The Semantic Government

Those who follow the “Web 2.0” hype know that the hype has moved on. People are talking now about “Web 3.0” or the “Semantic Web”. However, in government, we are behind the times. We’re still talking about “Gov 2.0”. The time has come for us to move on, too.

Here in Canada (and in many places elsewhere in the world) we are seeing a growing disconnect between the citizenry and their government. That’s why we see so many calls for “democratic renewal”. People are not seeing the “once every x years” election cycle providing meaningful dialogue between citizens and representatives.

As governments scale back due to economic pressures, more and more citizens are looking to other groups (community groups, advocacy groups, charities and other non-profits) to provide what they are looking for – not to government. That’s one of the reasons we see all of these calls for “open data”. So that the government data can be made into meaningful information and maningful services by other groups. People do not believe that the government itself can make its information meaningful and relevant.

Ultimately, people are finding less and less in government that is meaningful to them in their day-to-day lives.

It is time for us to close this gap. It is not enough for us to “open government” and release our meaningless information. We need to imbue our information with meaning and make everything we do meaningful to our citizens.

Who will join me in this crusade? We need to look beyond “Gov 2.0” to a government with meaning – a “Semantic Government”.

😉

Leave a Comment

31 Comments

Leave a Reply

GovLoop

Just don’t call it Gov 3.0 yet please 🙂

While one can argue whether some of these trends fit in Gov 2.0 or semantic web, I do think you hit upon a big key – most citizens aren’t thinking about their government and just want their information and services delivered and don’t care if that’s Wal-Mart delivering water in Katrina or FEMA.

Andrew Krzmarzick

Agree with Steve below – I don’t think we have a solid grasp of “Gov 2.0” yet and are still laying a foundation of education and implementation. In fact, next week, I am speaking to a group of managers from an agency who admit that they know very little about the web-based tools that can transform their interactions with the public.

That being said, I think a blog post by UK’s Foreign Service Office last week is a good example of a government entity taking data or information and making it meaningful to the public – do you agree? Or do you envision something else? Do you think citizens are viewing this kind of content?

David Tallan

A bit of context for this blog post. Some of us had been at a staff meeting where the “2.0”s had been flying fast and furious, applied to all and sundry. It was enough to start a drinking game.

This was reported to another staff member, for whom “2.0” nomenclature is a real pet peeve. Thinking 3.0, he asked when we would see people asking for “Semantic Government” (like semantic web). I said there was a blog post in that…

I posted it here and on our OPSpedia platform. But the smiley wasn’t enough. People are taking it seriously.

Sarah Bourne

You don’t get very far if you only look in the rearview mirror. Anyone involved in making government information available should be looking at semantic (aka linked data) technologies. We’re beginning to see some already in open data initiatives, with http://data.gov.uk/ leading the way. We have folks here in Mass. hoping to do something similare here with our data sources. And my personal vision is to do more in the area of embedding metadata into our unstructured data so it, too, an be manipulated similarly to raw data.

The tools for this kind of thing are in their infancy, but if you start thinking about it now, maybe you’ll be ready when they are more ubiquitous.

Not jokin’, just dreamin’ 🙂

David Tallan

@Sarah – I think that the future isn’t so much in embedding metadata in unstructured information as automatically deriving metadata from unstructured information. We’ve been experimenting with tools like OpenAmplify and OpenCalais.

Avatar photo Bill Brantley

“As governments scale back due to economic pressures, more and more citizens are looking to other groups (community groups, advocacy groups, charities and other non-profits) to provide what they are looking for – not to government. That’s one of the reasons we see all of these calls for “open data”. So that the government data can be made into meaningful information and maningful services by other groups. People do not believe that the government itself can make its information meaningful and relevant.”

This is a grand vision but where is the accountability? Let’s say I am a third-party group that takes government information, adds a little metacontent that slightly skews the meaning, and sends that out to the public. Some citizens rely on my information and they are harmed by their use of my information. Who is to blame? The third-party group that was just relaying the open information or the government agency that produced the information? Or if the third-party group only relayed some of the information (“cherry picking”) from the government agency?

