“Open 311” has been used to refer to a few different things. To clarify, the intent of this website is to create a standard specification to turn 311 services into an open platform. I call this Open311, but really, it’s is a standard that does not yet exist. There won’t be any one specific app or API implementation that can be referred to as “The Open 311,” but as an open platform there will be a multitude of Open311-compliant apps that work with a multitude of Open311 API implementations. This is the architecture of an open platform.
An Open Platform
Much of what distinguishes a platform from an open platform is covered in a recent post that describes the web and democracy as examples of open platforms, but the fundamental point is that an open platform can’t exist until an open standard does. In the context of 311, an open platform is one that makes service requests publicly available and enables participation through a read/write interface. Additionally, an open standard gives this platform complete accessibility, collective innovation, and a distributed model with scalability and stability.
The Many Faces of Open311
Some of the ambiguity regarding “open 311” was brought out at TechCrunch 50 during the launch of a new 311 service called CitySourced. After CitySourced founder Jason Kiesel presented the product a panel asked questions and offered feedback. Tim O’Reilly was one of the panelists and said:
I’m a huge fan of this type of application and ever since Thomas Steinberg built FixmyStreet.com in the UK it’s been pretty clear that this was the way to go. But I worry a little bit about the defensibility of the product. What’s your thinking about how you become a market leader? Is this something you just have to get out there and sell to a lot of people first? Because there’s so many people working in this area: SeeClickFix, a lot of app contests being run by cities are producing open 311 apps, there’s the open 311 API. What’s your competitive advantage and how are you going to go to market in such a way that you become a viable business?
Responding to Tim’s mention of open 311, Jason says:
Open 311: it’s great, I would say that the UI needs some work; overall, if a city adopts that platform, they’re locking themselves into their city alone. Our platform works nationwide, so if you download the app and you’re in the city of San Jose and you’re in LA and you report something in LA, it goes to LA. So that’s a big competitive advantage that we have.
Tim mentioned “the open 311 API”, but currently there are actually two drafts of APIs which call themselves “The Open311 API.” One is provided by a company, the SeeClickFix API, and one is provided by a city, the Washington D.C. 311 API. I’m not quite sure which UI Jason was referring to. In the case of D.C.’s 311 system, there are now a variety of different UIs. Much of the intent of opening 311 is to not be tied to any specific UI. As a standardized specification, Open311 would abstract the services of 311 systems into a consistent open platform that anyone can build a UI for.
Open311 is not meant to refer to a specific app or any one incarnation of 311 services. Instead Open311 intends to be a specification of an open platform for 311 services. This difference may seem subtle, but it’s an important one. It’s the difference between closed platforms like the iPhone and open platforms like Android or the web which are enabled by open standards: the Android operating system, HTML, and HTTP. The challenge we need to address is creating an open standard by looking at APIs like those offered by D.C. and SeeClickFix and coordinating the commonalities of different city infrastructure to distill core requirements for a universal 311 API. Once this core standard is defined, new user interfaces and custom workflows can be created by anyone and shared between cities to provide distributed innovation.
Open Platforms are Distributed
Jason Kiesel provided a good point, in order to guarantee that the platform actually connects to all cities and can provide a UI that works anywhere requires coordination with each city. Tom Steinberg put this eloquently when he asked us to build FixMyStreet in this country:
Your citizens deserve services that have that kind of usability of a single national service; that hides the splinter of federalism. And it’s challenging because of the federalism; which is precisely why you should do it. Because if you can overcome that you will have done something amazing and you will have a really simple example that will give legitimacy to investment in loads of other things.
It could be beneficial to have government coordinate this routing on a national level, but we don’t necessarily need higher level government to provide a single unifying force. Tim O’Reilly alluded to this in mentioning the government’s role in providing root DNS servers within the context of 311 services:
Can gov get in the business of providing some minimal services (a la the root servers of the DNS, originally govt chartered) in such a way that the public takes the ball and does everything from there.
Like DNS and the web, a distributed 311 platform can be facilitated with not only multiple inputs and endpoints, but also multiple hubs that route information. This is the underlying architecture of the internet and it’s the model now making its way to web technologies using an open stack that distributes streams with real-time methods like PuSH feeds. It’s no accident that I mimicked the diagram of the PuSH model to use as a visualization for distributed 311 services in my Gov 2.0 talk.
By not depending on a central server, or even a central router, distributed systems are more stable and more scalable. Imagine if multiple major cities relied on one single centralized service to provide their 311 platform. What would happen if this service suffered catastrophic hardware failure or if it were run by a business that wasn’t financially sustainable? These are things that happen on the web. Ma.gnolia is a social bookmarking service that competes with Delicious, but recently their infrastructure suffered a catastrophic hardware failure and all of the users’ bookmarks were lost. Similarly, Tr.im is a URL-shortening service that competes with Bit.ly, but recently they announced that they would be shutting down the service (and breaking all of their links) at the end of the year because they are not profitable. Even Twitter demonstrates its lack of stability on a regular basis and they’ve yet to prove that they are financially sustainable. Fortunately, there are signs of new life at Ma.gnolia and Tr.im, but preventing these types of systemic failures is exactly why the internet is designed with a distributed architecture.
Additionally, the internet has proven that the distributed model helps foster new innovation and new business while providing a better user experience by guaranteeing interoperability. A distributed model does not preclude software as a service just as the internet did not prevent companies from leasing infrastructure and selling hosting services. This is a model that simultaneously promotes growth and prevents systemic problems, but it’s only possible with cooperation.
A Call to Action
This is a crucial opportunity to think globally and act locally. Please join the effort to establish an Open311 standard; speak up on our mailing list or edit the wiki. For this to work, we really need to hear from cities. Describe your city’s 311 infrastructure and how collective action can improve these city services. We have organized an Open311 DevCamp on October 24th in New York City with developers of 311 systems from both inside and outside city governments, so please come to that and help move this forward. As a proactive developer community we can help build better communities for all.