Community Resource Directory Platform In A Box

In any community, there are many ‘referral providers’ – using call centers, printed directories, web apps, etc – that each collect information about the health, human, and social services available to people in need. However, these tend to be managed duplicatively, in fragmented silos, isolated from and even competing with each other. As a result, there are more and more sources of less and less trustworthy information – so it's hard for people to find services, and for communities to understand how resources are being allocated on their behalf.

We propose to build the 'community resource directory platform in a box' – a toolkit with which governments and philanthropies can publish canonical open data about all services they fund. By helping institutions publish better data about service delivery, we can make it easier for communities to develop new ways to use information about the resources available to people in need.

It's hard to 'see' the safety net. Which agencies provide what services to whom? Where and how people can access these services? These details are always in flux. Nonprofit and government agencies are often under-resourced and overwhelmed, and it may not be a priority for them to promote their own information in order to attract more clients. 

There have been many attempts to build a 'centralized clearinghouse for community resources' – yet few, if any, have succeeded. Instead, in any community, one will find multiple organizations competing with each other to maintain the same information, while guarding it as their property.

As a result of this wasteful and ineffective state of affairs:

• People in need have difficulty discovering and accessing services that can help them live better lives. 

• Service providers struggle to connect clients with other services that can meet complex needs. 

• Decision-makers are unable to gauge the effectiveness of programs at improving community health.

• Innovators are stymied by lack of access to data for tools to be used by any of the above.

One system can't possibly meet everyone's needs, yet many systems could use the same data. So Open Referral is taking a new approach to this problem.

Any person in need, and any kind of institution that is in a position to help them – these are the ultimate beneficiaries of our work. The adoption of open standards and data infrastructure will transform the way that all health, human, and social services are discovered, delivered, and even evaluated.

Yet people in need wouldn't use a data standard or data infrastructure, and neither would the service providers who help them. The primary users of Open Referral's products are the people who build the tools that are used by these organizations, service providers, and help seekers. Which is to say, we help database administrators, and technology developers, and information maintainers – we make it easier for them to develop tools, aggregate and share information, and integrate systems.

We do this by forming and facilitating pilot projects, in which local stakeholders test new partnership models and new technological solutions, through iterative, participatory research that prioritizes the interests of their users in the process.

In other words, we help the people who help the people who help people find help.

Open Referral has already developed open protocols for publishing data about health, human, and social services -- and the Alliance of Information and Referral Systems endorsed these as industry standards.

Members of our network have used these standards to develop open source applications – such as mobile-friendly websites and apps that help people find services -- which can be freely redeployed and customized, rather than having to build from scratch.

However, these steps still don't help answer the big question: who should be responsible for maintaining this information? Our next step is to make it easier for institutions to aggregate and publish information about their own grantees and contractors – as standardized open data.

Through Solve, we will upgrade a functioning prototype into a redistributable product. Specifically, we will develop a toolkit that will make it easy to produce adaptable forms, which collect information from providers and output it into a canonical "machine-readable" registry of services. 

The toolkit will enable any funding institution to develop an interoperable 'directory information platform' that serves reliable data about its funded services to an ecosystem of third-party tools and applications. 

The real users for this toolkit will be government agencies and philanthropic organizations that need a better way to keep track of the services  that they themselves are funding through grants and/or contracts. This toolkit will enable any government or philanthropic funder to quickly deploy an adaptable, standardized form, solicit compliance from its grantees and contractors, monitor participation, and publish the resulting information. 

The toolkit will include both technical components and a 'How To' guide that advises on the political process of engagement, change management, and evaluation – so that this change is effected not just by dropping a new technology solution from above, but rather through a social process of dialogue and deliberation.

We know that many governments and foundations want this functionality, as we've already helped several such institutions develop it. (For example, the NYC Mayor's Office on Opportunity is now publishing data on every one of their contractors; we've also helped the Florida Bar Foundation collect directory information about every legal aid service in the state.) This project will make it much easier for such institutions to solve this problem themselves – and will enable advocates in the community to promote this shift towards open data about the resources that are being allocated in their name.

