Printer Friendly

On a logical structure for the authoritative accounting literature: a discussion of the FASB's codification structure.

INTRODUCTION

As early as 1965, Goldberg asserted: "it is scarcely an exaggeration to say that the problem of communication is the axial problem in accounting" (Goldberg 1965). Despite that assertion, research into how formal accounting communications, such as the financial accounting standards (FASs), can and should be effectively articulated is conspicuously absent. This paper aims to address the significance of document structure in the drafting of the FASs. Improved document structure promises to more effectively communicate the information contained within documents and to enhance the capabilities and performance of a document retrieval system (Sumita et al. 1993)

FASs, as historically drafted, are organized across standards by standard setter and type of standard. However, within standards, the text lacks a consistent and sound structural arrangement, which impedes digital storage and retrieval methods and is not conducive to accounting research by most users (Fisher 2004). For example, markup languages such as the eXtensible Mark-up Language (XML), which represent the structure and content of documents, are predicated upon a hierarchical document structure that includes pre-defined sequencing and nesting. Further, despite the fact that the American Institute of Certified Public Accountants' (AICPA) special Task Force on standards overload recommended the development of systems to "facilitate access to the professional literature and improve the understandability of that literature" (Burke 1997) (emphasis added) little has been done to apply digital methods and technologies to the structure of FASs and the development of digital libraries of FASs.

The Financial Accounting standards Board (FASB) Codification and Retrieval Project has been engaged in a comprehensive effort to codify the authoritative accounting literature (AAL). The literature has been reorganized to present it in topical order, as opposed to the current presentation of the AAL by standard number where standards are classified under one of four possible levels of the Generally Accepted Accounting Principles Hierarchy as originally defined in statement on Auditing Standard number 69. The FASB Accounting Standards Codification[TM] was launched online on January 15, 2008 beginning a one-year constituent verification period during which users are provided free access to the codification. Free access is intended to enable the FASB to obtain feedback on the codification prior to its final acceptance and approval.

While the FASB has not publicly announced the technical underpinnings of its retrieval system, we suggest that as the codification project progresses, the FASB should employ an XML-based logical structural model. This paper proposes such an XML structure for the FASs.. The role of structure in facilitating digital communication and manipulation of textual information is highlighted throughout.

The remainder of this paper is organized as follows. Prior research and related work are examined. The results, to date, of the FASB Codification Project's efforts to structure the AAL are discussed. suggestions for additional consideration of document structure are offered along with a logical model for FAs structure. Finally, concluding remarks are provided and future research opportunities are mentioned.

PRIOR RESEARCH AND RELATED WORK

Written Communication

"The written document whether delivered in a printed or online medium is often essential to a product or service's success and ultimately to an organization's success" (Smart et al. 2001)

Increasingly, researchers indicate professional communicators should focus on issues of information design citing the design of information as critical to the usefulness and usability of information (Redish 2000; Schriver 1997; Smart et al.). Eunson (2005) states that communicators must focus both on writing and structuring the content of documents as well as on their appearance and navigability. Baker (2001), in turn, suggests that document headings should visually expose the information structure since this will increase access to the information contained in the document. Campbell (2006) extends the significance of document design even further by suggesting that words alone may be an insufficient means of communicating in the 21st century and calls for the incorporation of visual aspects into documents for more effective communication. Good design can not only facilitate document navigation, but can increase document readability and enhance document comprehensibility (Burnett 2005).

Belkaoui (1980) applies sociolinguistics in determining that professional producers and users of written accounting information differ in their understanding of selected accounting concepts. Specifically, Belkaoui (1980) finds that academicians use a formal language, professional accountants use a public language, while students use a mix of both. since all three of these groups are users of FASs, this suggests that there may be an inherent difficulty in writing the AAL in a manner that is equally understandable to all. It also suggests that the FASB might be well served by determining not only who the primary (intended) user is of the AAL in its original and unaltered form, but who actually does use the AAL and for what types of tasks. Flower and Hayes (1980) suggest that the effective use of rhetorical tools presupposes that distinguishing characteristics of the audience have been considered. The smart et al. (2001) study of user preferences for online versus printed documentation finds that there is considerable variability across tasks, concluding that organizations need to carefully consider their primary audience(s) and allocate resources to the types of documentation most likely to promote user satisfaction and productivity. While balancing the diverse interests of multiple stakeholders is not easy, Harrison and Freeman (1998) suggest it is possible and Seeger and Ulmer (2003) add that doing so may positively impact long-term viability of the related entity.

