On a daily basis, throughout the UN system, staff of conference and meeting services, publications offices, libraries and information departments face significant challenges to manage UN corporate documents and data effectively in order to meet the growing needs of all stakeholders – Member States, Permanent Representations, sister organizations and their staff, the media and the public at large.

The complexity, interconnection and cross-references of UN system normative and parliamentary documents have increased dramatically in the last decade. UN documents contain complex, multifaceted mandates that may concern several organizations and entities, require different compliances, and be regulated by different legal environments or governing bodies, which in turn may need to report through their own processes and parliamentary documents. Moreover, normative texts are frequently amended with cascading effects on mandated activities, considerably increasing the intricacy of the legal ecosystem of the UN system. Verbatim reports, summaries and votes, frequently addressing multiple items in a single document, pose similar complexities.

The 2030 Agenda for Sustainable Development and its implementation requirements have placed new and increasing demands on the ability of the UN system to harmonically manage information and knowledge to effectively measure and document progress, link policy development to data, and foster accountability of all stakeholders at all levels.

Traditional web pages (HTML) and portable document format (PDF) files are difficult for machines to interpret. They do not have a structure that is suitable for automated analysis; they are meant to primarily display information on a screen or on paper, and they are meant for publishing and not for processing information.

To fully exploit the computational power of information technologies to process the content of UN documents, the UN system must transform this web of documents into a web of data that may be interpreted by computers.

To do so, there is a need for the UN system of organizations to agree on standardised ways to describe the meanings and concepts of textual information in machine-readable fashion in order to enable the ability of computer systems to process and exchange data with unambiguous and shared meaning. There is a need to use shared vocabularies and establish associated links to formal models (ontologies) that allow knowledge to be represented in ways that enables machine computable logic, knowledge discovery, and data federation between information systems.

A UN common machine-readable format is aimed at enabling the UN system of organizations to:

Enhance the efficiency and effectiveness of production and dissemination of parliamentary and normative documents;

Respond to the growing demands of stakeholders for smart information services;

Improve the quality of information dissemination and accessibility; and;

Bring significant benefits in terms of governance, accountability and transparency.

High-Level Committee on Management (HLCM), in its 2013-2016 Strategic Plan, called for the “development of innovative and sustainable business solutions” and highlighted that “new technologies open entirely new horizons to re-shape the operational models of UN organizations”. The HLCM embraced the use of ICT as an agent of change and as a driver of improved knowledge management and developed its agenda on a platform of Digital Transformation of the UN system built on the pillars of transparency, standards and data, and information security.

The results of an HLCM system-wide survey carried out during the first semester of 2015 confirmed a very limited use of semantic technologies in the UN system, but, at the same time, showed significant support for their adoption as a way to generate considerable efficiency in document management processes, as well as for a system-wide approach to the introduction of information standards.

At the Thirtieth Session of the High-Level Committee on Management (London, 5-6 October 2015), the Food and Agriculture Organization of the United Nations (FAO) presented its pilot project “Achieving Greater Access, Visibility and Efficiency in FAO’s Normative and Governance work”, which aimed at testing the suitability of Akoma Ntoso (AKN) XML for the UN documentation environment. AKN is an OASIS XML candidate standard for parliamentary, normative and judicial documents originally developed by the United Nations Department of Economic and Social Affairs.

The meeting expressed general consensus on the adoption of open source standards based on the eXtensible Mark-up Language (XML) for efficiency gains in document management processes and underscored that approaching the critical domain of information and knowledge management with unity of intent was strategically important for the UN system.

The Committee noted the added value of a joint initiative towards the adoption of a UN Semantic Interoperability Framework across the system as part of the Committee’s strategic priority to use new technologies to re-shape the operational models of UN organizations. By approving the project “A system-wide semantic framework for UN normative and parliamentary documents”. The HLCM requested the establishment of a working group, open to representatives of the documentation, library and conference management offices, that, by building upon the FAO experience with Akoma Ntoso (AKN) XML, would have defined a common machine-readable format for parliamentary and normative documents for the UN System.

In June 2016, the HLCM Secretariat, in collaboration with FAO and UN/DGACM as co-leading entities, set up the Working Group on Document Standards (WGDS) to work towards the definition of a UN Semantic Interoperability Framework (hereinafter UNSIF). The activities of the WGDS have been jointly financed by the HLCM’s HBP Trust Fund, FAO and UN/DGACM.

