The language of business reporting in the Xbrl-xml era

Xml

XBRL (extensible Business Reporting Language, extensible Business Reporting Language) is an application of XML (Extensible Markup Language, extensible Markup Language) technology in the exchange of financial reporting information.

XBRL is advocated by the XBRL International Steering Committee and promoted its development. At present, there are more than 250 organizations in the world. The XBRL International Steering Committee is funded by the American Institute of Chartered Accountants, Deloitte, Ernst and KPMG, Microsoft, IBM, J.d.edwards, SAP, and other technical and financial companies. Since XBRL was born in 1998, XBRL has been developed rapidly, especially as the latest standard and technology of financial information processing, XBRL increased the transparency of corporate financial report disclosure, and greatly improved the efficiency and ability of financial report information processing. Take the financial industry as an example, at present, many international exchanges, accounting firms and financial services and information providers and other institutions have adopted or are prepared to adopt the standard and technology, such as the Tokyo Exchange's TD net system using XBRL technology to submit financial data, the Australian exchange is studying and preparing to use XBRL, Deutsche Bank of Germany used XBRL to process loan information and make its credit analysis process smoother. XBRL has also begun to be applied in China. Both the Shanghai Stock Exchange and the Shenzhen Stock Exchange have started research and experiment on XBRL technology. At present, dozens of listed companies have participated in the pilot work of using XBRL technology to submit the summary of the company's annual report. Of course, the current situation in China, XBRL application is still in the initial stage, there are a lot of problems. The following is an introduction to XBRL language.

　　XBRL Basics

XBRL originates from XML, so to understand XBRL technology, you first need to understand XML. For XML, there is no more introduction here, and readers who need it should refer to the relevant information themselves. The latest standard is XBRL Specification 2.1, which was passed shortly before the XBRL International Steering Committee. XBRL Information model is mainly composed of two parts, XBRL example (XBRL instance) and XBRL classification set (XBRL taxonomy). The XBRL instance contains the specific facts in the report (concrete facts), while the classification set is the definition of a specific fact, including its syntax, its form, and its attributes.

XML Schema and XLink are used in XBRL technology.

　　XBRL Instance

The XBRL instance encapsulates specific business facts. One of the most basic units is the fact (facts). For example, the sales in the last quarter ("Sales in the most recent quarter") is a fact. Simple facts without complex information structures are called entries (item). In order to express more complex information units, a set of related facts can be put together, expressed in relatively complex structures such as tuples (tuple). Tuples can contain either entries or other tuples. Entries correspond to business concepts, and the specific business meaning and definition description of these business concepts is done by the classification set (taxonomy set).

Specific examples can refer to some of Microsoft's XBRL financial statements, which can be downloaded from the Microsoft home page.

　　XBRL Classification Set

The specific facts in the report are represented by examples. But these specific facts should conform to what language specifications, their form and attributes, is defined by the classification set.

Different countries, different industries use different accounting standards. As an international standard, XBRL has to adapt to different situations, and the key is to establish a specific classification set that conforms to the accounting standards of various countries. This work is done by the XBRL organizations of each country to establish the XBRL classification set which conforms to the national accounting standards. The classification set first defines the financial statement elements that will be used in the financial statements, along with additional relevant information such as the correlation between the elements, the names and locations of each element in the report, and the accounting standards on which the financial statements are made.

The XBRL taxonomy describes the vocabulary (vocabulary) by defining the syntax of the lexical entries (term of vocabulary) by the XML schema (schema), and Xlink describes the metadata, which can be used to characterize the relationship between the item and the item. Each XBRL instance is associated to the corresponding classification (taxonomy) by referencing a schema (schema), while each pattern can refer to other patterns. A set of all schemas (collection) that are required to validate the syntax of a XBRL instance makes up part of the discovered set of categories (Discoverable taxonomy set (DTS)) of the instance. By referencing the XLink link set (linkbases), XBRL instances can also express more information about business concepts. XBRL uses xlink as a mechanism for defining the relationship between business concepts and corresponding metadata. In addition to the XBRL instance, the schema (schema) in DTS can also contain references to a set of links (linkbases). All of these schemas and linkbases together constitute what is called DTS, so that we can describe the business concepts of the entire report.

The XBRL classification set consists of multiple files. The primary file includes a taxonomy Schema (. xsd file), and calculates, defines, labels, displays, and references these five linkbase files. For all tuples, entries, and facts in a XBRL instance, define the name and data type (type) in the taxonomy set. The main data types used in taxonomy are the xbrli:datetimeitemtype representing the date, the xbrli:stringitemtype representing the string, the xbrli:moneytaryitemtype representing the amount, There are Decimalitemtype, Shareitemtype, Uriitemtype and so on. In principle, all elements should be included in either Xbrli:item or Xbrli:tuple (SubstitutionGroup), in other words, elements that are not in these two groups can in principle be written to an instance document (Instance Documents).

