to being ready for organizations and individuals to formally sign on. While there are other Codes of Conduct floating around – most are similarly modeled on the Red Cross Code of Conduct – I believe that this HFOSS Code of Conduct is the one most appropriate for consideration by the Sahana Software Foundation and other open source organizations. I’m hopeful that we can consider this for formal adoption at our annual meeting in a month’s time.

• Capacity Building – this point in the report beautifully stresses the importance of organizations building their financial and governance structures in order to become viable partners with humanitarian organizations. This is exactly why we have been going through a strategic planning process for the past few months, reviewing our organizational structure and what we need to do in order to become an organization that can sustainably deliver on our incredible promise and potential. We are not there yet, but we are getting there.  One way to measure capacity is through a Maturity Model. This is also something that many in our community have been working on. We started with a brainstorming session at the RELIEF event last February in Monterey, which included Chamindra, Eric Rasmussen (SSF Member and then CEO of InSTEDD), John Gage (formerly of Sun), Ray Buettner (Naval Postgraduate School), John Crowley (NDU/HHI), and some others. The current H-FOSS Maturity Model needs some work but the idea is to create an objective means for organizations to self-assess their readiness to participate in relief activities as partner’s of humanitarian organizations on the basis of the software and overall solution, the people and organization supporting it, and of course, their experience.

• Service Level Agreements – professionalization leads to the ability to put in place service level agreements, of the sort that MapAction has held with the UN system for years now.  In fact, I am quoted here in commenting: “Sustaining a volunteer effort is extremely difficult. This was as big and bad a disaster as you could imagine. But beyond 2-3 weeks, most volunteers need to go back to their jobs, their spouses, and their kids. Sustainability was an issue. Partnership agreements that can guarantee support and service can help address this problem.” (p.50). This is all a logical extension of capacity building for our organization. It won’t happen over night, but funded organizational growth will allow the Sahana Software Foundation to offer service level agreements to humanitarian organizations and give them confidence in the use and adoption of our products. This means having the resources (i.e. paid staff) to be able to meet peak demands for support – both remotely and with boots on the ground.

It has been amazing what we have accomplished as a community over the past six years. At times under the Lanka Software Foundation, we’ve had funding for developers to work on the core architecture and features and offer support and enhancements. For the past couple of years as the Sahana Software Foundation was established, we have been surviving on a volunteer basis. That is not sustainable any longer. I think everyone realizes that. It certainly is not for me personally.

And the expectations for what we can do are so high because of what we have accomplished in the past couple of years; from the incredible response to the Haiti disaster to highly successful projects with the City of New York, the National Library of Medicine, and new opportunities with IFRC’s Asia Pacific regional office, with the Asia Disaster Preparedness Center in Bangkok, the National Disaster Relief Services Center in Sri Lanka, and the City of Los Angeles, who just issued an RFP for Sahana customization and services. There is no going back to being a small project. We are now a global movement.  If we can successfully professionalize what it is we are doing, we will be able to be trusted partners of the relief and response community.  (See how I stepped onto my soap box into the middle of this post!)

4) Importance of open source, open standards and the effectiveness of Sahana’s solution

The Disaster 2.0 report contains a sidebar case study on Health Facility Mapping with Sahana, OpenStreetMap, and Crisis Mappers on p. 29. John Crowley and I spent some time reconstructing this story for the report from e-mails to the CrisisMappers mailing list, along with personal recollections and private IMs on Skype received from Eric Rasmussen who was in Haiti at the time. The challenge we issued to crowd source the geo-location of all known hospitals in Haiti is documented on our wiki and written up in the case study we published for the 2010 ISCRAM conference.

The key point is what we did with the data once we got it into the Sahana Eden Haiti response portal:

“What happened next is critical to lessons learned from the response. Sahana made the data available in open data formats via several feed formats, including XML, KML, GeoRSS, and the XML schema designed for tracking hospital data, EDXL -HAVE. This resource became one the best resources for health facility data for the next month. Over 8,000 unique individuals visited the site or pulled from these feeds. Crowdsourcing had taken a responsibility that would have taken OCHA days to complete and reduced it to a little more than a day of work. In the process, a group of V&TCs had built a process for locating health facilities—a process that is now being revised for the next disaster.” (p. 29)

As mentioned in Takeaways from Camp Roberts East on the Sahana Situation Room blog, we indeed are continuing to work on the hospital management system to turn it into something appropriate to serve as an global open data repository of geo-located hospitals and medical facilities. It coincidentally almost exactly matches the requirements outlined for a Global Health Facility Database Initiative that came out of the Annual World Health Care Congress in Washington last April documented on the Pacific Disaster Center’s website.  A more detailed post on this system was promised and is up next on my agenda. But Michael Howden recently gave a talk to the Pandemic Preparedness Forum in Bangkok, Thailand on Sahana that details some of these plans.

So we are on the right path.