Open government is desperately needed but the question of accountability should be considered when the information becomes free and used by the public.

Sarah Bourne

@David – There are a few benefits to embedding. One is that you have more options for integrating and managing deliberately added and automatically derived metadata (I’m going to refer to this as validated metadata.) Perhaps most compelling is that you can have metadata at both the content-as-a-whole level (this article is about dogs, written by…) as well as tagging people, places, dates, and other details in-line, especially useful when they are not the focus of the piece. Another is that that validated metadata can be available to anyone, regardless of how they come upon it. For instance, semantically-aware browsers may have all sorts of features and services a a client level, and they undoubtedly will use embedded metadata rather than referring to a separate data store. The current uses of the hCard and hCalendar microdata are a good example. I think it will also have a role in determining provenance- where did this content come from originally? And of course, if it’s embedded, it’s very easy to extract it as well for back-end, large-scale searching, sorting, filtering, etc.

The automation part is key, though, because we have found that most people do not have a knack for creating it manually at the time of creation – they are much better at editing it later. And then there are the millions of pieces of content already out there!

Andrew Krzmarzick

Woot! Another awesome conversation has erupted on GovLoop (love it when that happens!). And this is really fun seriousness. 🙂

Language is important…but probably not the real focus of your post, David – right? The real crusade [ triple smiley => 🙂 🙂 🙂 ] is about metadata, unstructured information, etc…but I’ve got to disclose something here – I’m right in the thick of these conversations, and that language still confuses me a bit (and my areas of study were semantics, philosophy and theology, so love this kind of language/convo…).

So if I don’t totally get it as someone in the midst of it, will the average citizen? I continue to ask if any of this is useful for my parents – late 50-somethings who get up every day, make their contribution at work, school, church, community – and who rarely use the web except for email, posting photos, Skyping with family, Facebooking a bit, etc. My mom actually has a blog and is trying to build a web-based business….so has taken some courses on SEO, web-based marketing, etc….so moderately savvy. But semantic web is beyond them.

I’m glad Sarah brought up http://data.gov.uk since there have been some really cool uses of the data that mean something to the average citizen, like this Postcode Paper experiment as “a great demo of what kind of services are possible with data-driven paper.”

Mantra: What got us here WILL get us there…and what gets us there must link to what has been to bring everyone there together.

David Tallan

@Sarah – I think it is a multi-part part process. Create the unstructured data. Use a semantic parser to automate creation of the metadata. (Validate the metadata.) Embed the metadata through XML or some sort of tagging.

When I saw your original content I was reading it as asking people to apply a lot of metadata to their content as they create it (embed at creation) . That is really difficult to make happen. I responded wit the call to automate. But I didn’t capture that final step to make it really useful for everyone.

Sarah Bourne

@David – Ah, good! We’re thinking the same thing!

@Andrew – The semantic web isn’t really about people, or at least, users will never have to understand it any more than they understand how their TV works. The semantic web is about taking human knowledge and understanding, and putting it where machines can make use of it. The end result for citizens means things like more useful search results, because there will be a way for machines to “know” about the difference between “reading” (as in reading a book) and “Reading” (a town in Massachusetts.) It will make it possible to quickly and reliably find related information from different sources. But most people will never have to understand a bit of how it does all that!

Steve Ardire

> Who will join me in this crusade? We need to look beyond “Gov 2.0” to a government with meaning – a “Semantic Government”.

Yes we will join you in this endeavor because we’re already moving along these lines 😉

Structured Dynamics are developers of Citizen DAN http://www.citizen-dan.org, an open source Community Indicators System (CIS) offered as a Cloud Computing Semantic Portal. It’s being driven by Open Source, Open World, Web, and Semantics or what’s expressed here http://www.mkbergman.com/873/changing-it-for-good/

The CIS will support both quantitative (statistical) information and qualitative (stories and narratives) indicators to identify and track economic, environmental, cultural, and social indicators presented as graphs, trendlines, timelines, map visualizations with “dashboard” views of indicator data that can be interactive and linked. Citizens can navigate, filter, analyze, and visualize these indicators to better identify, quantify, and track quality of life issues and community well-being.

