XBRL Specifications

ActivityMain PointsComment XBRL Strategic Initiatives6 initiatives Participation needed Volunteer now!!! Highly participative discussions, wide interest Inline XBRLBrowser html embeds XBRLWide interest Rendering Linkbase, rendering user projects New specs in development, custom pre-spec use Opposite of inline XBRL Formula features, formula user projects New CR modules, Project reports (BE, CN, ES, JA) VersioningStarting IFRS useUS-GAAP present Semantics and DatabasesHigh interest area International new facesChina participates(I enjoyed meeting China formula users)

Abstract Modeling is starting now

Oct 09 What are the long term strategic goals for XBRL? XSB provided suggestions and debated them on weekly XSB calls Nov 09 Face-to-face Meeting in Brussels XSB summarized the suggestions into 3 main goals Feb 10 Publication of Discussion Document XSB published their recommendations Mar 10 Feedback window was opened XII reached out to the community to actively gather feedback Apr 10 Feedback window was closed XSB analyzed and summarized the feedback that was received May 10 Face-to-face meeting in Brussels XSB defined the framework for the six initiatives Aug 10 What exactly are the initiatives? XSB defined each initiative in much greater detail Sep 10 Internal Review of Initiatives XSB circulated the document internally for feedback Oct 10 Preserve. Promote. Participate. XSB released the document on October 6, 2010 Identifying the Goals Analyzing the Market Feedback Identifying the Initiatives

InitiativePrimary BenefitDiscussion Document Goals Make XBRL easier for developers Improve XBRL comparability Make XBRL data easier to consume 1Create an abstract model An abstract model provides a conceptual framework for understanding XBRL and gives developers a strong foundation for their implementations. Yes 2Produce training materials High-quality training materials lend support to developers and those new to XBRL. Yes 3Define standard API signatures API signatures assist developers with their implementation of XBRL solutions. Yes 4Reorganise existing specification A reorganisation of the XBRL specification will make the specification easier to understand. Yes 5Enhance data comparability Data comparability widens the applicability of XBRL data across project and international boundaries. Yes 6Develop application profiles Application profiles reduce the scope of XBRL implementations by breaking up the XBRL specification into components. Yes

XBRL framework for modelling various domains a meta-model, not a model taxonomy models a domain UML framework for modelling software infrastructures XIIs relationship to XBRL is as OMGs to UML OMG Meta-Object Facility (MOF) models UML XSB wishes the same for XBRL

Primary model captures XBRL semantics: Core spec and Dimensions are mandatory Model should be void of syntactic details Should be portable across technologies, like SQL and OOP Secondary model binds primary model to XML This bridges the gap from model to current spec

Domain (taxonomy) modelling not an objective Parallel modelling efforts will be considered with BPB involvement Formulas, Versioning, Rendering XSB will consider

Focused group to delivery over 6-8 weeks Easier to get short commitments Deliverable is of high value to the community, so sooner the better Abstract model is a prerequisite to several other initiatives

Format will be: 2 face-to-face meetings over a 4-6 week period 1 st meeting scheduled for Boston on Jan Conference calls daily, or every other day, in between the face-to-face meetings Dedicated project manager to drive the process Preparations and materials will be assigned prior to face-to-face meetings Face-to-face meetings will be divided into minute segments, with firm deliverables defined

Use cases What processes is XBRL intended to work in? Modelling the core specification Facts, units, periods, entities, labels, heirarchy, etc. Modelling the dimensions specification Primary item, domain, dimension, member, etc. Binding the models to the current XML syntax How to bridge the gap from abstract model to the current XML specs

Review process will have to be open and as inclusive as possible: Review period will have to demonstrate community coverage Broad outreach (webcasts, conferences, etc.) will be a necessity Specific, targeted outreach will also need to be conducted where necessary

InitiativeResource Type 1Create an abstract model Program/Project Managers UML Modelers Software Architects XBRL Practioners Business Reporting Domain Experts 2Produce training materials Publications Coordinator Software Developers Tech Writers Academics 3Define standard API signatures Program/Project Manager Software Architects XBRL Practioners 4Reorganise existing specification Publications Coordinator Tech Writers Software Developers/Testers XBRL experts 5Enhance data comparability Domain Experts Financial Professionals XBRL Taxonomists 6Develop application profiles XBRL Technologists XBRL Practioners