Afterward – the Sahana Singularity

Some have already commented that these two reports already seem a little dated, so quickly is the growth of capabilities and new volunteer and technical communities such as the Standby Task Force. Patrick Meier wrote that “One could already write an entirely new UN/Vodafone Foundation Report on just the past 3 months of crisis mapping operations.”

Time Magazine recently had a fascinating cover story entitled “2045: The Year Man Becomes Immortal”  based on Kurzweil’s theory that we are rapidly approaching an event referred to as “the Singularity” where technical and computing capacity will exceed that of all human beings combined.   (The singularity defines the point at which the future becomes “qualitatively different and harder to predict.”)  This event may be like Terminator’s Judgment Day or much more benign, when we can leverage machines to discover the cures to all diseases and causes of human mortality. IBM’s Watson may be Jeopardy champ in 2011, but in 2023, the computer will surpass the brainpower of a single human being and beyond that the limits are not foreseeable as to what may be achievable. These infographics are helpful to understanding what this is all about.

The same is happening with Sahana as well. As we build capabilities and learn from experience and see the numbers of people interested in applying technology to humanitarian situations such as disaster relief as a social good, there is an incredible opportunity in front of us. But also many risks if we do not learn to leverage and grow our community and opportunities.

I wanted to keep this post short. Alas, brevity is not my strong suit.

Go forth and do good!

Musings from Tech@State

I had the pleasure of attending the Tech@State event at the US Department of State last Friday (11 February 2011).  The theme was open source. Thanks to Noel Dickover who invited me to speak on the panel on Open Source to the Rescue: Disaster Response and Humanitarian Assistance.

Heather Blanchard (CrisisCommons), Deb Bryant (OSUOSL), Mark Prutsalis (Sahana Software Foundation) and John Crowley (Harvard Humanitarian Initiative) at Tech@State

Big shout out to Deb Bryant of the Oregon State University Open Source Lab who gave Sahana a big shout out in a general session…. along with Lin Wells of NDU and Star-Tides, who also noted Sahana’s role in the Haiti response. Thanks guys.

The rest are some random notes and thoughts I noted on my iPad throughout the day (using Evernote, which seamlessly syncs and categorizes and tags notes between my iPad, Nexus One and Windows 7 Thinkpad T510… what a lifesaver… and it’s free software!!!):

Thoughts from the General Sessions:

• Aneesh Chopra, CIO of the United States, quoting the President: “I want a government more open and transparent.”
• The Executive Branch panel speakers (Aneesh, Macon Phillips, New Media Director of the White House, and the very impressive Todd Park, CTO of US Department of Health and Human Services) more about open data than open source.  Oddly odd. No problem. I can live with that.
• Someone later noted that the President recently gave a speech at the UN and devoted about 50% of it to asking the world to join in philosophy of openness. Cool. Get the quote. Tag it. Bag it. Put it on our web site.
• Aneesh described a voluntary industry driven standards process that is market and commercially oriented as the way forward. So be it. This means that the commercial-partnership approach suggested by the OSBR article I reposted here is more critical than ever.    We also need to see Challenge.gov as an example of how the US government is rewarding innovation through a competitive marketplace. Is the Sahana Software Foundation prepared to compete seriously in that marketplace? Hmmm…. It’s worth looking into.   I definitely need to spend some time on challenge.gov and see if there is something that we could hit out of the park. I’d be surprised if there is not.
• Ben Shneiderman of the University of Maryland, a colleague of Louiqa’s suggested that user interface standards are as important as data standards to ensure commonality of user experience across different open source solutions…. We need to check out usability.gov for more info and support these initiatives. (Is there an idea that is not now a .gov?). Based on a thorough 10-second look at this site, I’m pretty sure there is a ton we can learn and benefit from this initiative. Thanks Ben
• “Coopetition”: I hadn’t heard this term before, but I understand it’s been around for a bit (since 1913 according to wikipedia). I guess I need to attend more unconferences to keep up on the latest lingo. Love it. Best I’ve heard since “data esperanto” from Pulse Camp.  What about “collabotition?” – even better. But I’m pretty satisfied that we in the Sahana community are all about cooperation and collaboration with our external partners, and coopetition and collabotition only with each other.   Hopefully “civil” continues to be a modifier as well.
• Recall in 2004, the Office of the Management and Budget (OMB) issued  Memo 04-16 indicating that open source should be considered the same as proprietary software. A good step.
• A better one came in 2009, when a Department of Defense directive went further to suggest that open source should be preferred and encouraged DOD to participate in open source communities. More tag it bag it material.
• Gunnar Hellekson of Red Hat and Open Source for America declared victory before citing poor adoption statistics within the USG of open source outside of DOD (who tries everything once, let’s face it), and DOE (with all their scientists).
• Greg Elin of the FCC noted that the open source community needs to “climb the mountain”….. meaning meet us (govvies) half way by developing hooks into legacy proprietary systems. This will help ease the transition from proprietary to open source. So the message is clear: we should stop complaining about the lack of APIs and make our systems work with legacy systems.
• One more .gov site to spend some time on: Apps.gov

Funny comment from the GIS break-out session: In explaining the poor resolution of a picture on one of his slides, Jeff Warren of Grassroots Mapping noted: “It used to be a video.  Now it is a PDF.  Sorry.”   That’s how I feel sometimes… it used to be data… now it’s a pdf… but the next word is not “sorry” but something unprintable.  Sorry.