A proof of concept will be stood up in early May for a major Canadian city. More details here http://www.citizen-dan.org/details.html

Please ping me if you’d like to know more….

Steve Ardire
[email protected]

Andrea Di Maio

So in your view isn’t “semantic” a buzzword? Well, I have seen it haunting us since when we used to play with “A.I.” (and NOT the movie). In this respect, it is a vintage buzzword indeed 🙂

Avatar photo Bill Brantley

Trying this again: I think the question here isn’t about using technology to make getting government information easier. I think your question is whether the information government collects and produces actually has meaning in people’s lives. The short answer is that every bit of information that government collects has some meaning to at least one citizen out there.

What I feel is the problem is that the information is often compartmentalized and spread over different parts of the government to the point that meaning is loss. The more enterprising third parties realize this and spend time and resources and collecting the raw bits of data and put it together into something more meaningful. For example, taking weather data, economic data, and agricultural data to input into a computer model that produces information useful for citizens who want to determine where and when are the best places to plant crops for the maximum market profit.

What David seems to be proposing is that government make all of its data feeds open for these kind of mashups by third parties because government itself can’t create these innovative mashups on their own. This may be due to turf wars, functional silos, and the inherent barriers of a hierarchical structure.

And this is where I would disagree. I believe that if government employees were given the same tools and data feeds as the third parties, they could create innovative mashups that would make the work we do meaningful to citizens. Who else is closer to the processes and knows the information needs than the front-line civil servants?

So, I agree this is a worthy crusade. Where I would start is freeing the government employees so that they could take advantage of Government 2.0 and they will better serve the public.

David Tallan

“What David seems to be proposing is that government make all of its data feeds open for these kind of mashups by third parties because government itself can’t create these innovative mashups on their own. This may be due to turf wars, functional silos, and the inherent barriers of a hierarchical structure.”

I’m not proposing that here. (I may be elsewhere, but that’s another story.) Here I’m saying that others (the “Open Data” or “Open Government” proponents) are proposing that. And I’m saying that the very proposals are indicative of failures in our government (failures of meaning) that we need to rectify.

But that’s not what I’m really saying here. What I am really doing is making fun of our propensity for leaping on the next buzzword.

For a while, “2.0” was the buzzword du jour. And we still see a lot of that. However, while we still see “Procurement 2.o”, “Gov 2.0”, “Enterprise 2.0”, “Sales 2.0” and what have you, the web itself is moving on. We don’t see so much “Web 2.0” any more. It is all about the semantic web. This caused one of my colleagues to mutter “Soon we’ll see people calling for “Semantic Government!” as they look to apply the new buzzword to government.

And I said – “I could write a blog post doing that.” Apparently, I can do that very convincingly.

Firoze Lafeer

Oops, trying this again:

I’ve seen a lot of these discussion derailed into the semantics of semantics, or the proper use of the word “ontology”. I think that leaves decision makers with the impression that this idea is incredibly complicated or would require staff PhDs in information science or linguistics.

But really, we don’t need to mark up our data in a way that is complicated or 95% “correct” (whatever that means). Even a 50% effort here would make our data a lot more useful, and a lot more usable in mashups, than what we have today. That’s what I really like about data.gov.uk: it’s not very complicated or sophisticated. But it works because it’s live and available and now people can build on that and the government itself can continue to iterate this.

What I see some people doing, just my observation, is over-think this to the point that we end up shying away from all these “complicated” (loaded) things like “Semantic Web”, “Ontology”, LinkedData and RDF. But then we try to find our own non-standard (more difficult and expensive) ways in our own silos to represent entities and relationships.

So I wouldn’t worry if we’re “3.0” or “2.0.1” (or “1.99”), I think we can do a lot and learn a lot by just taking some basic baby steps in marking up our data better. It would be great if we could as a community push our own .gov’s to do just that much like what we are starting to see in the UK.