Couture (1992) identifies technical/professional writing as a category of professional writing distinguished by the intent to accommodate audience needs through compliance with professional readability standards. Riahi-Belkaoui (1995) details the results of 42 studies conducted from 1952 to 1993 on the readability of accounting communications, including accounting textbooks, excerpts from the Internal Revenue Code, financial statements and financial statement footnotes along with eight studies focused on their understandability. The studies overwhelmingly conclude that readability is difficult and that understandability by certified professional preparers (e.g., certified public accountants), certified professional non-preparers (e.g., chartered financial analysts), and college-educated, non-professional users is problematic.

Finally, Haried (1972) notes that there has been a "conspicuous lack of empirical support" for the semantic problems in external accounting communications. He identifies the designation of technical meanings to terms that express different meanings in common use or in other disciplines and the lack of standardization of terms used in financial reports as the primary problems noted.

XML and Document Structure

XML was developed by the World Wide Web Consortium (W3C), with the support of the contributions and efforts of many in industry and academia, in an effort to overcome limitations in both the standardized Generalized Mark-up Language (SGML) and Hyper Text Mark-up Language (HTML) (Abiteboul et al. 2000). XML permits users to define domain-specific (mark-up) tags. Thus, developers can create tags that are more intuitive to users in specific disciplines. The use of mark-up languages is cited as a productive means of formally describing document structure (Baeza-Yates and Roberio-Neto 1999). The addition of structure can add important value to documents, referred to by Baeza-Yates and Roberio-Neto (1999) as making text "smart". In accounting, Fisher (2004) calls for the use of formal structural standards in all FASs to support digital representation, storage and retrieval.

In the computer science domain, XML has gained acceptance as the preferred solution for document management and document versioning when documents utilize a web-based environment (Chien et al. 2001). Marian, et al., (2001) state that XML is becoming the standard for semi-structured data exchange over the Internet. The increasingly wide acceptance of XML has given rise to derivative languages such as the eXtensible Business Reporting Language (XBRL), developed to improve business and financial reporting at the account level (Bovee et al. 2005). While XBRL is being used for the definition of financial statements and other financial reports, XML is better suited for the definition of financial accounting standards, which are comprised primarily of text data (Fisher 2004). XML also has the ability to model unstructured data (Banerjee et. al. 2000). XML was ultimately chosen as the language for document (structure) definition here in view of its natural extensibility through the support for user-defined tags as well as the increased prominence of its use in both document version management (Wong and Lam 2002) (Chien et al. 2006) and the financial accounting domain (Geerts and White 2004) (Geerts 2004).

XML and Information Retrieval

Information retrieval (IR) is often confused with data retrieval. Data retrieval primarily involves determining which documents of a collection contain the keywords in the user query, which is usually not enough to satisfy the user's information need (Baeza-Yates and Riberio-Neto 1999). Instead, what the user of an IR system really wants is to retrieve information about a subject rather than to retrieve data that satisfies a given query. In order to better fulfill an information need, the IR system must somehow 'interpret' the contents of the information items (documents) in a collection and rank them according to a degree of relevance to the user query (Baeza Yates and Riberio-Net 1999). This 'interpretation' of document content involves extracting syntactic and semantic information from document text and using this information to match the user information need. The difficulty is not only knowing how to extract this information but also knowing how to use it to decide relevance. Thus, the notion of relevance is at the center of information retrieval. Relevance rankings are most often determined by keyword(s) counts. Keywords, in corpus linguistics, are words whose frequency of occurrence in a specific text exceeds that of chance alone. The implication is that such words are indicative of the meaning or essence of the text. The problem is that they are often not good indicators.