Open Source to the Rescue: Disaster Response and Humanitarian Assistance:

We had a great panel discussion in the afternoon on the use of open source for disasters. Ryan Ozimek, the force behind all of the snowmaggedon sites and the CEO of Joomla kicked it off with a fascinating presentation on the crowdmapping efforts that demonstrated the potential of using such platforms to connect neighbors in mutual assistance (or the occasional snowball fight).

I followed with a great 6.5 minute presentation on Sahana. Check out the slides below.   I don’t know, but I think all of the work on strategic planning is really making our mission and path forward clear… and it is compelling.  (Yes, I know you’ve seen the slides before, it’s the story that goes with it that matters… next time I’ll record a voiceover).

John Crowley then briefed on some of the amazing insights that have come out of the RELIEF/Camp Roberts experimentation process.

Finally, Heather Blanchard of Crisis Commons reviewed the lessons learned from the Haiti earthquake experience (link to her slides)…. I recommend a thorough review of the slides, because there is nothing there that I disagree with.   We need to document these and build on them as learning objectives for our own strategic planning.

The discussion that followed was lively and productive. I should mention Jon Gosier of Ushahidi’s SwiftRiver project was also there… great to meet him and we both talked very positively about the collaboration between Sahana and Ushahidi and OSM on interoperability and using common data standards, and a common humanitarian spatial data model.

I would have liked to have heard more from those from the US State Department, USAID and its Office of Foreign Disaster Assistance…. they were there in the room in force, but outside of the State Department’s Humanitarian Information Unit team, they remained mostly silent. I’m hopeful they returned to their offices Monday to write the appropriately supportive memos to the higher-ups on higher-floors. But some discussion as to how they could support our efforts – not even by using our solutions themselves, but simply by recommended us to governments and local emergency management authorities in the countries affected by disasters as options would serve the stated goal of the Administration to promote open source and open data. Well… maybe they read giist.

So, that’s it from Tech@State. I’m energized again and very glad to have spent the day with a lot of govvies who support open source.

Go forth and do good!!!!

Reflections on Pulse Camp: Lessons Learned for the Sahana Software Foundation

I was honored to be invited to attend the inaugural Pulse Camp that took place in New York City on December 1-3, just prior to the Random Hacks of Kindness 2.0. Sahana Software Foundation director Chamindra de Silva was also invited but unable to attend due to the travel costs involved. The purpose of Pulse Camp was to bring together stakeholders and thought leaders to help UN Global Pulse design “a new, global system architecture for its real-time data & analysis technology platform.” The initial scope of Pulse Camp was laid out by a couple of blog posts by UN Global Pulse director Robert Kirkpatrick, that are worth a read.  The first, “Self Assembly Required: A Real Time Platform for Global Pulse“, detailed the real-world requirements for the Global Pulse platform, while the second, “Global Pulse: An Open Architecture for Sustainable Innovation“, began to define the technical architecture of such a platform.

I have known and worked with the Director of UN Global Pulse, Robert Kirkpatrick, for over 10 years now, dating back to his time with Groove Networks and mine with the US Department of Defense as a contractor. Robert has a keen mind for humanitarian software design and has always been willing to take on big strategic analytic projects; UN Global Pulse is a great fit for him.

Ultimately, Pulse Camp was a big strategic planning session for UN Global Pulse, the third time in the past year that I have participated in such an event for another organization.  The benefit to the Sahana Software Foundation is admittedly limited, outside of the partnership with a UN program and relationships that come out of the attendees, but some ideas came up that I wanted to share with the Sahana Software Foundation community.   And also, by now, I have a great understanding of what we need to do to organize our own strategic planning workshop that I hope to schedule for the first quarter of 2011 with key stakeholders.   It was nice to be mentioned  in their post event write-up and included in their retrospective (see below).  I hope that my contributions have a positive impact on UN Global Pulse and I look forward to continuing to partner with them on areas of mutual interest.

