Many people want to be able to contribute to the development of government policy, either as key stakeholders or citizens. The problem is finding out about what is going on to which they can contribute. Key stakeholders get invited – for example the British Chambers of Commerce or the British Computer Society. Others are aware through looking at the website, setting up alerts or monitoring via RSS, or through other means including links and third party information.
What would be useful is one place to find all the consultations that are open at any time. Harry Metcalfe sought to do this in his service www.tellthemwhatyouthink.org, but found the identification of where all the consultations were and the different ways they are structured difficult in providing a full list. Of course a list that looks complete but isn’t is the most frustrating of all – potential contributors don’t know what is missing and may miss something important because they are not looking elsewhere.
The Consultations Code committed to a complete list of open consultations. This is now being formed with Directgov with a target date of the end of the year. And we’re doing it using semantic web mark-up (RDFa) so that anyone can extract the data and use it. I see the possibility of key stakeholders downloading the information about consultations directly onto their websites and providing online response forms using social media tools that can then be integrated and fed back to government.
The commonest question I get, is why not use plain old XML data streams? We could, but there are many useful aspects to open government if we use semantic web mark-up. Before that stage, putting all descriptions of consultations into a common form, helps people identify quickly what is relevant and allows people to bring together into new services. Ensuring each has a URI by a single web page to itself (using the URL as a URI), allows reference to where all the documents are. Making them indexable by Web search engines (not all were!) means that people can find all the different consultations, including relevant ones, even if you didn’t know of all the organisations that have consultations. Finally putting in RDFa markup means that they are re-usable, so third parties can encourage participation. Government providing a service by creating a single list from this data on all the different public sector organisations is just one use of that data.
Formal consultations are only one way of many that seek to engage the public. We could apply the same principles – and display them in a single place – for other time-limited means. And we could bring in statutory notices that include such items as planning notices, which are mini-consultations, encouraging people to comment. When I go onto my professional sites, I want to see relevant policy discussion. I also want to do so in my personal life – identifying all those changes that might affect my locality and my interests.
Posted in: Semantic web and information re-use
Tags: consultations, RDFa, search, XML
That’s the best and most readable argument for RDFa-enabling consultations that I’ve yet come across, thanks. I think you set out an interesting vision there – there’s obviously work to do on the content of consultations to make them more accessible to a wider audience (not least their length and language).
RDFa implementation is tricky though. I’m not scared of a bit of code, but I’ve struggled to pull together the spec for how our templates need to change. Can you point me (and others, I suspect) to some links to the source documentation and a readable implementation guide?
Steph Gray says:
September 16, 2009 at 8:16 pmAlthough the provision of a full feed of government consultations is a laudable thing, I feel as if I should play Devil’s Advocate and ask “so what?”
When we launched WriteToReply we suggested that Government departments should: a) publish an RSS feeds of consultations; b) using an autodiscoverable RSS feed.
[Ref: http://writetoreply.org/actually/2009/02/09/whats-consulting-now/ ]
This would not only have simplified Harry’s aggregator at TellThemWhatYouThink, it would also have provided an easier to locate channel from particular departments out to special interest groups who track consultations coming out of those departments.
Having a complete list of open consultations is a handy resource, but I worry about how that list will be used to actually improve participation in consultation exercises?
That is, even if we have a consultations fire hose, we can still ask: so what? How is this going to help engage people in actual consultations? Most people aren’t likely to be interested in most consultation documents. The fire hoses that currently exist at TellThemWhatYouThink and DirectGov are 80% full enough that we should be able to demonstrate what we’d do with the full 100% feed. But I’m not sure that any such compelling applications exist? (I’d love to be proved wrong on that:-)
One thing I might guess at is that a 100% aware site should be able to aggregate *and then resyndicate* consultations on particular topics that are likely to appeal to micro-targeted audiences. Which means an effective search algorithm, faceted views over consultations wherever they come from, and the ability to easily define filters. And also the ability for users to set up and subscribe to alerts about consultations on a particular topic. (It’s not just consultations that this might be useful for; tracking amendments to legislation is another obvious example – e.g. http://ouseful.wordpress.com/2009/01/23/tracking-uk-parliamentary-act-amendments/ )
Taking a further out view, I could imagine an ‘adserver’ like platform that would serve links to relevant consultations *as if they were adverts*. (For this, you have to be of the mindset that relevant ads are content/news in their own right!;-)
As well as having a list of consultations, all announced in a standard way, I also feel obliged to ask whether or not you see the consultation documents themselves being published in a common format?