Keywords, in the context of Internet search, also refer to the user supplied term (s) that defines documents of interest. The key for an effective IR system is to facilitate the likelihood that the user supplied keywords and the system-generated keywords correspond to relevant documents. The use of mark-up (e.g., XML) encodes document content to better represent its meaning. Such encoding enables machines to "understand" and thereby manipulate the content (Shah et al. 2002). XML tags, specifically designed to capture meaning, permit the usage of word/phrase counts in combination with mark-up to facilitate the interpretation of text and better determine relevance (shah et al. 2002).

Figure 1 provides an illustration of an XML tag, that provides both the structural definition of an element (subtopic.section) in combination with the semantic definition of attributes of the element (Keyword-one, Keyword-two), for a fragment of FAs 16 on prior period adjustments. The 'keywords' interim reporting and prior period adjustment would not be identified by simple keyword searches because these exact sequences of terms do not occur in the passage.
Figure 1: Illustration of the Use of an XML Tag to Indicate Structure
and Meaning in a Fragment of FAS 16: Prior Period Adjustments

<subtopic.section Keyword-one="interim reporting" Keyword-two="prior
period adjustments">

If an item of profit or loss occurs in other than the first interim
period of the enterprise's fiscal year and all or a part of the item
of profit or loss is an adjustment related to prior interim periods
of the current fiscal year, as defined in paragraph 13 above, the item
shall be reported as follows:

a. The portion of the item that is directly related to business
activities of the enterprise during the current interim period,
if any, shall be included in the determination of net income for
that period.

b. Prior interim periods of the current fiscal year shall be restated
to include the portion of the item that is directly related to business
activities of the enterprise during each prior interim period in the
determination of net income for that period.

c. The portion of the item that is directly related to business
activities of the enterprise during prior fiscal years, if any, shall
be included in the determination of net income of the first interim
period of the current fiscal year. </subtopic.section>

The 'subtopic.section' designation indicates a structural element.
The 'Keyword-one' and 'Keyword-two' designations provide possible
search terms indicative of the meaning or application of the section,
which would not be inferred through exact keyword matches to the text.


The most frequently occurring term in Figure 1 is the word 'period', which occurs 8 times, followed by the words 'interim' and 'item' occurring 6 times each. The term 'period' is too vague to serve as an effective search term (i.e., simply imagine how many times the term occurs in the accounting literature). The term 'item' does not convey the meaning of the FAS 16 fragment at all. While the exact phrase "interim period" occurs 6 times, the phrase "prior period" does not appear at all. Supplementing the literally occurring terms with keywords that can better approximate the meaning and/or application of the passage can most certainly increase the effectiveness of information retrieval.

More specific tag definition facilitates the retrieval of more precise information items (Norvag 2002) and, in fact, intelligent machine manipulation beyond simple (keyword) searches is often dependent upon such mark-up detail (Lauritsen 1993). In accounting, Debreceny and Gray (2001) note that tagged text facilitates automated information location and retrieval and enables the problems of resource discovery and attribute recognition to be (at least partially) overcome.

The current Financial Accounting and Reporting system (FARS) database, a digital and searchable representation of the AAL that is broadly available, supports only keyword searching. While keyword searching is popular and prevalent because of its simplicity, it is also widely acknowledged as imprecise (Garnsey et al. 2006, Fisher 2004, Go et al. 2003) and unreliable (Kon and M. Hoey 2005). A keyword query is an "all or nothing" proposition. A document either contains the keyword or it does not. In accounting there are often multiple terms (words) that denote similar concepts (e.g., company, enterprise) and searchers are frequently unfamiliar with the vocabulary (Garnsey 2006). smart, et al. (2001) identified the lack of knowledge of appropriate keywords as one of the most significant factors that cause users to prefer printed documentation over online documentation. Keywords can also be inadequate in identifying the concepts represented in documents (Kon and M. Hoey 2005). Finally, keyword searching results in the retrieval of an entire document when possibly only a portion of the document is relevant to the query. Mark-up languages such as XML can enable a more precise answer to a query (search) by using the tagged document structure to beak up the document and return the part of the document that is most relevant to the query (Moens 2005).