Pulse Camp retrospective video

So here’s the thoughts I wanted to share with everyone that came out of Pulse Camp:

Dangers of Total Information Awareness: The core of what Global Pulse wants to achieve is to provide tools that allow analysts to look at large amounts of public and open along with sensitive data sets – that may be available only to government agencies or commercial companies – together to identify indicators that allow governments to intervene early to avert or mitigate crisis. Its intentions’ are wholly good – for example, Global Pulse might identify rising milk prices in poverty stricken countries as an indicator that families can’t afford to buy milk for their children – thus suggesting a government-subsidy program to be initiated before malnutrition sets in. But it reminds me a lot of the Total Information Awareness or TIA  program launched by the US Defense Advanced Research Projects Agency in the early 2000s as a means to combat terrorism but raised many Big Brother objections on a civil liberties basis. [This is something I actually have a lot of first-hand experience with, spending several years in the early 00’s funded by DARPA’s TIDES program which was a part of the TIA program].  There is a lot of data that is available to governments – efforts to collate this data together in ways that can make connections between agricultural data and economic data and health data might make sense, but can easily be misused.

Systems can be designed to do good, but can be used for other purposes. It is easy to see how Humanitarian Free and Open Source Software can be misused. This lesson is valid for Sahana software as well, as we see Sahana software applications as aggregaters of multiple sources of data as a means for giving disaster responders the best information available for decision-making:  But, are you registering victims at a shelter, or prisoners at a detention center? Collecting missing persons information or those wanted for political crimes? Registering volunteers or child laborers? The software can do all of the above.

Tracking credit card, spending data, cellphone and text messaging data might provide a real-time early warning indicator, which is something that Global Pulse is interested in doing, but it can also be used for insider trading-like advance warning on market indicators; analyzing phone records and SMS data might show locations of interest, or political dissonance. And we haven’t even mentioned the potential privacy concerns that are raised by the sharing and aggregating of large amounts of until-now-disconnected data sets. The potential for misuse is grand and labeling it as a humanitarian UN project is not going to be enough. Significant safeguards will need to be put into place to ensure that the tools developed are not misused.

This made me think of the terms of service agreement which used to be a part of the download page on Ushahidi’s web site. It used to include a checkbox where they asked you to confirm as part of a terms of use that the software would only be used for humanitarian purposes. However, there was always (as there should be) a short-cut direct to their launchpad account where one could download the software directly without confirming any such agreement. I had considered this a good idea for Sahana software as well, but realized that it conflicted with the terms of the open source license under which Sahana software projects are distributed by imposing additional conditions on the use or reuse of the software. So, unless the Sahana Software Foundation elects to create its own open source license and seek OSI approval (and I’m not suggesting this… at least at this time), similar to how the Apache Software Foundation created its own license government its software projects, there is little we can do legally to limit the use of our software for good and not for evil.

That said, Ushahidi is maintaining a voluntary registration system as a means by which to track deployments and to contact users as required.  Their download page reads:

To request a download of Ushahidi source code to run on your own server, please provide the information below. We will send you a link to download a tarball of the application. Only the email and name field is required, but filling them all in will allow us to update you when a new version becomes available, collect feedback and help us keep track of deployments.

This is something the Sahana Software Foundation should do as well – to track deployments and users and create an announcement list (not a discussion list, outgoing only and only for important announcements such as new releases being available).

Data Esperanto: we may have just coined a new phrase at Pulse Camp. To address how we get a single data model out of several sources – many into one and try to make sense of them. This is one of the goals of Global Pulse and is similar to how Sahana Eden has been designed with S3XRC engine in the middle, with REST controller and yummy consumable feeds at its core. UN Global Pulse is to be open source and open standards – similar to Sahana software projects, so there is a core kinship involved here. This ties in particularly with the following concept:

Global Directory Server: One of the ideas I brought to the Pulse Camp discussions was one of the concepts for Sahana software applications that came out of discussions at the RELIEF experiments at Camp Roberts last year – and initially proposed by Chamindra de Silva – that of supporting a global directory server that would record the location and other metadata about different Sahana servers. Extended to the Global Pulse infrastructure, there was definite interest in having Sahana instances register as data providers with the Global Pulse network, along with Ushahidi and GeoChat and other humanitarian FOSS systems. The basic idea is that Sahana servers should be able to register with Global Pulse as a data provider and as an available tool for data collection.

But returning to our own needs, I think the time is ripe to begin working on this concept and implementing it for our own applications, along with an app-store/module-manager-type architecture (see next). Such a service could provide the backbone of Sahana mobile applications and better support autonomous deployments and use of Sahana software in post-disaster environments.

Sahana Zero-Configuration Architecture Design by Chamindra de Silva

Here’s how part of it would work: as part of the installation process for any/all Sahana server instances, the user would be prompted to register the server’s details with a global directory server, defining what emergency or country it is collecting data on, what modules/registries are being used, what are the data outputs (feeds) available (kml, edxl, rss, etc.), how to reach it (ip), and its data feeds, how to push data to it and pull data from it for mobile devices/applications, etc.

This would enable us to build out the following types of capabilities. It would:

• Allow disaster responders to find Sahana servers with data of interest to them from a single portal.
• Allow Sahana servers to find each other, enabling full or selective synchronization of data between them. (and helping head off the overdeployment of Sahana servers for a single incident (see Stop Ushahidi before it’s too late).
• Enable Sahana servers to be discoverable services on a local or mobile network.