The HLCM Working Group on Document Standards was tasked with defining a common set of guidelines for the semantic mark-up of UN documents based on Akoma Ntoso, an open document standard for parliamentary, legislative, administrative, executive, and judicial documents defined by the LegalDocumentML Technical Committee1 of the Organization for the Advancement of Structured Information Standards2 (OASIS).

The objective was to provide the foundations for the deployment of smart services for the management of parliamentary, normative, judicial and official documents to increase the efficiency of the parliamentary processes, reduce the administrative and financial burden, facilitate the re-use and preservation of official documents, and realise significant gains in terms of accountability and transparency.

These Guidelines are the outcome of the work (Jun 2016 – Mar 2017) of the HLCM Working Group on Document Standards, composed of members from the following participating organizations:

The WGDS has met regularly online from July 2016 to March 2017 to carry out joint analyses and consultations and to work toward the definition of the UN Semantic Interoperability Framework for normative and parliamentary documents (UNSIF) as formed by:

Akoma Ntoso for the United Nations System (AKN4UN), which defines the guidelines for the localisation of the Akoma Ntoso XML standard to the specific requirements of UN parliamentary and normative documents.

The United Nations System Document Ontology (UNDO), which provides a formal representation of UN document basic entities and their relationships.

This document, “Akoma Ntoso for the United Nations System: Guidelines for the mark-up of UN normative, parliamentary and judicial documents” (AKN4UN), defines the technical specifications that the UN system of organizations should follow to localise the Akoma Ntoso OASIS standard to the specific requirements of each organization and document type.

AKN4UN is a living document, which will be subject to regular updates to incorporate new developments and requirements. It will be available at the following webpage: akn4un.unsystem.org. The online version shall be considered the latest and authoritative version of the Guidelines.

The Guidelines set different levels of requirement for the mark-up of the different documents. In this document:

Mandatory requirements are indicated by the capitalised verb “MUST...”.

Highly desirable requirements are indicated by the verb “SHOULD...”.

Desirable requirements are indicated by the verb “MAY ...”3.

Each requirement is labelled as: (M) = Mandatory; (HD) = Highly Desirable; (D) = Desirable.

The Guidelines are to be used as a reference tool by information managers and information technology managers of UN system organizations, as well as the software vendor/integrator community who plan to develop, deploy or manage systems that use (or intend to use) Akoma Ntoso XML documents.

The OASIS LegalDocML official documentation is accessible online4.

On the Akoma Ntoso website5 you will find a wide range of information and material and you may also join the Akoma Ntoso mailing list6 to discuss issues related to its implementation.

XML element names are denoted with angled brackets and blue colour font. For example, <element> is an XML element.

XML attribute names are denoted with an “@” prefix and blue colour font. For example, @href is an XML attribute.

In the narrative of the documents you will find other terms in angle brackets, e.g. <attachments> or <docType>. This means that the terms, such as “attachment” or “doc type”, are used with the specific meaning that Akoma Ntoso assigns to them.

While notes to highlight specific issues will be formatted like this sentence

Terms in { } curly brackets are placeholders that should be replaced with the proper contextual values, e.g. in /akn/un/act/codesOfCoduct/un-ga/{date}/{number}: {date} must be a date value from the identification metadata of the document, and {number} must be the recognized official number of the document.

In AKNthe data-type of the attribute value is prescribed by the schema and in some cases the attribute value is restricted further by the schema to include only a set of prescribed values. In these Guidelines we follow the following convention for the values of attributes:

{akn:list} – this indicates that the value for the attribute may only be set to the values prescribed in the schema. For example, the @status attribute has a set of 10 prescribed values in the AKN schema (“removed”, “temporarilyRemoved”, etc.) No value other than the prescribed set may be used in this attribute. These values are schema restricted, so setting invalid values will result in a technical failure during validation of the XML using the AKN schema.

{akn4un:value} – this indicates that the set of value or values for the attribute is recommended by the AKN4UN guidelines. Note that these values are not prescriptive, i.e. setting something other than these recommended values will still allow the XML to pass technical validation using the AKN schema, and compliance with AKN4UN.

{value} – this indicates that the value may be set to anything as long as it meets the data type rules for the attribute as set in the schema. If the schema says a string is allowed, any string may be safely set, if the schema says only a date value may be set, any date value may be set.