4 Design issues (Non-Normative)

4.1 Simple data model

• 4.1.1 Akoma Ntoso XML-Schema Defining an XML language goes through four different specifications: This 4-part distinction is explicit in Akoma Ntoso, and is used to ensure the long life and widespread usefulness of all documents expressed in this language.
  • The namespace, i.e., the official and unambiguous identifier and name of the language (in Akoma Ntoso, that is http://docs.oasis-open.org/legaldocml/ns/akn/3.0)).
  • The vocabulary, i.e., the set of reserved words that will be used for the language. In XML the vocabulary is used to specify the name of elements and attributes of the language. Currently, Akoma Ntoso defines 310 names for elements and 69 names for attributes and it uses the lower camel case naming convention for both elements (e.g. mainBody, amendmentList) and for attributes (e.g. showAs, refersTo).
  • The grammar, i.e., the rules that are used to build a correct (or, in XML, valid) instance of a document in the XML language being defined. The grammar is composed of rules that dictate what content is legal to appear within any element (which is called the content model), both in terms of other elements and characteristics of the text itself.
  • The semantics, i.e., the mapping between the vocabulary and rules being used in a valid document, and the actual meaning inferable from its markup. The semantic of an XML markup is absolutely dependent on the kind of use the markup document is subject to (often called the downstream application). For ensuring multiple different uses, both by humans and computer applications, declarative semantics is preferred, where by declarative we refer to the description of the element as declaring the content as it is in terms of structure, role or purpose, rather than as it should be handled by any specific downstream application.

4.1.2 URI/IRI

URIs/IRIs, or Uniform Resource Identifiers/Internationalized Resource Identifier, are standard mechanisms for referring to documents, languages and concepts on the World Wide Web. A good URI/IRI has either an identification purpose (i.e., it provides a way to universally refer to that resource in a manner that does not change with time, computer systems or software versions) or a location purpose (i.e., it provides a way for a software or a human to unambiguously and rapidly access the resources wherever it is stored), or, on some situations, both.

Akoma Ntoso gives a lot of importance to URIs/IRIs, and provides systematically specific URIs/IRIs for all documents, concepts of the ontology, and even for the markup language itself. All such URIs/IRIs are described in the Akoma Ntoso naming convention.

4.1.3 FRBR

The Akoma Ntoso standard defines a number of referenceable concepts that are used in many situations in the lifecycle of legal documents. The purpose of this section is to provide a standard referencing mechanism to these concepts through the use of URI/IRI references associated to classes and instances of an ad hoc ontology. The referencing mechanism discussed in this document is meant to be generic and evolving with the evolution of the underlying ontology.

The most important concepts of the Akoma Ntoso informal ontology are related to documents that have legal status. All discourse and all description of legal sources can be characterized as referring to one of the four levels of a document as introduced by IFLA FRBR (International Federation of Library Associations (IFLA) - Functional Requirements for Bibliographic Records (FRBR) http://www.ifla.org/VII/s13/frbr/frbr.pdf)):

• WORK: the abstract concept of the legal resource (e.g., act 3 of 2005).

• EXPRESSION: any version of the WORK whose content is specified and different from others for any reason: language, versions, etc. (e.g., act 3 of 2005 as in the version following the amendments entered into force on July 3rd, 2006).

• MANIFESTATION - any electronic or physical format of the EXPRESSION: MS Word, Open Office, XML, TIFF, PDF, etc (e.g., PDF representation of act 3 of 2005 as in the version following the amendments entered into force on July 3rd, 2006).

• ITEM: the physical copy of any manifestation in the form of a file stored somewhere in some computer on the net or disconnected (e.g., the file called act32005.pdf on my computer containing a PDF representation of act 3, 2005).

4.1.4 Ontology

In computer science, an ontology is an organized collection of facts and assertions about a specific domain. Ontologies identify a number of classes of relevant concepts and their properties and the relations between these classes. In a properly organized ontology, through classes and properties it is possible to derive (technically, infer) new properties relating instances of the classes even if there are not explicitly present in the description of the instances themselves.