Implementation of standards such as EDXL-DE would allow all Sahana servers to track the provenance of the data (where each record came from) to allow for better filtering, analysis, and to implement controls (as needed) over the redistribution of the data.

But there needs to be more than one directory server. While maintaining a global directory server is of great value alone and would be something hosted on the Sahana Software Foundation infrastructure, there is also a need for a federated network of directory servers, which could support both local network and private network implementations.

The App Store / Module Manager / Plug-In Manager: This brings us to the concept of an App Store for Sahana software projects, a concept that was worked on for Sahana Agasti during the past Google Summer of Code and brought to RHOK 2.0 by the CUNY SPS team.  Having a modular architecture has always been a part of Sahana’s grand design, but a true plug-in architecture has not yet been a part of any Sahana software projects. As the tools and capabilities of Sahana software projects continues to grow, there will be a greater desire to not just turn-off or hide applications not being used, but to not have them installed at all. Free-standing Sahana apps (like Sahana OCR – which runs wholly independently of a Sahana server instance) are a first step in this direction, but need to be more integrated with the technical project’s releases as a true plug-in or optional module to install through a standard installer script.

And in keeping with a vision for the Sahana Software Foundation – and this is my vision – (not to claim ownership but to avoid the assumption that this is necessarily shared by everyone) – of it being a home for multiple humanitarian FOSS projects, akin to the Apache Software Foundation with its over 100 projects, an app store / module manager / plug-in manager are all concepts that need to be a part of our strategic plan for future growth.

All that by way of introduction, I led one working group session at Pulse Camp that sought to identify the characteristics of an App store for UN Global Pulse. We had a bit of difficulty thinking outside of the box and getting beyond our collective experience with the iTunes store – and even the Google Android apps market. But there are many concepts and ideas worth borrowing from, noting and considering for our own purposes:

• Ability to register an app (this ties into the global directory server that tracks both apps and servers)
• Ability to make it discoverable (either globally through an app store, or locally as part of the federated deployment network); this is particularly important for mobile networks and devices where smartphone or java-capable cellphone users could be notified and prompted to download an app based on their preferences or location services.
• A lifetime manager (tracks versioning, including how recently the app has been updated, as well as usage statistics)
• A reputation rating (based on who is using it, as in the organizations (local, national, UN, charity), as well as how many overall users)
• Meta data about app (such as tags, categories, standards used, etc. that better allows people to find the app that they need)
• A recommendation engine (that will suggest apps to users based on who is using similar things based on the emergency or other registries active)
• Any restrictions on use (either of the app itself or the data likely to be contained therein due to legal or privacy concerns; for example, the use of a mobile app designed by a national ministry of health to collect biosurveillance data in a particular country might be restricted to staff of official partnering agencies)
• A federated system (which would allow local servers to run their own app store)
• A standards adherence score (based on the number of open data standards supported by the app
• An expert support/advice/comments capability allowing multiple reviewers of an app to make comments about its use and effectiveness.

CODS: A new initiative endorsed by the UN Interagency Standing Committee on 1 November 2010 to provide a Common Operational Datasets (CODS) in Disaster Preparedness and Response.  While nascent at present, it promises a means to build out data schema for different registries that meet the requirements of UN agencies and NGOs working within the UN system. This is an opportunity for the Sahana community to engage with UN and humanitarian agencies, because any solution we offer for managing shelters or food aid distributions or health assessments or logistics systems are going to have to be compliant with the CODS or there will be a technical barrier to adoption.

There was a great one-hour session held at the end of Pulse Camp that I participated in with senior officials from UN OCHA, UNICEF and UNHCR where this was presented. We discussed ways to implement these standards with humanitarian voluntary technology projects and communities like Sahana. One of the suggestions I had made was to work together to bring these standards to the W3C Emergency Information Interoperability Framework Incubator Group, which is now organizing its next phase as a working group.

Final thoughts: A common theme of a lot of a lot of conferences and meetings I’ve been to over the past year that was also featured at Pulse Camp is this what should be rather obvious thought: The best tool is the one people are already using. If they have to adopt something new, it is already too late. The challenge for us is how do we make Sahana software the tool that people are already using?

That’s it from Pulse Camp.

I’ll migrate some of this over to the ideas: and/or standards: namespace on the Sahana Software Foundation wiki so we can build upon these ideas as part of our strategic planning.

Go forth and do good.

Developing a Service Industry to Support the Sahana Disaster Management System

Reprinted/republished from the Open Source Business Resource, December 2010.