Steve Ardire

Firoze Lafeer did you read the link I posted below Changing IT for Good http://www.mkbergman.com/873/changing-it-for-good/ Open Source, Open World, Web, and Semantics to Transform the Enterprise

Here’s the last few paragraphs of the post that says embracing the open semantic can be accomplished with minimal disruption which is its most compelling aspect 😉

Open SEAS is explicitly designed to facilitate becoming an open semantic enterprise.
Namely, this means an organization that uses the languages and standards of the semantic Web, including RDF, RDFS, OWL, SPARQL and others to integrate existing information assets, using the best practices of linked data and the open world assumption, and targeting knowledge management applications. It does so based on Web-oriented architectures and approaches and uses ontologies as an “integration layer” across existing assets.

The foundational approaches to the open semantic enterprise do not necessarily mean open data nor open source (though they are suitable for these purposes with many open source tools available). The techniques can equivalently be applied to internal, closed, proprietary data and structures. The techniques can themselves be used as a basis for bringing external information into the enterprise. ‘Open’ is in reference to the critical use of the open world assumption.

These practices do not require replacing current systems and assets; they can be applied equally to public or proprietary information; and they can be tested and deployed incrementally at low risk and cost. The very foundations of the practice encourage a learn-as-you-go approach and active and agile adaptation. While embracing the open semantic enterprise can lead to quite disruptive benefits and changes, it can be accomplished as such with minimal disruption in itself. This is its most compelling aspect.

We believe this offers IT an exciting, incremental and low-risk path for moving forward. All existing assets can be left in place and — in essence — modernized in place. No massive shifts and no massive commitments are required. As benefits and budgets allow, the extent of the semantic interoperability layer may be extended as needed and as affordable.

The open semantic enterprise is not magic nor some panacea. Simply consider it as bringing rationality to what has become a broken IT system. Embracing the open semantic enterprise can help the New Normal be a good and more adaptive normal.

Andrew Krzmarzick

@Sarah – so let’s say that it doesn’t matter that citizens (and I!) don’t get it. Our agency leadership will need to understand this conversation for us to move in that direction. I think @Firoze gets at this point:

“I’ve seen a lot of these discussion derailed into the semantics of semantics, or the proper use of the word “ontology”. I think that leaves decision makers with the impression that this idea is incredibly complicated or would require staff PhDs in information science or linguistics.”

Buzzword or not, it’s inevitable that people are going to move to ‘the next thing’ – “Semantic Government” – or a new Administration will come in and we’ll abandon current progress. The language we’re using around this stuff is important because it’s what makes it stick…and shows progress toward something better?

Avatar photo Bill Brantley

@David – “But that’s not what I’m really saying here. What I am really doing is making fun of our propensity for leaping on the next buzzword.”

Well, that’s a given. 🙂 My email box and postal mail box are full of invitations to conferences on the next “greatest” thing and how it will solve all of our problems and make us all “happy, shiny people holding hands.” I’ve written a couple of blogs on the very subject – http://blog.billbrantley.com/category/humor/

Steve Ardire

Sorry to be pedantic but it’s really quite amazing that people here are not acknowledging the key message from http://www.mkbergman.com/873/changing-it-for-good/

The open semantic enterprise is not magic nor some panacea. Simply consider it as bringing rationality to what has become a broken IT system. Embracing the open semantic enterprise can help the New Normal be a good and more adaptive normal.

Firoze Lafeer

@Bill: Yeah, it’s buzzword fatigue. And I think in this case that fatigue might be harming open government and transparency.

Putting context and meaning around data is not a new idea. And adopting existing standards rather than reinventing the wheel is not a new idea.

I think if we look at it in those terms then we hopefully US .gov’s can make substantive, albeit imperfect, steps to making all this data more useful.

Avatar photo Bill Brantley

@Firoze: Totally agree. I think a large part of the fatigue is the attitude that whatever new technology is in vogue, it is claimed that it will solve all of our problems. As a professor, I received a slew of emails about how to use iPad in the classroom. Before that it was Twitter and before that, Facebook.

Like you say, none of this is new. And it sounds like you are also saying it goes beyond technology. Maybe dropping the version numbers will help to reinforce the idea that OpenGov is more than the technology.