Within the World Wide Web, the discipline of ontologies is gaining visibility and widespread adoption, thanks to the initiative called the Semantic Web by the W3C. Within this initiative, several languages have been defined, including RDF [RDF], RDF Schema and OWL. Such languages allow specific ontologies to be defined, mixed and interchanged for a wide number of purposes.

Akoma Ntoso allows multiple different ontologies to be created about the document it describes. Rather than defining an ontology of the legal matter being discussed in the legal documents (which could be overly wide and all-encompassing, since the legal matter may be by itself rather wide and allencompassing), the Akoma Ntoso specification provides mechanisms to define arbitrary ontologies and connect them to the various parts of the document and of the text. Akoma Ntoso informal ontology. The Akoma Ntoso informal ontology is therefore centred on the concept of document, which is considered in a very precise way through the specification of the FRBR conceptualization of documents (about which see the next section). Besides the FRBR classes for document, the Akoma Ntoso informal ontology also lists a number of supporting classes (such as Person, Organization, Place or Event, that are used to provide further meaning to the main classes of the ontology.

4.1.5 Design patterns

Patterns are the abstraction and distillation of past experiences in designing and resolving design problems. They are general and widely applicable guidelines for approaching and justifying design issues that often occur in XML-based projects.

In Akoma Ntoso, patterns are used to create categories of content models (and thus correspond to only those content models that have been found to be actually useful) and, more generally, in schema design (and thus correspond to guidelines on how to make the schema more modular, flexible, and understandable to by users). Both approaches are well known and well established in the literature, although by different experts in different ways.

4.1.5.1 Categories in content model

Categories of content models is the term used within Akoma Ntoso to refer to families of elements that share the same conceptual organization of the internals. The Akoma Ntoso schema uses six categories of content models. This means that all content models and complex types used in the schema follow precisely the form of the relevant category, and all elements can be simply described and treated according to their category rather than individually.

These categories are: The markers: markers are content-less elements that are placed here and there in the document and are meaningful for their position, their names and their attributes. Markers are also known as empty elements or milestones. There are two main families of markers in the Akoma Ntoso schema: placeholders in the text content (e.g., note references) that can appear in any position that also has text, and metadata elements that only appear in some subsection of the <meta> section. In Akoma Ntoso, all metadata elements are markers, so that metadata values are not part of the text content of a document, but rather become attribute values. The inlines: an inline element is an element placed within a mixed model element to identify a text fragment as relevant for some reason. There are both semantically relevant inlines and presentation oriented inlines. There is but one content model using inlines (and markers), which means that all mixed model elements (i.e., those that allow both text and elements) also allow a repeatable selection of all inline elements. For a discussion of why this is only a trade-off decision, and not the ideal solution, see the discussion at the end of this section. The blocks: a block is a container of text or inlines and placeholders that is organized vertically on the display (i.e., has paragraph breaks). Most blocks in Akoma Ntoso are based on the HTML language. There is only one content model that uses blocks, and it allows a repeatable selection of all available blocks. This means that wherever any block is allowed, all blocks are allowed, as well: e.g., wherever a paragraph is allowed, a table or a list is also allowed. The subFlow: a subFlow element is an element placed within a mixed model element to identify a completely separate context that, for any reason, appears within the flow of the text, but does not belong to it or does not follow its rules. subFlow elements are containers appearing in the middle of sentences but containing full structures (with no direct containment of text or inline elements). The containers: containers are sequences of specific elements, some of which can be optional. Containers are all different from each other (since the actual list of contained elements vary), and so there is no single container content model, but rather a number of content models that share the same conceptual category. The shared characteristic of containers, is that no text is allowed directly inside them, but only a collection of other elements. Text therefore can only be placed within a block within the container. The hierarchy: a hierarchy is a set of sections nested to an arbitrary depth, usually provided with title and numbering. Each level of the nesting can contain either more nested sections or a container. No text is allowed directly inside the hierarchy, but only within a block element that is contained within a container element (not considering, of course, titles and numbering). Akoma Ntoso uses only one hierarchy, with predefined names and no constraints on their order or systematic layering. There are two exceptions to the systematic use of patterns:

• The <li> element allows both inlines and other nested lists (<ul><ol><li>). The pattern would require elements to contain only text, and nested lists to be direct child of the main list. Since this goes against universal HTML practice, we have decided against full pattern adherence and in favour of HTML tradition.
• There are some inline elements that only make sense in the preface and/or preamble of the document: for instance there, are <docTitle> and,<docNumber> , for numbered documents such as acts or bills, or , for judgments. They are, in fact, part of the one inline content model and thus are available everywhere in the document. There is no simple way to define blocks within <preamble> and <preface> to allow these elements and blocks elsewhere to reject them, so it has been decided that it is better to allow them everywhere rather than uselessly complicating the schema.