Our solution is innovative because it creates conditions in which it is easier for others to innovate. 

Before Open Referral, any new effort to connect people to information about services would have to start from scratch – building a database, collecting (and maintaining) all of the information, and developing the interface to use it. All together, these challenges have stymied innovation for decades. 

Our approach provides a standard protocol in order to a) make it easier to deploy a resource directory without having to design a database from scratch; b) make it easier to repurpose the contents of an already-existing resource database, so that the information can be put to new uses; c) make it easier to redeploy already-existing technology to work in new contexts without requiring custom integrations; d) help institutions prevent technology vendors from "locking in" a community's data.

By making it technologically easier to deploy new technology that uses resource information – and by organizing the political conditions in which such information is published as an open data resource – we can lower the barrier to new innovations that might come from anyone in a community, not just the elite organizations that happen to have access to capital and data. 

At the root of Open Referral is a conviction that, when it comes to making change, the value of technology is secondary to vision and trusting relationships. So we first formed a community of people who were drawn to our vision – of a future in which reliable information about resources is findable through any number of channels, regardless of which technology someone happens to have available. 

Through dialogue in that community, we developed a common vocabulary that could be used to translate information about services from one context to another – an 'interlingua' for machines, so that data could flow across technological and organizational boundaries.

We developed this common vocabulary in a Github repository, which publishes to a user-friendly website through which readers can submit comments and suggestions via the Hypothes.is annotation feature.

Over time, this common vocabulary evolved to include a reference table, a logical model, formatting instructions for publishing resource data in CSV files, and tools for data transformation and validation. We've also developed a set of Application Programming Interface protocols in the OpenAPI scripting language, which is a kind of machine-readable blueprint for APIs – enabling developers to either rapidly deploy a standardized API, and/or translate their existing one into a standardized format. We also deployed an open source live documentation site that third parties can redeploy and adapt to serve as their own developer portal. 

Upon these components, we will build a standardized form-building toolkit in our next chapter.

We've already demonstrated the value of open standards for resource directory information in a range of contexts. For example, Open Referral’s standardized format has been used to develop and redeploy open source websites, such as Code for America's Ohana, Zendesk's Link, and Boston Children's Hospital's HelpSteps. Organizations building or redeploying these tools estimate that they have saved anywhere from $25,000 to $250,000 or more each, across a dozen or more projects. 

We've also demonstrated that this approach to standardized data collection and publication can work in the specific model at hand: Open Referral worked with the Florida legal aid community, in partnership with the Florida Bar Foundation, to develop an open register of legal services, maintained through deployment of a standardized form to collect this information directly from grantees. In this model, a funder can mandate submission of canonical information, via an easy-to-use, interoperable form, which publishes to a database that can be accessed simultaneously by many third-party clients. This program worked, as described in our final report, and the resulting data is now being used in at least three projects that take different approaches to connect people to legal services in florida. 

Now we must make it easier to replicate and scale the solutions that are enabled by interoperability. Through this project, we will evolve this prototype into a product for the market of funders and government agencies.

In the coming year, we aim to demonstrate the viability of new business models for information and referral providers that can sustainably provide resource data as an open data resources.

We also aim to replicate our successful Service Registry model, in which a funder requires their grantees (or contractors) to publish information about their own services in an open registry. This model was prototyped in Florida, with the Florida Bar Foundation, in which every grantee received a standardized form, through which to update information about their own services; the submitted information is subsequently vetted, aggregated and made available via standardized API. We now are seeking opportunities to repeat this with other funders and governments, while demonstrating the many kinds of value that it yields to service providers, help-seekers, and funders themselves.

Finally, in the next year we aim to demonstrate the viability of resource data federation, in which multiple resource directory maintainers can reduce their costs and increase the quality of their data by sharing information across a cooperative network. We have at least one pilot project testing this approach (Benetech's Service Net initiative in the San Francisco Bay Area) and will aim to launch a 2nd project and open source the resulting code and findings. 

