Thursday, May 13, 2010

eGovernment Challenges and Information Architecture

Government work provides profound, often deeply frustrating, and generally amazing wide-reaching opportunities to apply our IA/UX powers in the service of millions. But this year’s Information Architecture Summit had no sessions specifically about IA in government settings.

And so I asked... “Shall we lunch?”

A group of Washington, DC government IAs and representatives from Canada and Sweden formed a topic table to start the discussion. During our hour we had time for lots of questions, a few answers, and some very interesting lines of inquiry.
Going Beyond Silos
There are political and developmental challenges to delivering access to applications and data across organizations. But there is a desire among government project owners to share, pointed out Duane Degler, who consultants for U.S. government agencies.  Yet it’s hard for them to find others in the ecosystem with whom to share. Most government employees work either in delivery programs or in operational support of delivery programs.  Civil servants often find their program mandates don’t reward, enable, or even allow collaboration.  There is a need for agencies to change their processes, and opportunities to use new technologies and conceptual models to support that.

New paradigms, like Open and Transparent Governance, are indeed emerging.  Under this aegis we can see examples of data sharing and public transparency on for the US and in Great Britain. 

As another example, Murray Thompson brought up the new, Open311 API for service requests from citizens to local government.  FixMyStreet lets citizens throughout Great Britain report local problems.  San Francisco set up a Twitter hashtag to tweet issues to the 311 service, and has just released its CitySourced mobile app.  Washington, DC, San Francisco, Vancouver, and Toronto are among those who are building apps and mashups under the new API. 

The eBenefits portal that Ironworks Consulting is engaged in is a joint Veterans Affairs and Department of Defense project by mandate.  While inter-departmental and inter-organizational cooperation therefore had the green light, from the beginning it’s been a challenge to broker permission into a reality. The key was to start with those business partners who see the vision and are willing to play, and then build a sucessful platform that everyone wants to get involved with. More thoughts on challenges and timelines for big eGovernment projects are in the blog article Citizen-Centric Portals In a Nutshell.

 Cross-channel service design
Jonas Sonders talked about a Swedish social security help desk project in 2007 with the goal of increasing the number of hard questions staff have to deal with, while also reducing the number of staff.  To achieve this, they made the simpler questions self-service, put an AI chat avatar into place, and created a “my page” for answers.  

But to make this work, productivity expectations have to change (e.g. it must be okay for help desk staff to take longer to answer questions).  Content has to be written to answer the hard questions.  Everyone and everything has to be ramped up. In fact, the help desk vision is now out-stripping the underlying information structure. To keep up, solutions from an architecture point of view include, for instance, putting hooks in the system between data entry and information.

Semantics, persistence, ontologies
Thom Haller, who teaches and consults on writing for the web, introduced a discussion on “persistence.”  What are ways of capturing classifications and organizing schemes so they get carried with the related data into the future?  This is a challenge for NARA, Library of Congress, and other long-term archival receivers-of-information. Lea Taylor of MITRE pointed out that content owners ask that their blogs be archived and distributed via CD, although no user interactivity via comments are possible then.  Furthermore, what needs to be discoverable in the web-enabled government?  For instance, are tags legally discoverable? What’s the disposition policy and where does it come from?  (In the US the answer to that question is that while the National Archives would have the ultimate voice, someone at the agency in question proposes a disposition policy.)

A follow-on discussion was held on sharing vocabulary models and classification schemes.  We touched on how we might share vocabulary models (thesauri, tag models, ontologies) across groups (agencies, departments, federal/state/local, or even international) to support linking data and cross-fertilizing content. 

The good news
The bottom line is there are ever-growing opportunities for IAs to build platforms for citizen-centric government.  And to do so successfully, we need to continue bringing an enterprise-focused point of view to the table, including strategic and cross-channel thinking.

No comments:

Post a Comment