4.1.5.2 Patterns in schema design

Design patterns are a distillation of common wisdom in organizing the parts and the constraints of a schema. Some of them are listed in http://www.xmlpatterns.com/. Whenever there has been a design choice to be made that was not immediately obvious and naturally acceptable, a relevant pattern has been sought and properly used. In fact, http://www.xmlpatterns.com/ also contain immediately obvious and naturally acceptable patterns that have been used in Akoma Ntoso, but only the not-so-obvious and not-so-natural ones have been explicitly mentioned and referred to. You can find the relevant references in comments within the schema itself, and in the documentation.

4.2 Widest scope

4.2.1 Support for all the types of legal documents

Akoma Ntoso provides explicit support for many different document structures within the context of parliamentary and judiciary activities: legislative documents (e.g. bills, acts, etc.), amendment documents (e.g. amendment), parliamentary documents (e.g. debates, hansard, report, etc.), judiciary documents (e.g. judgments), collection documents (e.g. Official Gazette, etc.), general document (e.g. annexes, memorandum, etc.).

The information organized within the documents corresponds to various typologies of documents.

Akoma Ntoso Document Types	Category / Legal Document	Definition
bill/act	Akoma Ntoso type : hierarchicalStructure Legal Document: bill/act/ordinance/decree/subsidiary legislation/executive orders/ normative standard/ administrative regulation/etc.	These are deliberative documents produced by parliamentary activities or from other empowered bodies (e.g. Committee). They are usually drawn up according to a hierarchical structure in which the text is subdivided into sections or chapters. These are subdivided into clauses or articles, sub-paragraphs, etc.
debate	Akoma Ntoso type: debateStructure Legal Document: debate record/Hansards/judicial verbatim	These are texts resulting from the transcription of the any deliberative works. The structure reflects the different section of the debates and alternation of questions and answers that takes place during the parliamentary works. This structure could be used, for instance, for parliamentary/judicial verbatim, UN/FAO transcript of assembly or council, etc.
debateReport	Akoma Ntoso type: openStructure Legal Document: committee minutes/judicial minutes	These are texts that are minutes or reports usually of the committee used to describe official meeting sessions.
judgment	Akoma Ntoso type judgmentStructure Legal Document: judgments//caselaw/precedents	These are documents in which a court of law makes a formal decision or specific determination following a lawsuit. The structure reflects a typical narrative of sentences.
doc	Akoma Ntoso type: openStructure Legal Document: any other type of document/Executive SummaryMemorandum/etc annexes/table/judicial documents	These are texts that are legally valid but do not have any particular structure. These include any parliamentary procedure document that has no particular textual structure and is not a collection of other documents. An example could be also the Report of the Amendments of a Bill, the Memorandum of a Bill, Order of Business, Legal Notice, judicial documents, etc.
documentCollection	Akoma Ntoso type collectionStructure Legal Document: collection of documents	Used to represent documents which are collections of other independent documents. A typical example is the electronic folder related to a bill. This folder is composed of several independent documents (committee reports, initiative, bill, memorandum, etc.) and by different expressions over time such as versions of the same bill. Another example is the U.S. Code: it is a documentCollection composed of various Titles of positive and non-positive law. In European institutions, the committee report, for example, can be considered as a document collection, as it includes documents like Resolutions, Explanatory Memorandum or Opinions from other committees.
amendmentList	Akoma Ntoso type collectionStructure	Used to represent a special document that includes all the amendments, collected and submitted to the official deliberative body for the discussion.
officialGazette	Akoma Ntoso type collectionStructure	Used to represent an issue of an official publication body such as Official Gazette, Journal, Bulletin or Federal Register.
amendment	Akoma Ntoso type amendmentStructure Legal Document: Amendment document	Used to describe specific amendment documents. It is a special document or a component of an amendment list, presented by the member(s) of parliament to the committee or to the assembly for discussion and vote.
statement	Akoma Ntoso type openStructure Legal Document: US resolution	Used to represent those legal documents that may or may not be normative, but they are fundamental for the life of an official institution. Often they have a mixed structure (blocks and hcontainers): an example is a resolution issued by the US Congress, or an UN resolution/declaration. Other examples are the resolution or decision from European Parliament.
portion	Akoma Ntoso type portionStructure Legal Document: chapter of a document	Used to represent a portion of any document at manifestation level.