Within five years, we aim to replicate and integrate these models through industry-wide best practices and globally scalable infrastructure.

In the coming year, we aim to demonstrate the viability of new business models for information and referral providers that can sustainably provide resource data as an open data resources.

We also aim to replicate our successful Service Registry model, in which a funder requires their grantees (or contractors) to publish information about their own services in an open registry. This model was prototyped in Florida, with the Florida Bar Foundation, in which every grantee received a standardized form, through which to update information about their own services; the submitted information is subsequently vetted, aggregated and made available via standardized API. We now are seeking opportunities to repeat this with other funders and governments, while demonstrating the many kinds of value that it yields to service providers, help-seekers, and funders themselves.

Finally, in the next year we aim to demonstrate the viability of resource data federation, in which multiple resource directory maintainers can reduce their costs and increase the quality of their data by sharing information across a cooperative network. We have at least one pilot project testing this approach (Benetech's Service Net initiative in the San Francisco Bay Area) and will aim to launch a 2nd project and open source the resulting code and findings. 

Within five years, we aim to replicate and integrate these models through industry-wide best practices and globally scalable infrastructure.

People often assume that the development of standards is itself a futile act – as summarized by XKCD's cartoon on standards. However, this assumes that the development of a new standard is a merely technical venture, rather than the active process of community organizing, field-building, and institutional alignment that Open Referral facilitates. We help communities solve the collective action problem of "too many competing standards," by cultivating cooperation across previously-competitive systems.

Nevertheless, some proprietary vendors are simply hostile to this concept. It's easier for them to lock their customers in and extract as much value as possible than it is to accommodate the emergent needs of an open ecosystem. 

This is also true of startups. One can easily make a technology sound plausible. It's harder to ensure that technology is accountable to a broad ecosystem of stakeholders, and that it aligns with a complex institutional ecosystem. 

Finally, even among organizations that do want to cooperate, there tends to be a lack of capacity in their fields that inhibits the ability to adopt new methods even if those methods save time and money. 

There are some technical challenges: we need to develop protocols for feedback loops that enable resource directory maintainers to receive and act upon notification that information needs to be corrected. We also need to address the challenges of unique identification of services. These problems are less challenging than the issues above. That's why small amounts of resources from contests like this can go a long way.

We overcome resistance to cooperation by building relationships among those who do wish to cooperate, helping them succeed through cooperation, and recruiting more cooperators. 

This takes a long time, but it works – our record demonstrates that. 

Our path has now been greatly facilitated by formal endorsement from key institutions such as the Alliance of Information and Referral Systems. Open Referral is now the industry standard for resource data exchange – which means vendors are expected to adopt it as a default method for interoperability. If any vendor won't cooperate, their prospective users have many other options that will. 

Increasingly, governments and philanthropies are using Open Referral to publish information about the services that they contract and fund. This makes it easier for organizations to be leveraging this standardized data than it is for them to continue maintaining information manually in isolation. 

Through these trends, and the proliferation of open source tools that use Open Referral to accomplish basic functions that are common across sectors, we're steadily making it easier for organizations to cooperate than not to cooperate. 

As the value propositions become clearer, it becomes easier to help our partners build their capacity to experiment and innovate, which in turn catalyzes the development of more software and best practices that can be shared across our network. 

And finally, small amounts of funding such as that which is offered by this prize can help us evolve prototypes into sustainable products to accomplish key objectives in our strategy of ecosystem change.

The Open Referral Initiative is fiscally sponsored by Aspiration, a 501c3 that provides facilitation and capacity building support for open source nonprofit technology projects.

We also have an LLC, Open Referral Consulting Services, that does contracting work on demand.

We have one full-time lead and a network of dozens or so people who come in and out of projects as necessary. It would be difficult to calculate the number of contractors as we work through partnerships in which partner organizations manage their own teams.

