Hackathon, Challenge or Code for… Best practices for events on programming with open data

5 min reading
Development / 01 October 2014
Hackathon, Challenge or Code for… Best practices for events on programming with open data
Hackathon, Challenge or Code for… Best practices for events on programming with open data

BBVA API Market

If you are considering organizing a conference, contest or long-term initiative with programmers to encourage innovation and create applications with open data, this recent series of best practices may serve as inspiration. Take note.

In hackathon format. If the purpose is to create a community among programmers, conduct a quick experiment or, alternatively, you want to draw attention to a particular problem for a short period of time, holding a hackathon suitably meets these objectives. Just remember that its main limitation is that, with few exceptions, the initiatives that emerge often lack continuity.

An example of best practice that proposed to turn this limitation around from the outset is the Edinburgh Apps, the first civic competition organized by the local Edinburgh government (City of Edinburgh Council). This event has placed this municipality among the most innovative in Scotland for three main reasons. First, because in addition to programmers, designers, journalists, entrepreneurs and groups with a direct interest, it was decided to involve other municipal officers, civil servants involved in the management of information, digitization, administrative technology, open government and citizen service. The large representation of public managers helped to make Edinburgh become a benchmark and inspiration.

In addition, the organizers made ​​it clear from the beginning that they didn’t want it to be a one-off, isolated initiative. Municipal officers explained how they are actually working to expand the open data sources, and individuals in senior roles in the council made the most of this contest to publicize the official open data  strategy that Edinburgh is following. That is, they showed “everything behind it”.

Finally, and very importantly, several of the projects managed to achieve business continuity as start-ups. For example Neatbox, which aids urban transit for pedestrians with impaired vision. This is a small device installed at junctions that connects, firstly, with the signal of traffic lights that have a push button for pedestrians and, secondly, with the phone of any user. So when a blind person reaches the zebra crossing, their cell phone activates the green man without having to search to press the button. Or Airts, a project that combines local data on support services with private data of companies in need of logistics, warehousing and distribution to multiply the effectiveness of their services and workforce or better distribute resource allocation in a project.

In app challenge format. An app programmingchallenge  has the advantage that, with less time constraints and physical presence than a traditional hackathon, it is closer to the ideal framework of involving developers to create lasting solutions, finding more allies in the government and among the business network, and the possibility of finding better data sets to work on.

Here’s an example that is still current. The phone operator AT&T has a particular interest in the convergence between telephony and vehicle traffic and navigation. Its application AT&T DriveMode seeks to reduce distractions caused by incoming calls and messages. Being one of the member companies of the New York City Lab, AT&T has teamed up with a consortium of technology companies, public agencies and several universities to organize an app contest designed to reduce the almost 15,000 accidents between vehicles and pedestrians (including cyclists) that occurred in 2013 in New York.

Connected Intersections, which is the name given to the challenge, part of a well-documented hypothesis on the effect of the distractions caused by mobile devices while driving and how to apply those technologies to improve the safety of pedestrians. Participants must start from the data from this study to design their proposals, with the requirement that the application does not distract users, but does alert them while carrying their smartphone. Great care has therefore been taken in order to have access to high-quality data: from information relating to vehicle collisions collected by the NYPD to the state of highways and urban roads, speech recognition APIs from participating companies, open data from the Department of Transportation, real-time traffic cameras or software development kits like Google GlassPebbleSamsung and many others. Organizers will distribute 50,000 dollars in prizes, and everything — beginning with the technical requirements — is described accurately and specifically to ensure that the projects meet the organizers’ expectations.

Code for… Civic programmers integrated in the government . These medium- and long-term innovation programs (that are increasingly widespread) are based on the goodwill of a group of programmers to create and donate technology solutions that solve local problems and that of a local government that welcomes them to find out the intricacies of institutions and enable them to detect obstacles in public services that can be solved with a technical solution.

We have already discussed the examples of Code for América, Code for Europe and others here . Code for Germany is in its infancy, and the Mexican case may serve to illustrate some best practices in terms of implementation. Código para la Ciudad de México (Código CDMX) is a call from Laboratorio para la Ciudad, the area of civic innovation and urban creativity within the Federal District government, which is dedicated to promoting the creative capital of Mexico City as does the MediaLab-Prado de Madrid: that brings together people from different disciplines around a forum for innovation and development of ideas. The particularity of this case – as the blog YoGobierno.org points out –  is that, while Code for America and similar organizations emerge from civil society, Laboratorio para la Ciudad and Código CDMX emerge from the Mexico City government. It was the government who “came out to seek” the support and expertise of citizens, and formalized this with the signing of a letter of intent between Miguel Angel Mancera, head of the Federal District Government, and Nigel Jacob, representative of Code for America. Código CDMX invited two different groups – developers on a full-time basis and university students on a part-time basis – to carry out a nine-month immersion. Each of the six developers (chosen from 256 candidates) had two university student volunteers (from 104 candidates) on their team.

The result of this immersion in the local administrative machinery has been six recently-presented open code applications. Traxi (from the Ministry of Transportation and Highways) is a mobile app for choosing safe taxis with high-quality service. By simply entering the taxi’s license plate number, you can find out if it has any outstanding fines, complaints or poor scores, and it includes a “panic button” and localization if you feel threatened. Verifícalo (Ministry for the Environment) is for drivers to look up their inspection obligations, payments and driving limitations. Cuídate and Chécate aquí (Health) promote healthy activities informing you of events or showing you the clinic, hospital or health center closest to your location, the services they provide, their opening times and phone number. Check out the other apps here .

And back to the basics… the data

Involve other actors, encourage the business continuity of ideas, show the long-term open data strategy… But don’t forget that if you organize an event, contest or program to develop applications with open data, everything begins by providing quality data to motivate participants. The prize or compensation is not the only incentive: projects must have a civic or commercial purpose that is appealing to all. Remember that there are ways and means to open data. For your contest with open data make sure you release large amounts of high-quality data, include an accurate description of the data sets provided, make the data freely available to anyone (and if there are exceptions, define them well), use accessible formats, encourage their reuse and distribution, facilitate the publication of linked data  and ensure that the data are as current as possible.

It may interest you