After e-Commerce, Can XML Drive e-Democracy?

1. Parliamentary Information and Democracy

Democracies are continually confronted with demands for ever greater dialogue between electors and elected. In recent years, the explosion of the World Wide Web has increased the possibilities and range of such dialogue. Coupled with this has been a need to bring ever more information and documentation on parliamentary and legislative work into the public arena and onto the Web.

In response to the dangers of "information overload", many parliaments have been examining how to manage their information and documents in such a way that improves transparency and access and permits greater understanding of the democratic process.

Although each Parliament has its unique characteristics, all parliamentary democracies have a number in common:

• 'Actors': the parliamentarians themselves, political parties and parliamentary groups, staff and advisers. (in some, also the executive and/or elements of the judiciary);

• 'Structures': the 'plenary' or full House, committees, delegations, Presidency and/or Speaker, political group leaderships;

• 'Procedures': legislative consultation/approval, debates, voting, formal sittings, questioning of executive, and terminology associated with them;

• 'Acts': (draft) legislation, rules of procedure, basic laws and Treaties, through to order papers, agenda, minutes and verbatim reports of proceedings;

• 'Information': Official Journals through to library studies and press releases.

All such Parliaments share a desire to organise information and knowledge in a way that is accessible and understandable. In addition, there is a growing demand to share information between Parliaments.

2. A Common Vocabulary for common building blocks

During Spring 2000, the European Parliament — as part of its internal review of information and document management — published a "Call for expressions of interest" to assess the level of support for working together to agree a common information architecture.

If parliamentary information systems, are all built from similar "building blocks", so the argument went, identifying these blocks in a similar fashion would go a long way to allowing greater integration and exchange. However, divergent and mutually unintelligible information coding and management standards have always made this a difficult objective to attain. Language diversity (both human and IT) in different systems and Parliaments, merely adds another level of complication.

During a series of meetings within the Parliament and beyond, it was argued that a critical success factor for moving forward would be the ability to stake out an "information territory" within which participating bodies would be happy to work.

3. Top-Down

Throughout this period of discussion and pursuasion, mention of technology in general — and of XML in particular — was actually kept to a minimum. This was part of a very deliberate strategy targetted at senior management, politicians and budgetary authorities who were considered as a priori unreceptive to detailed IT-talk (with the noteworthy exception of the World Wide Web).

It was considered of particular importance to appeal to these potential sponsors in terms they would respond and relate to:

• ease in following the work and activities of specific politicians, political groupings or other identified actors — graeter visibility

• ease access toinformation and documents on a common theme, procedure and activities — bring Parliament closer to the citizen

• possibilities of greater dialogue

• strengthen possibilities of legislative tracking — fear that private setor consultants or other interest groups would get there first.

In addition, a critically important factor in the 15 member—state European Union (EU) was that of transcending the language barrier: EU legislation has to be applied in every member state in the language of that state ¹.

The most common objection raised in these preliminary discussions has been: why do we need XML if most if not all of these projects can be carried out without it? Two key themes have been important in answering this question: interoperability and re-use, firstly within a particular legislative assembly and, at a later stage, between assemblies.

An important argument in favour of cooperation is that it has borne fruit on other projects, as can be seen by the existence and extensive support of Eurovoc, a 55,000 head word mulitlingual thesaurus that already offers a comprehensive common vocabulary and taxonomy covering all areas of legislation and activity of the EU. It is used by the EU's own institutions and national parliaments alike, both within the EU (for whom the 11–language thesaurus is maintained by the European Parliament) and beyond (a number of initiatives have been undertaken to translate the entirely of the thesaurus to a number of other languages).

4. Bottom up

Whilst building support amonst senior management, it has also been important to start coordinating IT initiatives in the field.Many have been lured by the promise that XML offers, particularly in those areas — notably web publishing and other "downstream" information processing trades — where "traditional" IT project management discipline is possibly weaker. The failure to coordinate and raise awarenes at the develop level can present a number of dangers.

Firstly, XML — like HTML before it — is a very democratic standard: open to all to use, easy to learn the basics (particularly through the simple reverse engineering approach that the 'View Source' property made ubiquitous in HTML) and relatively simple to deliver some added value relatively quickly. This strength also has its downside: the ease with which a whole raft of small projects will appear acoss an enterprise, each one focussing speicifcally and exclusively on its own needs and no more.

In the public sector this can be particularly acute as such a high proportion of development work will be sub-contracted: with different contractors unaware of other projects underway, and in the absence of any standards, they are likely to create what they need as they go along.