“No innovation matters more than that which saves lives”

Avelino J. Cruz, Jr.,
Secretary of National Defense of the Philippines,
on the use of Sahana following disastrous mudslides in 2005

Abstract

The Sahana Software Foundation governs humanitarian free and open source software (HFOSS) projects that address the information-coordination challenges of disaster management. Sahana software has been proven effective and is recommended by numerous industry sources from both the emergency management and business sectors. It fulfills critical needs to enable organizations responding to disasters to share information across organizational lines and to track and effectively manage disaster efforts. These needs are substantial and are growing because of climate change and urban population growth, which is leaving increasing numbers of people vulnerable and susceptible to the effects of disasters.

Sahana fills a unique niche in the toolkit of emergency and disaster response agencies because it facilitates critical information sharing and coordination of efforts across all types of organizations and individuals involved, and is readily flexible to the needs that arise from any particular disaster. The software has its roots in the open source business community and has always been strongly supported by that industry. In this article, we describe an industry opportunity in HFOSS and provide evidence that Sahana software can be successfully commercialized. These factors make it a ripe time for open source businesses to engage cooperatively with the Sahana Software Foundation to support the growth of a service industry around Sahana software and HFOSS.

Introduction

Sudden-onset natural disasters have a devastating political, economic, social, and human impact on a society. The 2004 Indian Ocean Earthquake and Tsunami left almost a quarter of a million dead, 500,000 injured, 5 million homeless, 1 million jobless, and caused at least $7.5 billion in damages across several countries. Hurricane Katrina submerged New Orleans the next summer, and dispersed thousands of the evacuees across dozens of states; it was the costliest storm on record in the United States, causing over $81 billion in damage. Approximately 75,000 persons were killed by the 2005 Kashmir earthquake, mostly in Pakistan, with 3 million left homeless. The 2010 earthquake that decimated Haiti left 230,000 dead, 300,000 injured, 1 million homeless, and almost 300,000 homes and businesses destroyed. Monsoon rains left almost one-fifth of Pakistan under water in 2010, affecting 20 million persons left homeless and destroying infrastructure and businesses. As the trends of population growth, urbanization, and global climate change converge, the scale and impact of disasters will only continue to grow.

Disasters of this scale require a corresponding massive relief effort to save lives and help disaster victims recover. For agencies responsible for helping disaster victims survive and recover, some of the biggest challenges involve those of effective coordination and management of the multitude of requests for assistance and information. Lifesaving decisions need to be made quickly: where to dispatch search and rescue teams, where to set up shelters, how to effectively distribute aid, how to manage donations, how to trace missing persons, and how to assure security of persons and property in the areas affected by the disaster. The best decisions are the most informed ones. Without access to information needed to make good decisions, relief may not get to those who need it most, aid dollars may be wasted, and lives may be lost.

Ten days after the 2010 Haiti earthquake, those coordinating the relief effort for the US Government did not know the location of all of the operating hospitals and medical facilities in Haiti, what services they offered, their status, and how many beds were available. This information was needed so relief workers would know where they could send the injured or ill, and for those hospitals already overflowing with patients, where they could transfer non-critical patients so they could continue to receive needed medical care. It is hard to imagine how this is possible. The Sahana Software Foundation stepped in to meet this need and through hard work and research by volunteers, we were able to provide through our public portal the first comprehensive data set of all functioning hospitals in Haiti, including their coordinates such that they could be located on a map, within 24 hours, and provided multiple data feeds so everyone could use this information within their own systems. This information was used by the Pan American Health Organization (PAHO), responsible for coordinating the international health response to the Haiti earthquake, the medical planners for US Southern Command, and dozens of other relief organizations.

The Sahana Software Solution

The experience of the 2010 Haiti earthquake illustrates the mission of the Sahana Software Foundation: to help alleviate human suffering by giving emergency managers, disaster response professionals, and communities access to the information that they need to better prepare for, and respond to, disasters through the development and promotion of free and open source software and open standards. In the aftermath of a massive natural disaster, some degree of chaos is inevitable, but this where Sahana’s disaster management systems can play a critical role. It can provide a scalable systematic approach by managing large amounts of information, being able to efficiently distribute that information and make it accessible to those who need it on demand, providing a means to automatically collate, aggregate, and calculate based on all available data, and provide reports that are updated live and in real-time.

Sahana is one of several HFOSS projects that are now revolutionizing information management for international disaster response operations. Among others, this landscape includes the following projects and organizations, many of which provide visualization and maps from static sources of disaster information:

• Ushahidi
  
• InSTEDD
  
• OpenStreetMap
  
• FrontlineSMS
  
• Google Crisis Response
  
• Crisis Mappers Network
  
• OpenMRS
  

Compared to other projects, Sahana software is unique in two ways. First, Sahana software is the only system designed to provide management tools that allow responding agencies to work with the large amounts of data available to them; it allows them to assign individuals to tasks and facilitates management of people, places, and things that are important in disaster relief. Second, it provides a powerful platform for interoperability such that other tools and users can use Sahana software to share information, for example by pulling data from Ushahidi, OpenStreetMap, Google Person Finder, and other sources, aggregate it in one place, and redistribute it via open data standards such as EDXL, KML, or RSS. This allows Sahana to bridge information gaps across the diverse organizations and individuals responding to an emergency, each seeking to manage maximize the impact of their efforts.