If you are defining a specific set of classifications for a particular industry or organization from scratch, you should define a schema that includes entries and tuples (items and tuples) for the business concept of that industry or institution. However, more often than not, the required industry classification is already in place, and it needs to be changed slightly to suit the needs of the user, such as adding some business concepts and then connecting them to the standard concepts and so on. In this case, the industry classification set pattern (Industry taxonomy schema) can be introduced first (import), and then some new schema components (schema components) are added to reflect the user's business concepts.

The taxonomy Schema (. xsd file) also defines a LINK element, Link:linkbaseref, whose properties define the roles of each linkbase and the files that are connected. In addition to the type attribute, the Arcrole property, the Actuate property, the Role property, the href attribute, and so on are included.

The label link (label links) and the reference link (reference links) associate Business concepts (business concepts) with metadata (metadata). For example, we know that a text string (the text strings) can be used to mark (label) or document the concepts in a report, while a tag link (label links) can relate the text string to the corresponding concept, such as You can use a label link to associate the text string "Last quarter gross income" (revenues in most recent quarter) with the "REVENUEMRQ" entry defined in the taxonomy. Single-stature business concepts can be labeled with multiple tags in many languages. The creator of the specific XBRL instance (author) determines the trade-offs of each label for the instance.

Figure 1 illustrates three things to do to define an associated link (label link) associated with a business concept and label: Mark the business concept with a LABEL element, define a locator (locator) for the concept with the LOC element, and use a Labelarc An element defines an ' arc ' (ARC) that is associated with this business concept on the specified label. A reference link is also a similar way to link a reference to the authoritative literature in the business domain (domain), using a mechanism similar to a tag link: Defining a reference link and a locator that locates business concepts and related document references, and finally builds a REFERENCEARC element Associates the previously defined locator with the reference.

Tabbed links and reference links are used to correlate business concepts and metadata. The other three types of links are precisely the opposite, and are used to establish the association between different business concepts. For example, a computational link is used to define the computational relationships of the relevant business concepts. For example, the concept of "after-tax profit" ("Profitaftertax") is derived from the concept of "pre-tax profit" ("Profitbeforetax") minus the term "income tax" ("Taxpaid").

Profitaftertax = weight (1) * profitbeforetax + weight ( -1) *taxpaid

Figure 2 is an example of how to relate these three business concepts.

The definition link describes the relationships among several business concepts, such as generics and a generalization-specialization relationships, such as: "PostalCode" is a generalization of "ZipCode". Show links (presentation links), as the name implies, define the relationship between different business concepts from the perspective of presentation (presentation perspective), for example: In a presentation report, a parent-child relationship should appear as "Sales" and " Printer Sales "Printersales.

　　XBRL Processors

XBRL processors (XBRL processors) not only validate the syntax of the instance document (instance documents) by classification mode (taxonomy schemas), but also by category link set (taxonomy linkbases) To verify the semantics of the instance. When processing an instance document, the processor should be able to extract the metadata needed for the application based on the classification of the instance. This process is similar to XML Schema (schema) validation (validate) to generate PSVI. XBRL validation should take a taxonomy (taxonomy) to validate the instance and send the metadata (metadata) derived from the validation process to the application (application), such as the processor should be able to detect a metadata label for an item (item) in the report instance. In addition to validation, XBRL tools (XBRL tool) include: Category editor (Taxonomy editors), XBRL Instance generation (XBRL instance creation), XBRL repository (XBRL repositories) and query reporting engine (queries and reporting engines), and so on.

　　The application of XBRL

XBRL provides us with a framework for business reporting. Business statements cover a wide range, from the standardization of public institutions (regulatory compliance) to the enterprise Intelligent (Business Intelligence) are applied. As shown in Figure 4, XBRL can create value in different links in the reporting chain. The standardization of business information processes on the reporting chain provides opportunities for many applications. Packaged (canned) applications can be used to write (composition), analyze (analysis), display (presentation) information.

　　Summarize

At present, there are many kinds of reporting languages based on XML technology, and XBRL is becoming the mainstream of commercial report language. Through XML technology, XBRL provides a truly extensible framework (the framework) that organizations can use to report important business information. In the business reporting chain, XBRL provides an advantage for each link, from information preparation (preparation), analysis, until application (consumption). After two rounds of revisions-public review and implementation feedback (implementer feedback), XBRL has matured and become widely used.