Open Referral emerged out of years of experience with and observation of failed efforts to find 'quick fixes' to this problem. Across five years working in the field with health, human, and social service providers, I learned about the complexity of the challenge through interpersonal relationships with stakeholders on many sides of the problem.

My background in community organizing and cooperative development enabled me to bring facilitation and deliberation skills to bear on a situation that resists unilateral, competitive attempts to establish simple solutions. (I have found that many people from technical and/or business backgrounds falsely assume that strategies involving open data standards are not feasible – because they have not experienced the process of community-building, and struggle to imagine the pragmatic paths that can lead to collective action.)   

Along the way, I have developed a network of people with a wide variety of skills, which can be brought to bear in specific combinations as needed. Notably, many of these people came to Open Referral having had experiences with efforts to solve this problem through apps, 'centralized' systems, and associated startup business models – which led them to conclude that conventionally-entrepreneurial approaches are poorly suited to address this wicked problem. Together, we work to find alternative solutions, usually by helping stakeholders find solutions that are appropriate and accountable to them. 

Open Referral works through partnerships:

• Aspiration is our fiscal sponsor and often works with us to design and co-facilitate events and partnership processes.
• Open Data Services Cooperative has served as our technical leadership for standards development.
• The API Evangelist leads the development of our API protocols.
• Loup Design & Innovation helps with stakeholder engagement and participatory research.
• Benetech is leading one of our key pilot projects, the Service Net Initiative in the San Francisco Bay Area.
• Sarapis supports pilot project implementation and has also led work applying Open Referral in disaster response arenas.
• Digital Public advises us on legal frameworks and governance models for collectively owned data assets. 

This is in addition to dozens of partnerships through local projects and issue-specific initiatives. A sample of these partnerships can be reviewed on our blog. 

The Open Referral Initiative is an open network that includes hundreds of service providers, information-and-referral specialists, policy officials, funders, researchers and technologists. 

For our first several years, we acquired small grants and in-kind contributions in order to patch together shared capacity to develop specifications for data exchange, associated tools, and activities that facilitated collaboration around the use of these specs and tools. 

To the extent that this collaboration was strategically relevant to our network members' own objectives, some of them contributed time, talent, and resources to support this process. Over time, these collaborations became more formal, and more integral to real-world implementations – and through these collaborations, funding followed in the form of contracts and consulting relationships.

Now, Open Referral has 501c3 fiscal sponsorship to enable us to receive additional grants that facilitate development of our core resources – and we also have a profitable consultancy that supports members of our network through the process of strategic design and implementation, fundraising, facilitation, evaluation, etc. 

In any given community, significant money is allocated to unsustainable, fragmented, siloed resource directory projects. (In the U.S. we estimate this may be hundreds of thousands and in some cases millions of dollars per community per year; at least one study in the UK has reached a similar conclusion.) Yet demand for resource information is nowhere near being met. This is the core of our market, which also extends to any institution that has an interest in the health of its community — from government agencies and universities, to anti-poverty service providers and health insurance companies, to philanthropies that invest in education and the safety net.  

We develop open source knowledge and code, and generate sustaining revenue by providing expertise for deployment, support, and operations.  Our path now involves replicating those successes, and evolving such prototypes into scalable products. From tools such as the community-platform-in-a-box that we're proposing to develop this Solve opportunity, to globally federated infrastructure that we believe we can achieve within the next five years, our work can enhance the value of existing investments and unleash new forms of value. 

By focusing on building the capacity of partners rather than competing with them, we're able to leverage potentially-transformative systemic change while remaining small-scale ourselves. That said, we're aiming to reshape the world – in a way that may transform Open Referral itself into something entirely along the way. 

We've been able to get funding to prototype specific tools that are critical to our vision; we prototype these tools with specific partners in particular contexts in which we can help them solve a particular problem that they have – and we choose to solve for actionable problems that might also be common to other stakeholders across the field. It's a challenge to build the capacity to evolve these prototypes into products. 