Sahana software ensures responders and recipients have current relevant information to respond effectively. Sahana users can:

• register individuals or families at a shelter, and link requests for aid (e.g., food, water, and blankets) or services (e.g., medical assistance, counseling, and family reunification) at individual or location levels

• manage inventories and update the status and capacity of operating hospitals and medical facilities. Missing persons registries can be referenced against those registered at shelters.

• pull requests for assistance and information from crowdsourcing solutions such as Ushahidi, FrontlineSMS, and Twitter. Requests can be managed by assigning tickets to each one, which are used to track responsibility, validity, priority, and status.

By facilitating sharing of relevant information across the diverse set of individuals and organizations that need to communicate with each other, Sahana software fundamentally transforms the chaos of information from a hindrance to an asset, enabling greater self-sufficiency during responses on behalf of communities.

The Sahana Software Foundation

The Sahana Software Foundation is the first and most mature HFOSS project and inspired the movement itself. Sahana was originally developed by members of the Sri Lankan IT community who wanted to find a way to apply their talents towards helping their country recover in the immediate aftermath of the 2004 Indian Ocean earthquake and tsunami.

Since 2004, Sahana has grown into a global free and open source software project supported by hundreds of volunteer contributors from dozens of countries and it has supported national and local authorities and relief agencies in their response to numerous large-scale, sudden-onset disasters. The effectiveness of Sahana software is demonstrated by a track record of successful deployments:

• In the 2004 Asian Tsunami, the Sri Lanka Center for National Operations used Sahana to identify the location of all relief camps and the demographic breakdown of over 26,000 victims (by age, gender, and illnesses) to help them target relief.

  
  

• Since 2007, the City of New York’s Office of Emergency Management has used Sahana for managing its all-hazards sheltering plan, which involves over 500 shelters capable of housing over 800,000 persons and staffed by over 60,000 city agency employees and volunteers.

  
  

• After the Chendu-Sitzuan Earthquake of 2008 in China, Sahana was used by the Chendu police to track over 40,000 families. Following the earthquake, 42 separated family members were reunited using Sahana.

  
  

• After the Haiti earthquake of 2010, Sahana was used as a registry of nearly 700 organizations involved in the response. The software tracked almost 10,000 requests for assistance and information collected by Project 4636 provided the most accurate and complete registry of the 162 operating hospitals and medical facilities, along with bed availability and status, and aggregated 41 data layers from various sources onto one situation map. More than 230 registered users entered data into the system, which was accessed by over 8,600 visitors. The site was used daily by dozens of responding agencies in the first month of the disaster response.

  
  

• The US National Library of Medicine has developed a set of medical-related tools with Sahana as part of the Bethesda Hospitals Emergency Preparedness Partnership.

  
  

• Internationally, Sahana has been adopted by national and local governments, including Sri Lanka, the Philippines, Bangladesh, India, Indonesia, Pakistan, Peru, Taiwan, and China.

  
  

The Sahana Software Foundation is recognized as a leader in humanitarian free and open source disaster information management software. It was established in 2009 as a non-profit organization to serve the needs and requirements of a diverse group of customers: government at the national, provincial or state, and local levels, United Nations (UN) agencies, international and local charitable organizations (non-governmental organizations), communities and disaster victims, and technology companies and software developers.

When considering the future of HFOSS and the Sahana Software Foundation’s objectives, we recognize the following trends:

• The world’s urban population will increase to 6.4 billion by 2050, when 70% of the world’s population of 9.2 billion will live in urban areas, according to the UN.

  
  

• Urban infrastructure spending is expected to approach $350 trillion over the next 30 years, excluding water and sanitation, based on calculations by Booz & Company.

  
  

• Spending on disasters will triple to $185 billion per year by 2100, according to estimates in a UN and World Bank report.

  
  

Together, these trends represent an incredible opportunity. The Sahana Software Foundation seeks to leverage its humanitarian free and open source software, its own resources and expertise in disaster management, and an active community of volunteers, to significantly enhance the world’s ability to mitigate and respond to disasters that threaten increasingly large numbers of vulnerable persons concentrated in densely populated urban environments. But we can not do this alone.

The Path Ahead: A Cooperative Business Model

While Sahana is primarily a global,volunteer-driven, open source project, it also has deep roots in the open source business community and has always been supported by that industry. Google and IBM, along with Accenture, have made investments and contributions to Sahana’s development. Also, Virtusa provided testing help for the Haiti earthquake and Pakistan flood responses and provided a significant contribution at the start of the project in Sri Lanka. Further, Sahana has spurred the formation of two commercial companies that were started by experienced Sahana developers to provide Sahana-based services: Respere and AidIQ. Sahana has also been a part of several commercial service offerings and successful bid responses.