When we set up WriteToReply, first as a commenting platform for the Digital Britain Interim report, then as a more general public document consultation platform, we soon realised that no two consultation documents are structured in the same way!
However, with a bit of processing, cut’n'pasting and hand-finishing, we have been able to shove the various documents we are hosting into a quite simple HTML representation.
Just having a document hosted on WTR is only the first step towards getting people to engage with it, however. Finding an audience willing to engage with the consultation is the far harder next step. One approach we have taken is to use simple polls related to the consulation document (for example, http://ouseful.wordpress.com/2009/07/05/idle-thoughts-on-micro-consultations/ ). Another is to use the the links generated for each paragraph for a WTR hosted report as hooks back the consultation document that are baited with provocative questions or calls to action posted to Twitter and as comments on third party blogs.
We are currently using this approach with a set of pre-scheduled tweets that are aimed at engaging members of the educational technology community with the current JISC Strategy Review consultation [ http://writetoreply.org/jiscstrategyreview/ ]. Each tweet has a question and a link back to the appropriate paragraph of the consultation document. (Each paragraph can be commented upon individually.)
Another approach we are starting to explore is the use of automated semantic tagging to broker links *between* sections of a particular document as well as between related sections on different documents. Whilst this approach won’t necessarily bring in new people to a consultation, it will hopefully provide a way of navigating through related consultations once you have engaged.
Looking forward, we have begun implementing a system for allowing individual paragraphs within a consultation document to be transcluded (that is, embedded) elsewhere on the web. The idea here is that from a common source, third party sites will be able to include ‘remote quotations’ that pull content from the source consultation document into the third party web page. Why? So we can increase the surface area of the consultation document, and re-present it elsewhere, as well as providing ways of tracking (and hence re-aggregating) the remote commentary wherever it happens to occur.
Hmm… I’ve been rambling again, haven’t I..?! I probably should stop now!
Tony Hirst says:
September 16, 2009 at 11:12 pmDavid that is very well put – so why do I feel like it has now become too complicated? I understand your RDFa points, and understand the work that needs to happen – yet, working in a department, and looking at this with a long term view, I am a bit *hmmm* about it all.
I think that if you, and knowing you I do mean ‘you’, read this, you would question the broad brush of consultation.
The data, RDFa, linked and semantic web work, is what will enable great services to be developed. I am not sure that it resolves getting the right people engaged with the policies that affect them the most. We seem to be being drawn into a tech hole here that actually gets away from the point.
Policy owners need input as the policy develops. Citizens need to input to the subjects that are relevant – but not be deluged with information. Local government issues need to be separated from central government, without the big brother: ‘we know you live here, are interested in this so, here’s your neighbour’s planning app, waddya reckon?’and without the aforesaid, how can semantic/linked data be used effectively?
It’s all doable. The idea: ‘I see the possibility of key stakeholders downloading the information about consultations directly onto their websites and providing online response forms using social media tools that can then be integrated and fed back to government’ is fine, except that often just the stakeholder is the one who could have the knowledge to feedback.
I don’t know. I think that I am wary of offering up tools because we can to *all* consultations, when in fact, not every consultation is appropriate for general chat.
I agree that we should make it so that we can do this, but disagree that this should be a government wide policy. Does that make sense?
Emma Mulqueeny says:
September 16, 2009 at 11:53 pmThanks for nice comments. We do need to develop some guides to how to make consultations accessible as part of open government. We circulated the first draft in hard-copy of what needs to be added in to all heads of e-comms last week for their comments and advice, before we circulate more widely. That has the links to the google.code store where the primary documentation is. I’d be interested in whether this answers your questions or not, perhaps we need something more specific?
David Pullinger says:
September 17, 2009 at 10:13 amWhen I was writing I was thinking about how many of the key stakeholders are also member organisations – British Chambers of Commerce, British Computer Society and Royal Yachting Association, etc., as well as the private sector. They solicit the views of their members before replying, as I expect the large distributed commercial companies do in order to gather all their expertise. It is an interesting question about whether you get better engagement and policy formation through this approach, or by encouraging individuals to reply directly. The latter will pick up some expertise that may not otherwise be covered and is in line with seeking a new collaboration and conversation with citizens. In this time of exploration, it may be best to give a variety of routes and see what flourishes. This is what semantic web can help offer.
David Pullinger says:
September 17, 2009 at 10:30 am