For example, in this case we are proposing to rebuild a successful prototype (a standardized form, commissioned by a funder to collect directory information from their grantees) into a re-deployable product, and package it with materials that guide its usage. This funding is just (barely) enough for us to go from an 'alpha' prototype to a 'beta' tool, which we can then – if successful – evolve and sustain from implementation contracts – in this case, by providing this form-building toolkit to governments and funding institutions that wish to collect and publish more useful information about the services provided by their own contractors and grantees. We know there's a market for this out there, but we need a boost to demonstrate the replicability of an already-proven prototype!

Beyond just the funding, it seems like the Solve network is an ideal context in which to test these ideas, and promote our findings – given your network of emerging service providers, funding institutions, and specialists who work among them.

In general, we work with referral providers (who maintain resource directories) and community based organizations (who need resource information to provide better services) as well as the governments, funding institutions, researchers, and technology providers with whom they work. 

In this particular case, we are seeking to evolve a specific tool that is relevant to funders that invest in service delivery, and/or network institutions that manage the allocation of resources (and/or licensing, certification, etc). 

On a local level, the Florida Bar Foundation was our partner for developing this prototype: they now have a standardized form through which they can collect information about their grantees – and a registry through which they publish this information as open data. On a global level, institutions like the UN, the World Bank, and the Red Cross come to mind as similar organizations that have known challenges collecting and disseminating information about the resources that they allocate. In between, there are federal, state, local governments and any number of philanthropies that can be better sources of information about the services that they fund – and who regularly pay for ineffective, unsustainable websites that present this information with unsatisfying results. We can help them address this need in a new, and much more effective way.

Through this project, Open Referral will build a toolkit specifically designed to help governments and funders to collect and publish standardized open data about the services that they fund. 

This information currently must be manually maintained by third parties. Many of those parties are community-based organizations who need this information to provide care to people with complex needs; some of those parties, on the other hand, are rent-seeking intermediaries who treat the information as their property to be sold for a fee. Our project changes this dynamic: by making it easy for funding institutions to publish standardized open data about the services that they fund, we make it much easier for community organizations to focus on what they do best (helping people in their local context) and for information intermediaries to innovate and add value without enclosing public information.

Open Referral recently had a success story regarding refugees and service delivery: in British Columbia, PeaceGeeks developed the "Arrival Advisor" app specifically to help refugees and immigrants find resettlement services. PeaceGeeks used Open Referral to develop the data exchange that enabled them to repurpose 2-1-1 British Columbia's already-maintained resource directory. This enabled Arrival Advisor to quickly launch and succeed in BC, and it can be redeployed in any place where resource data is made available in this format. 

Through our current project, Open Referral will build a toolkit specifically designed to help governments and funders to collect and publish standardized open data about the services that they fund.

This information currently must be manually maintained by third parties. Many of those parties are community-based organizations who need this information to provide care to people with complex needs; some of those parties, on the other hand, are rent-seeking intermediaries who treat the information as their property to be sold for a fee. Our project changes this dynamic: by making it easy for funding institutions to publish standardized open data about the services that they fund, we make it much easier for community organizations to focus on what they do best (helping people in their local context) and for information intermediaries to innovate and add value without enclosing public information. 

Through this particular prize, we could specifically target institutions that fund services for refugees, in order to help them more effectively collect and promote information about those services.

Through this project, Open Referral will build a toolkit specifically designed to help governments and funders to collect and publish standardized open data about the services that they fund. 

This information currently must be manually maintained by third parties. Many of those parties are community-based organizations who need this information to provide care to people with complex needs; some of those parties, on the other hand, are rent-seeking intermediaries who treat the information as their property to be sold for a fee. Our project changes this dynamic: by making it easy for funding institutions to publish standardized open data about the services that they fund, we make it much easier for community organizations to focus on what they do best (helping people in their local context) and for information intermediaries to innovate and add value without enclosing public information.