It is essential to establish among developers, a high level of "transveral" awareness at an early stage. As with database developers, who will refer to common data dictionaries or other reference data sets rather than building their own, developers keen on venturing into XML must be pursuaded of the needs and benefits to be accrued from establishing and then using a common vocabulary.

The democratic approach should be favoured, bringing on board all developers and agreeing a common reference standard together. If that does not work, if there are those who stubbornly want to go it alone, project sponsors should however have the authority to call for a restoration of the monarchy: the divine right of the enterprise to impose its will on its subjects.

5. Project development

In the case of the European Parliament, there has been sufficient evidence of a common "information community". Making the case for cooperation with other legislative assemblies has been relatively straightforward, demonstrating the possible economies of scale and the advantages of graeter information access and tranparency. In other enterprises, this identification may not be so straightforward. In either case, however, a first step must be to assess whether any standard exists or is under development, that would match specific needs for XML deployment. the number of existing standards and initiatives is ever growing and highlights the need for publicly accessible and understandable repositories.

The first assessment of the European Parliament was that there was nothing available or in the pipeline. Various bilateral and multilateral contacts revealed an interest in some possible joint venture, despite the relatively poor level of peer group communication ²

The feedback has been encouraging, with some 60 parliaments across the world expressing an interest so far in some common development initiative, and a number of others wanting to keep abreast of developments.

The next step was equally important: staking a claim to an information territory. This involves not only making a clear statement as to the scope of a development project — what is out of scope is as important to establish, as that which is in the scope of a common vocabulary — but assessing what other initiatives exist close by that might encroach through all too commonscope creep. Although this parliamentary initiative seemed very clearly defined, the existence of others involved in adjacent areas — such as legal texts ³ — needs a permanent open line of communication to be established.

The next important step, once a level of cooperative interest was established, was ensuring a firm foundation to start development work through seed funding from a project sponsor. Finding and convincing a project sponsor is key to unlocking valuable resources and to gaining high–level management approval. Potential sponsors often have little or no interest in information technologies in general, let alone a concept as specific as XML. Gaining support means speaking in terms and language that your sponsors are likely to appreciate and understand.

The European Parliament was lucky as there is a long term research project that provides funding for initiatives promoting the exchange of data between administrations ⁴. In addition, the ParlML initiative fits in well with the current interest in e-Government and e-democracy.

As the development process gets underway, and interested parties start collaborative work, an eye needs to be kept on other projects that come into scope of the identified information territory: dovetailing ongoing work with standardisation work has often been a cause of failure for such processes in the past, as they have to be rescoped or redefined as each such element comes into view. With XML, this has proved so far to be less of a problem than might have been thought. We have witnessed a higher than imagined willingness to update and change tagging and tagsets to conform to a common vocabulary: the advantages for eveyone are clear and it is precisely one od the strengths of the XML family of standards that a project's entire XML tagging can be transformed to conform with a standardised vocabulary under development elsewhere.

The process of identifying interested parties has been relatively straightforward for the ParlML initiative: public authorities are in a position of natural monopolies, an organisational advantage that private sector initiatives would not have. The similarity of the entities and artefacts that constitute the work of legislative assemblies worldwide has also facilitated the identification of semantically equivalent building blocks, the work that constitutes an important part of the initial grunt work of any XML standardisation exercise.

Apart from the parliamentary administrations, there is a whole range of other interested parties lining up to get involved in the project, all interested in the benefits that a structured presentation of legislative work can offer them. Although the exercise is to be an open one, it was decided to restrict the very initial discussions and exchanges to those directly involved in the work of legislative assemblies, in order to preserve some initial reflection and thinking space and attempt to map out a clear development process and core standards approval.

6. Conclusions

Once an "information territory" has been identified, it is vital to:

• proselytise: inspire potential sponsors and partners with a vision of the benefits that greater interoperability and transparency.

• provide top-down guidance for a controlled and managed development and comprehensive view over your information territory, and minimising the risks of change

• envision the future: assess the organisational changes that may be needed to take on, and make most use of, the proposed new XML-centred architecture.

• deliver initial value and benefits as early as possible, in order to start a snowballing of initiatives through positive feedback.

• prepare to scale fast

Critical success factors include the ability to demonstrate that an XML-centred strategy meets a real need and that cooperating with others who share the same information space is to mutual advantage, whether competitors or not. The objective should be to promote an overall project and particularly any standardisation effort as a management led initiative without getting into the details of XML standards at all.

Design and Development by deepX Ltd.