4.2.2 Support for all the uses of legal documents

Akoma Ntoso is designed for use in all applications that use legal documents. This includes applications both inside and outside the deliberating bodies that make the law. Akoma Ntoso includes, without being limited to, support for:

• the initial drafting of bills;
• the legislative lifecycle including amendments, publication in the official gazette, and the recording of debates;
• the comparison between different version of the bill;
• the enactment and consolidation of those bills to produce the law;
• the codification, recasting, coordination of the acts when some changes are issued;
• the publication of the law and comparison between two different versions of the same law; and
• applications that involve the research and tracking of laws and legislation.

Akoma Ntoso also gives to the applications a representation of case-law, precedents, and judgments including those produced by the Constitutional Court that can affect the law.

4.2.3 Support for all the actors dealing with legal documents

Legal documents are important to many different people and organizations. These range from the people who originally request or propose new laws, the person tasked with drafting the legislation, the legislators who sponsor and debate the legislation, the people who want to alter that legislation, the person who signs the legislation into law, and the people affected by the resulting law. Akoma Ntoso provides support for this diverse group of actors involved in the legislative process.

4.2.4 Support for all the processes affecting legal documents

There are numerous processes that involve legal documents. Some processes within governments or institutions involve the production and issuance of legal documents. Other processes, by other government agencies or external entities interested in following or conforming to the law, involve the tracking and consumption of legal documents. Akoma Ntoso is designed to support all the processes that involve legal documents, whether it be the initial drafting of legislation, the process that results in the enactment of laws, or the follow-on processes to track and comply with those laws.

4.2.5 Support for the characteristics of legal documents in all countries and jurisdictions

Every country and every jurisdiction has unique requirements. This is a simple consequence of separate development of legal traditions around the world over time. However, upon further examination, it is quickly apparent that all the varying traditions found around the world stem from a relatively small set of legal traditions originating back in history. Akoma Ntoso has been designed, through careful examination of the world's legal practices, to take advantage of the common heritage found in all our legal systems while also providing enough flexibility to adapt to all the variations.

4.2.6 Support for all legal documents of the past and of the future

It is important that a legal data model support not only the future needs of legal information systems but also the past. Akoma Ntoso is designed to anticipate the future needs made possible by a uniform standard for legal documents while also being flexible enough to adapt to past practices, allowing all the variances that have occurred in the past to be modelled in a single document structure.

4.2.7 Long term preservation

Dematerialized legal documents modelled and represented in XML preserve their legal validity over time, maintaining a clear separation between original content (as formalized in the enactment stage) and the reworking of that text during the reporting process. This allows us to include a digital signature in the XML document, thus freezing authenticated documents, even digital ones, so that it can be represented in the future without subsequent modifications.

4.2.8 Self-explanation

It should be possible to understand the markup of a legislative document without having to first study and understand the associated schema or having to possess any knowledge of any special theory behind the design. For this reason, the vocabulary should adhere as close as possible to the legal domain terminology, while also being as neutral as possible with respect to any legal-specific tradition.

4.2.9 Self-containment

A good legal XML schema must encapsulate knowledge in one self-contained document without fragmentation in the logical schema of a database or document processing application. This preserves a document’s neutrality with respect to applications, platforms, and technological developments. It also keeps intact the expressive power of the legal knowledge contained in the document so that the document can move freely throughout the network.