The term precision is used specifically in information and library science to denote the fraction of retrieved documents (or document fragments) relevant to the user and is one of the traditional criteria employed to evaluate retrieval performance (Baeza Yates and Riberio-Neto 1999). Sumita et al. (1993) assert that retrieval precision is increased by the use of document structure as an aid in the location of relevant document fragments or terms. Experiments conducted by Wilkinson (1994) and shah et al. (2002) confirm that knowledge of the document structure improves retrieval performance by increasing retrieval precision, while other experiments support the ability of structure to enable more powerful query facilities (Lee et al. 1996) and to provide semantic clues that can facilitate automatic classification of document text (Yi and Sundaresan 2000). Larsen et al. (2006) demonstrates that searchers find the (XML) structural breakdown useful when browsing documents retrieved from a digital library.

The benefits of the formal incorporation of document structure appear evident. This provides compelling motivation to explicitly address the issues involved in structuring the AAL. The following section provides an overview of the FASB's efforts to deal with a document structure for FASs.

THE FASB CODIFICATION PROJECT

Throughout the codification project, the FASB worked to reorganize the accounting literature so that it will consist of AAL only, presented by topics and subtopics (McEwen et al. 2006). Document structure has been addressed in this effort and is evidenced by the development of both a topic and subtopic list and a standard section structure for each subtopic (FASB 2007). The AAL was divided into four general topic areas: principles and objectives, overall presentation and display matters, transactional guidance, and industry specific standards. some topics may be further subdivided into subtopics. For example, McEwen et al. (2006) describes the topic of debt as being further divided into: overall guidance, convertible debt, debt issued with detachable equity instruments, participating mortgage loans and product financing arrangements. Each of these subtopics is organized according to a standard section structure sequence of (up to) sixteen sections. Figure 2 illustrates the codification structure for the topic of debt.

Amendments to the codification will directly replace the existing verbiage. Guidance that applies to more than one of the four broad topic areas is categorized either under industry specific standards (when applicable) or broad transactional categories and then linked to and from other more detailed and related topics (McEwen et al. 2006).
Figure 2: Codification Structure for Debt

The Codification Structure for Debt

I. Principles and Objectives
II. Overall Presentation and Display Matters
III. Transactional Guidance
 Debt
 Status
 Overview and Background
 Objectives
 Scope and Scope Extensions
 Topical Definitions/Glossary
 XBRL Definitions
 Recognition
 Initial Measurement
 Subsequent Measurement
 De-recognition
 Presentation Matters
 Disclosure
 Implementation Guidance and Illustrations
 Relationships
 Transition and Open Effective Date Information
 Links to Grandfathered Material

 Copyright [c] FASB

IV. Industry Specific Standards


A DISCUSSION OF THE CODIFICATION STRUCTURE

The codification project is the first undertaking by the FASB that explicitly addresses the importance of a uniform and consistent document structure for the AAL. The effort is both significant and commendable. As the codification continues, we suggest that a review of the different methods by which the text will be accessed will help determine if the codification is likely to be effective in communicating and facilitating access to the AAL.

Optimistically, the development and incorporation of a topical arrangement with corresponding subtopics provides the beginnings of a hierarchical structure. Routen and Bench-Capon (1991) argue that hierarchical formalizations of text are preferable for the creation of knowledge-based systems as they are said to better disambiguate text and to allow for the incorporation of meta-level information. The formal specification of prescribed ordered sequences of topics, subtopics and sections provides the consistency in structure that can facilitate the development and implementation of indexing and searching procedures (Fisher 2004). A consistent hierarchical structure is more compatible with the development of XML document type definitions (DTD) and schemas.

The codification can be searched using simple keywords or by accessing an advanced search feature. The advanced search option allows users to search by: 1) codification topic, subtopic or section number, 2) keywords in the general text, 3) keywords in titles or headings only, or to 4) restrict the search to specific areas of the codification. Any or all of the advanced search features can be combined to refine a specific search. The advanced search offers significant improvement, in terms of search specificity, over simple keyword searches. Figure 3 provides a screen shot of the advanced search feature in the FASB Accounting standards Codification[TM]. The codification research system also offers users a navigation bar on the left side of the screen which presents the areas of the codification and can be expanded, by placing the cursor over any specific area, to reveal the topics within. The navigation bar clearly is an attempt to more effectively communicate the codification structure by allowing the user to visualize the structure and thereby to aid access to the AAL. Finally, the codification research system also allows the user to join multiple sections from different topics or subtopics into a single (online) document.