Mary Groebner

@Savi – I totally agree.
It doesn’t matter what we call it. IT folks will inevitably latch onto buzzwords that seem like software versions (1.0, 2.0, 3.0), while folks from other disciplines and backgrounds will grab for their own sort of equivalent of that buzzword (like ‘semantics’).

Of course, the average citizen doesn’t care what we call it – they want us to be efficient, effective, transparent, open, inclusive, etc. because they want us to provide service (yes, even in the form of collaboration/partnership) to get the things done that they think government exists for in the first place.

To me, it’s all about rebuilding trust in government, which at root is no different than rebuilding trust in another human being at an individual level. Government is afterall just a bunch of individuals (of, by, for the people, afterall). If you try to rebuild trust on an individual level you start by admitting your part in the problem (I didn’t trust you, I didn’t include you, I didn’t respect you, etc.), admitting that you’re human and make mistakes (yes, this data is dirty and I need your HELP in cleaning it up or yes, we know there are useful applications that could be built upon this data and we need your HELP in identifying what they are so here, you do it – kinda like the civic apps contest in PDX or the newly announced apps for climate change contest in BC), and moving forward from there, rebuilding trust a little bit at a time over a lot of different issues. (And yes, putting data out there without metadata or context would be the equivalent of an insincere apology, where you are saying the words but you don’t really MEAN that you want to include/involve/respect them).

I think we’ve got seriously enough problems trying to convince people within government (due to a variety of reasons almost all starting with the words ‘fear of’) to go forward with this approach regardless of what we call it. So I think it’s most important to frame it in context/language that we all AGREE on that are much more foundational than any new buzzword possibly could be (i.e. public service, accountability, efficiency, of/by/for the people)

Bob Woolley

Buzzwords seem kind of useless to me. From an IT perspective e-gov needs to recognize the varying context of citizen users, who quite frankly, go to government primarily when they have a need that government may fill. We need to do a better job of making government organizational stuff transparent. Citizens need to be able to meet their needs without having to navigate government. This is a major theme for utah.gov.

Steve Ardire

Structured Dynamics are developers of Citizen DAN http://www.citizen-dan.org an open source Community Indicators System (CIS) offered as a Cloud Computing Semantic Portal. A proof of concept will be stood up in May for a major Canadian city.

The CIS will support both quantitative (statistical) information and qualitative (stories and narratives) indicators to identify and track economic, environmental, cultural, and social indicators presented as graphs, trendlines, timelines, map visualizations with “dashboard” views of indicator data that can be interactive and linked. Citizens can navigate, filter, analyze, and visualize these indicators to better identify, quantify, and track quality of life issues and community well-being.

Then at http://semtech2010.semanticuniverse.com/ this presentation will reinforce above
“Sizzle for the Steak: Rich, Visual Interfaces for Ontology-driven Apps”
http://semtech2010.semanticuniverse.com/sessionPop.cfm?confid=42&proposalid=2960

Dave Manzolillo

Good to see all the conversation on this topic. I attended (and presented a Semantic Wiki for Climate Change findings) the ISWC2009 conference in VA, and one of the Seminars was how Semantic Technologies can be applied in Government (here is the link). We used a product (TopQuadrant) to drill into Data.gov datasets to identify IT spending trends across agencies.

Besides mashing up Data.gov feeds, there is a (IMHO) a huge need to categorize and organize government content. There is already a slew of products that do linguistic analysis on a corpus of docs & data (Attensity and Content Analyst in Intel Community). The IC is definitely the early adopter in the Semantic Content Analysis front, and other agencies now desperately need this capability of making sense of the exponential growth of content. I am hopefully exploring opensource alternatives in the meantime, some more manual process maybe involved (using SME’s to tag), but still offer a lower TCO.

Call it Semantic or Web 3.0 or whatever, the need is there. Web 2.0 tools have empowered all of us, now we need to be able to understand the meaning.

Keith Moore

David, we t Open Government TV join you in helping to create jobs and making government more efficient and more realistic to stakeholders who seek to engage and do business with government to provide solutions that reflect increased diversity and promote innovation.