Although the software will always remain free, part of creating a sustainable HFOSS project includes the successful and viable commercialization of a product with a service industry surrounding it. However, despite the success and recognition that the Sahana Software Foundation has already received, there has admittedly not been enough user adoption to generate enough demand for widespread industry growth; it is simply not being deployed fast enough. One of the factors holding Sahana back is the lack of a qualified service sector to support it. The Sahana Software Foundation hope to grow the service industry that can provide technical support, while providing:

• core codebase maintenance and support functions

  
  

• professional user documentation and training

  
  

• help in guiding the roadmap for new features based on stakeholder and user input

  
  

• promotional assistance to encourage the adoption of Sahana and the principles of collaborative open source approaches to disaster information management

  
  

Despite these challenges, there are many indications that Sahana is on the cusp of widespread global adoption. Sahana has been acknowledged by important stakeholders in emergency and disaster management, a message that will reach those who have the financial resources to invest in disaster risk reduction initiatives. An article in the Bulletin of the International Association of Emergency Managers recently noted the uniqueness of Sahana’s capabilities that include many useful features that are not included in commercial Critical Incident Management Systems (CIMS) that emergency management agencies typically use (IAEM Bulletin, Vol 27, No. 5, May 2010). Also, a US State Department White Paper on the Haiti Earthquake response noted, partly in reference to Sahana:

“Within hours after the report of the Haiti earthquake, a new community of virtually connected volunteers affiliated with ICT consulting companies, private corporations, open source software proponents, academic/research institutions, NGOs, and even the Haitian diaspora community began applying new ICT applications to the earthquake response… This new community needs to be recognized as a new player in the humanitarian information environment…”

The business and open source sector has similarly acknowledged the maturity of Sahana through the following awards:

• 2010 Cool Vendor in Risk Management and Compliance, awarded by Gartner Inc.

  
  

• 2010 Best Practices Award, awarded by Private & Public Businesses, Inc. for our collaborative response to the Haiti earthquake disaster

  
  

• 2006 Free Software Foundation Award for Social Benefit, an award that was itself inspired by Sahana

  
  

Having reached this level of recognition, the challenge is to deliver the support that is now expected. Capacity is needed to respond to new requests to have Sahana adopted at national and local levels as part of disaster risk reduction programs. Some of this demand will be fulfilled by the Sahana Software Foundation, but further capacity is needed. We hope that a robust service sector develops to support the efficient and cost-effective deployment of Sahana globally with customers who need it most. These customers are primarily in the public sector, but international organizations also have substantial needs for support services. Industry can provide critical personnel resources to market, deploy, and support HFOSS projects like Sahana. Through marketing, the private sector can help convince organizations of the value of this solution. Through the deployment of Sahana, private companies can enter commercial contracts to provide hosting, along with customization and enhancements on request. Finally, through their support of Sahana, private companies can enter maintenance contracts with a service level agreement that guarantees a professional level of technical and customer support in exchange for a fee. In these ways, industry can help create enough demand to make Sahana sustainable and commercially viable.

One of the means by which an industry around Sahana can grow cooperatively is through a certification and sponsorship program designed to professionalize the industry. This would provide a means for supporting a strong Sahana Software Foundation as a means to ensure the development of best practices around Sahana deployments, ensure that the codebase is well maintained, and provide a guarantee to Sahana’s customers that support is being provided with quality control and competence in the provision of emergency and disaster management services.

The needs of disaster victims are urgent and widespread and organizations providing disaster relief have a corresponding need for Sahana’s information-management tools. We hope to inspire the open source business community to join with the Sahana Software Foundation in a cooperative model to grow the capacity to address these widespread and growing needs so that there is an industry-level capability to do good. Doing good by helping people is fundamentally what HFOSS is about and this also fits in with corporate social responsibility programs.

In step with Sahana’s own social responsibilities, any engagement with the foundation must be done with the highest of ethical standards. The Sahana Software Foundation’s Chief Technical Officer, Chamindra de Silva, has been leading an effort to develop an HFOSS code of conduct designed to address issues of competitiveness and collaboration in the name of humanitarianism. Many other HFOSS projects have already agreed in principle to sign on to such a statement and we would expect the same of any company who seeks a partnership with the Sahana Software Foundation or seeks to use Sahana.

Conclusion

The open source industry has an interest in supporting a strong Sahana Software Foundation, whose experience, mission, and non-profit status allows it to serve as a trusted agent between a service industry and its customers. The number and impact of disasters is growing and will continue to grow, along with tremendous spending growth on disaster response and mitigation. Because of HFOSS projects like Sahana, there is an opportunity for open source businesses to contribute and to benefit from increased spending on initiatives that mitigate and reduce the risk of disaster. The challenge will be to develop a cooperative and collaborative industry that meets the needs of its growing customer base, while retaining a commitment to the principles around which HFOSS projects are organized. Most importantly, the solution should be born of a desire to help alleviate human suffering.