The codification project is also focused, to the extent possible, on the elimination of redundant information and is attempting to ensure that content is only presented at one location in the AAL. The elimination of redundant content may certainly increase both the readability of the AAL and the precision of searches performed on the AAL (Fisher 2004). In addition, presenting information only once will facilitate maintenance of the knowledge base and reduce the likelihood of errors in the update process (Fisher 2004). When amendments to the AAL are necessary, if the information has been presented in only one location, then only one location will need to be updated, thereby eliminating the possibility of inconsistencies in the database.

[FIGURE 3 OMITTED]

Changes in the AAL will be effected through amending the codified text. Prior versions of the codification along with all supporting documentation will be maintained. Document maintenance is an important consideration for the FASB as the codification task continues. To simply publish an amendment separately (for example on the FASB's website) will not sufficiently provide for the determination of authoritative guidance as of a previous point in time. To do so would require professionals to keep a 'library' of amendments; selected amendments could then be 'subtracted' conceptually from current guidance to arrive at the principles as of an earlier point in time. This could be a very cumbersome and certainly an inefficient task over time. Alternatively, one could keep all versions of the AAL over time, also highly inefficient.

Access to prior versions of the AAL is an important issue. The application of accounting principles occurs as of a specific point in time. Questions regarding the proper application of accounting principles often arise many years subsequent to the application date. For example, while the Securities and Exchange Commission investigation of Enron began in 2001, litigation of the case was still ongoing as of 2006. The events in the case pre-date both the investigation and the litigation requiring the ability to review the AAL applicable as of the prior points in time. The probable migration from U.S. GAAP to International Financial Reporting Standards (IFRS) in the very near future presents additional concerns regarding the maintenance of a digital library of the AAL that provides temporal access.

We anticipate that the FASB will consider the design of and provision for an automated path through versions of the literature over time. Generating document versions and creating linkages between respective versions are nontrivial tasks and the subject of a sizable body of research in computer science. The design of systems to support the temporal generation of document versions requires that consideration be given to such things as: (1) the definition of a page or document, (2) the type of data comprising the document (e.g., text, numerical, or multimedia), (3) the degree of structure in the dataset (e.g., highly structured, semi-structured, or unstructured), and (4) the level of granularity at which document changes can occur (e.g., entire documents, pages, sections, paragraphs, sentences or words) (Chien et al. 2001) (Salimen and Tompa 2001).

We anticipate that the FASB will disclose, to the extent possible, its technological specifications related to the codified text. Such technological specifications may be considered proprietary by either the FASB, the development consultants, or both. However, the technological specifications have considerable impact upon whether the objective of making the AAL significantly more accessible is achieved. For example, will keyword searching be augmented through the use of semantic mark-up and if so, how?

Finally we suggest that the development of an explicit vocabulary in the accounting domain could enhance access to the documents comprising the AAL. Bench-Capon and Visser (1997) state the advantages of "explicit specifications of domain conceptualizations-now popularly called ontologies-in legal information systems ". They argue that knowledge systems are committed to particular conceptualizations although they are far too often left implicit. Implicit conceptualizations impede knowledge sharing.

Explicitly defined ontologies facilitate a common understanding of represented knowledge including the objects, concepts and the relationships that exist between them, within a particular domain or sub-domain. Such explicit conceptualizations also provide a defined framework for knowledge acquisition and make knowledge reuse possible. An ontology can be a reusable component in the design of new applications. Finally, Bench-Capon and Visser (1997) argue for the creation of libraries of dedicated ontologies, competing as well as supporting, that differ in scope, purpose, and level of abstraction. such libraries are seen as providing a valuable resource for the integration of diverse research (Bench-Capon and visser, 1997). Additional comments regarding the technical representation of structure are offered in the section that follows.

