On May 2, Government Technology published a great article called Apps Contest Winners Need Better Government Data to Sustain Innovative Services. It was a very well-written article about the challenges of sustaining the make-data-available-for- private-sector-innovators model.
By now, many of us are aware of events like Apps for Democracy, an app developing contest in Washington DC that yielded $2.3 million worth of applications at a cost of $50,000. These events have been highly successful because:
Government wins by
- getting applications developed for free
- building positive relationship with development community
- gaining a PR win from the public for being innovative and transparent
Development Community wins by
- Competing for cash prizes
- getting their name out and showcasing their talents on a public stage
Public wins by
- obtaining many applications to help them engage with government
- gaining transparency into government
There’s no doubt valuable apps emerge from these contests, but what happens after the novelty of the contest wears off? This can’t be just a once and for all event, then everyone goes home and things revert back to the way they were. How do you sustain the viability of this relationship so value is still being created in the long term?
Before answering this question, there are several challenges that must be addressed:
1. Who is going to maintain these apps?
The GovTech article mentions that most of these apps don’t get updated after its release because government lacks the resources and developers lack the motivation. It doesn’t make sense for developers to keep playing if they can’t generate revenue and it’s very difficult to monetize off of apps given to the public.
2. How do we get through the “valley of disinterest”?
Former DC CTO Bryan Sivak believes open data has lost some of its luster from a few years ago, dropping from the “peak of inflated expectation” to the “valley of disinterest.” Now that the hype has subsided, are we still interested in this type of model?
3. Not enough data is being offered by government
Daniel Odio, CEO of a mobile Web consulting firm, envisions “dazzling possibilities for private-sector mash-ups… but doesn’t think enough open data is offered.” Ideally, data should be delivered to the public before they have to ask for it, but we’re still a long ways off from realizing that potential.
4. There’s a lot of “dirty data” out there
The article also cites a common frustration among developers trying to work with sloppy, inaccurate or unreliable data. A lot of times agencies are just pushing data out for data’s sake, and if it’s not carefully inspected for accuracy or if it’s not in a usable format, then developers can’t do much with it.
I’d like to add a few more challenges to this list
5. Cultural shift required
Technologically speaking, making data available is easy. There are plenty of tools available to help government push out data in formats that can be easily consumed by the public. The bigger challenge tends to be cultural. It requires a shift in the way government thinks about their data, and there are still a lot of agencies who aren’t ready for this shift.
6. People want quality, not quantity
It makes great PR to brag about hundreds of apps being developed with open data, but how many of them would actually be used by the public? Do we as consumers really want 52 apps telling us where the nearby bus stops are? Sure, it’s not up to government to decide what’s valuable or not, but if you get 300 apps from a contest and only a handful get used, the value of your efforts might be a bit overstated.
7. Can the networks support it?
One of the most frustrating things for a mobile user is to open up an app and wait and wait and wait. Carriers are promoting 4G, but many people don’t have devices that make it worth their time and battery life. For now, 3G is still predominant and if you have a data intensive app, waiting too long is one of the fastest ways to lose interested users.
8. Are we creating the next wave of the digital divide?
Smartphones are increasing exponentially, but the overall market share of iOS, Android and Blackberry devices is still a minority. Obviously, certain demographics are adopting smartphones at a much faster rate than others, so if government is offering innovative ways to access data that can only be experienced by certain demographics, are we essentially creating a new digital divide?
In the long term, I think the general idea of open data with private partnerships is sustainable, but maybe not in the form of contests. It would have to be a concerted effort with long term incentives for all stakeholders involved. The challenges will make it more difficult, but if true value is being created, the benefits can overcome these challenges.
Excellent thoughts that I’m going to share with people at EPA considering apps contests.
Thanks!
Let’s compile some answers to the weaknesses with apps contests. Ideas:
Fund the prize and ongoing support through a private partnership;
Make the prize a 3-year maintenance contract. If procurement challenges are a problem, use a sub;
Outline clearly what your agency wants to have after the contest is done. Extras are gravy;
Develop in html 5 and Titanium to reduce device-specific apps …
Thanks Jeffrey! I can’t take credit for the first four. Those came from the Gov Tech article.
Adriel, you sound like you have a blog post forming… I like your thoughts; definitely on board with getting a contract with the top app developers.
All apps should be open source;
Apps should have a biz plan outline;
Apps should have an infrastructure/hosting plan outline
What is an RFP and competitive bid solicitation but a form of contest?
Interesting, yes contests are very much like RFPs… But if we go through the RFP route, are we precluding some vendors because there are always so many requirements that have nothing to do with the actual solution? Maybe this is a way of updating the procurement process as well?
I want to be able to thumb up individual comments 😉
A few more answers and examples to some of these issues from my perspective.
#1. Mainting the apps going forward – we have attempted to take these apps in house and support them going forward through our own development efforts or an actual contract with the original developer(s). Our iphone trip planner was originally created for another geographic area and we contacted them to create the same app for us. Unless developers are making money on an ongoing basis there are very few who are going to maintain an application going forward. The feeling is if we are going to make an investment in this type of mobile infrastructure it will have to be something we pay for and paying for maintenance of an application as opposed to the original development is much cheaper. This also gets to #6, if it is an “official” supported app from an agency then customers will focus on that; our original targets for downloads and usage have been blown out of the water even though there are many other apps that support the same function for the same geographic area. One simple reason is they know who to complain to if it does not work correctly.
#2 (Disinterest) – this is a tough one; we track how many people download/access our data sets and besides regular existing application updates it is very few. But I think this ties to #3 new data sets have to be offered to peak interest and ideas from developers. If a problem space using a specific data set has been solved numerous time there is very little incentive for developers to think of something new. New data sets though and linkage between existing dats sets allows for the creative juices to flow. #5 is also related to this we do have people internally who think “should this data be published to the general public for consumption and how should it be”. This type of thinking allows for the up front process of making data accessible to happen. And designing data/systems to be accessible during design, development is a much simplier task and leads to much cleaner data #4. Really I think of this as more of a “tipping point” sort of issue not everyone in your organization has to think this way just enough people to make it effective.
#8 (digital divide) – we have thought about this one, but our current thinking is that the barrier to entry is much much lower for mobile devices than it was for computers back in the 10-20 years ago. When we do surveys and such there is always a fairly high percentage of people who have (or are planning to have) these types of devices. Especially (and this is important for a lot of agencies) the next wave of consumers.
Thanks Richard! Which agency are you with, by the way? I think you, Adriel, myself and whoever else should do a follow up post on ways to keep app contests sustainable. Great stuff!
Capital District Transportation Authority (CDTA) in Albany, NY.
One other point that Adriel made, since HTML 5 is making headway I think it and javascript should be the standard for these types of applications. Originally the argument was always well if you build mobile applications they should be device specific so they can take advantage of the specific api and functionality provided by the phone. This is starting to become less of concern now with the use of HTML 5, with functionality like media playback, improvements to offline data storage (not officially html 5), geo location (again not officially html 5 but its own w3c spec), canvas, cleaner forms, even regular expressions. This functionality tied in which javascript makes for very robust applications that can and should work on all types of devices. Also using these technologies does not take very device/platform specific knowledge about different development environments, languages and apis.
So if an app contenst can be structured in such a way that puts requirements or emphasis around the use of certain tools, like html 5, then I think it will be a big win for developers, government and the public.
Was CDTA part of the Big Apps contest or did you guys do your own thing? Can you send me a link with what you did? I’d love to read more.