InitiativeVolunteers from 1Create an abstract model US,CA, DE, ES, FR, PL, CN, AU 2Produce training materials US, CA, DE, PL, AU 3Define standard API signatures US, CH, DE, PL, AU 4Reorganise existing specification US, CA, FR, PL, CN 5Enhance data comparability US, CA, DE, FR 6Develop application profiles US, CA, AU

atives

NumberDescriptionTestcasesDate 368 Calculation linkbase checking currently infers precision, any value 0 implies precision 0, which is then always inconconsistent. Proposed spec change to infer decimals. (Testing SEC filings to determine impact of this change) Different dimensions share domain members SharedDo main Units with unbound xbrli prefix namespace for pure and shares 304-v12a, v15a s-equals, v-equals, functions: id- attribute sensitivity, drop Xpath 2 eq use, assigned to Formula WG ,

Items reported in precision 0 always invalid Items reported in decimals with value 0 Infers precision 0 Technically always invalid Changing to infer decimals Known financial reporting is in decimals Eliminates issues with 0-valued items, edge cases Test suite has been updated

The total of a binding calculation is defined to be the sum of the rounded values of the contributing numeric items in the binding, each multiplied by the value of the weight attribute on the item's associated calculationArc. This multiplication takes place after any necessary rounding is performed. The rounded value of a numeric item is the result of rounding the value of the numeric item to its precision decimals or inferred precision decimals (see Section and ). A binding calculation is defined to be consistent if the rounded value of the summation item is equal to the total rounded to the precision decimals or inferred precision decimals of the summation item. (If any item of the calculation has a precision attribute value 0 then the binding calculation is deemed to be inconsistent.)

Extension module FeatureStatus AspectCover FiltersCR Concept Relation FilterCR Custom Function ImplementationCR Generic MessagesCR Validation MessagesCR Instances (multi-instance, and variable-set chaining)CR Tuple outputOn-hold, PWD

1. CEBS member formula issues 2. Dimensional aspects of fallback values 3. Debugging capabilities 3. Rendering of new FINREP and COREP 4. Rendering ideas and formulae editoring

26 CR PWD status

Unchanged from business perspective Concept and dimensions aspects changed A, B, C and X are primary items D is a dimension d1, d2 and d3 are domain members Version 1 Version 2 A X ( D = d1 ) B X ( D = d2 ) C X ( D = d3 )

Predicate expression (Xpath 2) Identifies fact by relative location 500 Cash 500../gl:xbrlinfo/gl:summaryReportingElement eq Cash < verdim:concept href=…fr.xsd#cash"/>

../gl:xbrlinfo/gl:summaryReportingElement eq Cash

balance-brought-forward 5100 Supplies account D beginning_balance … standard … …../../gl:qualifierEntry eq balance-brought-forward and../gl:account/gl:accountMainID eq 5100 and../gl:xbrlinfo/gl:xbrlInclude eq beginning_balance

Inline XBRL - REC embeds XBRL fragments into an HTML document Requirements for linkbase – Drafting Linkbase spec – Modularized & new specs

inline XBRL Requirements Linkbase Future

Separation of Specification Improve understanding and extendsibility New Models Inheritance mechanism Injection model New Specifications Identification and Ordering Context Grouping Generic Preferred Label Manifest 42

Public Taxonomy author Identification and ordering Taxonomy Detailed rendering structure Extension Taxonomy Extended table of contents User specific Rendering Structure Instance / Inline XBRL Taxonomy 2.1 spec dimensions Data preparer Put as much as information here Less information here 43

Public Taxonomy author Identification and ordering Taxonomy Detailed rendering structure Create report Taxonomy 2.1 spec dimensions Report from Taxonomy information Ex: Manual Guidelines etc… 44

Current spec provides document model to rendering modeling: document layered model. Uses linkbase mechanism to represent entire reporting document model. Additional model introduced: Injection model. Uses existing rendering format (such as XHTML ) then inject rendering definition into the format. 45

Document Components Containers relationships XLink resource XLink arc Legend Component Container Document 46

Rendering Components Legend XHTML etc.. image Tables, lists, paragraphs are injected into exiting format (XHTML as such) 47

Heavy use of definition link by XBRL Dimensions, same mechanism such as preferred label in presentation link solves rendering issues. Generic preferred label is a new mechanism to preferredLabel attribute on definition/calculation/generic arcs. 48

Specification are separated into: Semantics Semantic modeling definition regarding the specifications Structure One basic model to represent reporting document structure Axes Set of axis definitions which could be used to rendering component Components Set of component definitions that could be used in the report structure Values Set of value definitions that could be used to represent cell pattern. Rules Set of rule definitions that could be used from any part of rendering information. Identification and Ordering (Table of Contents) Mechanism to identify statements (build table of contents explicitly) Context Grouping Mechanism to create group of contexts Generic Preferred Label Mechanism to specify preferred label into linkbases other than presentation linkbase. Manifest Mechanism to package XBRL doucments 49