4.3 Strong distinction between authors and editors

The first important point is the explicit and complete separation between the role of authors (who decide and write the actual content in terms of sentences, words, and punctuation) and of editors (who decide and organize the final layout and publication of the document). We often say that the author has created the content and the editors have created the metadata. Another way to put it is that the author is the creator of the FRBR Expression, and the editors are the creators of the FRBR Manifestation.

In the field of legal publishing, the author is often an abstract concept (e.g., the legislator), whose content is the result of a formal action (e.g., a final vote of approval for a highly discussed text), while editors can be involved at all stages of the publication process. In this regard, the identification of what constitutes the content and what is an editorial addition is in many cases subtle and difficult to establish. A rule of thumb is to try to determine the state of the document at the moment it left the hands of the author and was taken in by the editors. For instance, even the publication of the document on the official gazette does not determine clearly the “official” content. Some published data (such as, for instance, the number of the gazette itself) were not in the hand of the official authors and as such should be considered metadata and not content.

This basic distinction generates, on the other hand, a few secondary reflections that need to be discussed briefly.

4.3.1 The official form is the guarantee of the authorial intention

Many types of legal documents have a “more important” form of publication than others. We will call this the authoritative (or official) form. More often than not, this is a version of the document printed on paper and published within official channels (e.g., the official gazette) after a number of well-known and highly controlled editorial steps. All conversions into electronic formats, by their very nature, have an authoritative status that is of lesser authoritativeness than the official form. Any doubt arising about the correctness of its content should therefore be redeemed by comparing the content of the electronic format with that of the authoritative form, which remains the guarantee of the authorial intention.

4.3.2 Markup is an editorial process

Markup is the act of adding annotation and labels to the fragments of a document in order to allow computer-based processes to be carried out (from publication to print to storage to technical analysis, etc.).

The markup process is the process of actually adding these annotations and labels to the original text according to a specific syntax that is dictated by the computer environment (in the case of Akoma Ntoso, this means adding XML tags around the text fragments that have been identified and classified).

Of course this means adding to the original content, and in this sense, according to the definitions above, it has to be considered an editorial process and not an authorial process.

4.3.3 Naming is an editorial process

Akoma Ntoso defines a series of rules for giving a name (technically, a URI/IRI) to all the electronic versions of the legislative documents -the Akoma Ntoso naming convention. For these electronic versions to work correctly within the applications that make use of the Akoma Ntoso standards, it is necessary that the URIs/IRIs are correctly determined. These URIs/IRIs are derived in many ways from the “official” or the “most used” names of the documents, but do not necessarily look like them. Because they are defined by the creators of the XML representation, they are considered editorial in nature. Nonetheless, it is of the uttermost importance that they are created correctly and precisely according to the Akoma Ntoso Naming Convention or any other functionally equivalent naming convention.

4.3.4 Metadata items are editorial additions

Editors (i.e., the creators of the FRBR Manifestation) have two main tasks in the production process of Akoma Ntoso documents: on the one hand to identify and label (i.e., mark up) the fragments of the original content according to their role and structure, and on the other, to provide additional information about the document itself that is not contained in the official text as created by the original author.

Collectively, this additional information is called metadata. Since these are metadata elements, and since they are added at markup time, their specification is an editorial process, and not an authorial process, and come under the responsibility of the creator of the markup.

4.4 Descriptive markup and prescriptive markup

Another pillar of Akoma Ntoso is that it is both descriptive and prescriptive. By descriptive we mean a standard that accurately describes with tags the document’s various organizational functions (articles, chapters, sections, headers, etc.), allowing an expert to read the document under the guidance of the vocabulary used to enclose the text into sections.

A standard is descriptive when it uses a vocabulary of tags representing the domain where it should be applied. The tags are selected by domain experts, not by computer technicians, so that the tags enable the markup to convey the true semantic meaning they contain.

A standard is prescriptive when it defines the tags’ coercive behavioral rules, thus determining not only the vocabulary but also how it should be applied. In legal drafting, we usually deal with codes of rules that define behaviours and conventions for the correct formation of laws: in a prescriptive XML standard, these rules can be translated into technical delimitations included in the standard itself to facilitate compliance with the rules of legal drafting. For example, an XML document consisting of articles could be set-up in such a way that articles will always have a unique number; i.e. paragraphs are sequentially numbered and the structure is hierarchical. Otherwise, the XML document will not be standard-compliant; hence it will not be valid.

For example, a legislative official can open an XML document marked in Akoma Ntoso and, without knowing anything about XML, guess the function of each of the document’s parts that are referred to with tag names which matter to the expert and not to the computer technician. On the other hand, other standards have chosen to use technical terminology and vocabulary where the item is not enclosed within tags such as <article>, but are enclosed within more neutral terms such as <paragraph> or <block>.

Secondly, Akoma Ntoso uses its own schema to provide a set of rules for good regulations requiring a minimum set of quality links (e.g., numbering the articles). Thanks to this feature, tools can verify the correct composition of legislative text.

4.5 Content, Structure, Semantics, Presentation

Akoma Ntoso maintains four levels - clearly and strongly separated for the representation and description of legal documents:

• Content: what exactly is written in the document (e.g. the text);
• Structure: how the content is organized (e.g. articles, chapters, etc.);
• Semantics: the conceptual framework of knowledge needed to understand the document (e.g. for understanding what is the <def> you need to know what is a definition in the juridical domain and to combine the term to an ontological class);
• Presentation: the typographical choices to present a document on screen or on paper (e.g. right aligned, bold, italic).

Contrary to other XML-schemas, Akoma Ntoso separates the levels for maintaining the independence of the content from the semantic and from the presentation. In this way it is possible to semantically annotate the same content fragment several times without forcing the user to mark-up the document again. The same principle applies to the presentation: using the class attribute it is possible to define the semantic approach to the presentation, but the values of those parameters are defined externally to the XML file. This permits changing the layout several times without intervening on the physical XML document. Using XSLT and CSS it is possible to assign a proper presentation to a defined class. For example, if we have <title class="bigger right italic bold">, it is possible to define the main characteristics of presentation of the content <title>, but the specific values of these parameters are defined in a CSS (e.g. bigger means size 14, right means on right aligned, and so). When a media needs to have a different presentation only the CSS is properly adapted to the new needs.

4.6 Ability to evolve

A critical characteristic of a successful XML model is its ability to evolve over time. This “evolvability” has been a key concern in the creation of the Akoma Ntoso model. Thus, although the language is built to change over time, the language can be customized at will for local needs and purposes, and still be made compatible with the overall Akoma Ntoso infrastructure and the general language.

Furthermore, the language is built to withstand changes even regarding the number of actual functions provided: features such as the number and type of metadata values, or the automatic generation of amended text, or the activation of special analysis tools on the text that may require the language to evolve in time. In these cases, it can be guaranteed that existing documents already marked up according to initial versions of Akoma Ntoso will be either immediately compatible with the new schemas, or easily convertible to it via a single XSLT stylesheet that is provided.

4.7 Custom elements

Akoma Ntoso provides a flexible legal vocabulary for reflecting all the legal tradition requirements. The <heading> element in each hContainer part could be before or after <num> or <subheading>.

However in some legal tradition some elements are located in a precise order and some customization of the grammar is necessary. It is the case of Japan legislation or Scotland subsidiary regulation where the <heading> precede the number of the article:

(Definitions) 1 Article 2 (1) The term "Insurance Business" as used in this Act means the business of underwriting the risks listed in the items of Article 3, paragraph (4) or the items…

• 1 http://www.fsa.go.jp/common/law/ins01.pdf

For instance, the following example shows a Schematron2 fragment verifying that the <heading> appears before <num> in the <article> element:

<xsd:appinfo> <sch:pattern eId="article" xmlns:sch="http://purl.oclc.org/dsdl/schematron"> <sch:rule context="article"> <sch:assert test="[1][self::heading]">heading must be first child of article</sch:assert> <sch:report test="[2][self::num]">num must be second child of article</sch:report> </sch:rule> </sch:pattern> </xsd:appinfo>

Finally in Akoma Ntoso exists some elements that permits to include extra-elements:

• 1. in metadata block we have: proprietary (see also 5.10.13.1), presentation (see also 5.10.13.2), preservation (see also 5.9.1.1), and otherAnalysis (see also 5.10.9);
• 2. in the content block we have: foreign (see also 5.14).

• 2 http://www.schematron.com/