A PROPOSED STRUCTURE

Figure 4 provides a partial logical view of a proposed structure for the AAL. The AAL is defined as consisting of a series of topics. Each topic has a title and is further comprised of subtopics. Each subtopic has a title and consists of sections. Topics are assigned an attribute 'number' to capture and preserve the location and sequencing of the XML elements. Such information has been shown to be critical to the update of text when the literature is amended (Chien et al. 2001 and 2006) (Fisher 2003). Figure 5 continues the logical view of the proposed structure.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

Figure 4 illustrates that subtopics are divided into sections, which in turn can be divided into subsections with titles. The explicit representation of document structure has been shown to be useful to searchers (Larsen et al. 2006). Larsen et al. (2006) found that when searchers were given the choice to explore either full documents or elements of the documents in a retrieval set they most often chose the document elements.

Each section is also shown to possess the previously mentioned attribute of 'number' along with the attribute 'keyword-one'. The attribute 'keyword-one' is an example of the use of semantic mark-up to augment the structural mark-up. This tag allows the assignment of keywords to best represent the meaning of the section's content. The model can incorporate the addition of as many keyword (e.g. 'keyword-two') attributes as is considered useful to the best representation of content meaning. Keyword attributes can be incorporated into the model at deeper levels in the structural hierarchy (e.g subsections) to enable the more specific representation of content meaning. Thus, searches can now be performed on pre-assigned keywords rather than the literal terms in the written passages. This assists in overcoming the problem of the inconsistent or ambiguous use of accounting terminology (Gangolly et al. 1991), which can cause the results of searches to suffer from poor precision (the inclusion of irrelevant documents in the search results) and low recall (the failure to include relevant documents in the search results). In addition, relevance ranking functions can be augmented beyond keyword and phrase counts to include the differential weighting of terms depending upon the structural elements in which they occur (Moens 2005).

Finally, simply because a user-specified search term is not specifically observed in an XML element does not necessarily mean that the XML element is not relevant to the user's search (the accounting vocabulary contains many synonyms). XML document structure can be used in information retrieval to specify that XML elements whose parent (hierarchical ancestor) elements contain a specific search term are preferable over elements whose parent elements do not contain the search term (Moens 2005). such specifications can increase the level of recall in information retrieval. Recall refers to the number of relevant documents (or elements there of) retrieved from the set of all actually relevant documents.

Additional semantic mark-up can be incorporated into the model as well and is recommended here. However, the addition of such mark-up should only include information that is useful to users (searchers) of the AAL. This requires that studies be conducted to determine what types of information users will, in fact, find helpful.

Relatively little research has addressed the question as to how AAL searchers generate their queries. Query formulation (also referred to as generating retrieval cues) is critical to obtaining relevant results (Garnsey et al. 2006). Garnsey et al. (2006) found that providing students with a list of potentially relevant terms taken directly from the FARS database increases their ability to locate the answer to a research question. Further, in a term association test given to seven accounting professors, Garnsey and Hotaling (2005) found that for 13 of 30 randomly selected accounting terms, at least one of the ten professors could not generate any related terms. However, they also found that the number of terms with no response declined significantly when the subjects were provided with a list of potentially related terms (Garnsey and Hotaling 2005). This suggests that it would be useful for the structure of the IR system for the AAL to include semantic mark-up to facilitate user generation of search terms.

Greater constraint on structure could be provided in the model by defining the 'title' for each division of the AAL not as an element but rather as an attribute of the element topic, subtopic and section. That would allow for the enumeration of the specific FASB defined topics, subtopics and sections to be the only titles 'permitted' in the AAL. The choice between element or attribute representation represents a trade-off between flexibility and constraint. The choice should be made based upon whether flexibility in the model is considered beneficial. That is, representing titles as elements allows for the titles to change without impacting (i.e. requiring changes in) the model. The decision to use elements here was predicated upon the acknowledged failure of enumerated schemes' ability to service change (e.g., alterations, additions, deletions of titles) (Svenonious 1992). Such changes in systems are typically expensive.

We propose a document type definition (DTD) for the representation of the AAL in Figure 5.
Figure 6: A Proposed XML DTD for the Authoritative Literature

<?xml version="1.0" encoding="UTF-8"?>
<!ELEMENT AuthoritativeAccountingLiterature ( topic+)>
<!ELEMENT topic ( topic.title, topic.subtopic+)>
<!ATTLIST topic number CDATA #REQUIRED>
<!ELEMENT topic.title (#PCDATA)>
<!ELEMENT topic.subtopic (subtopic.title, subtopic.section+)>
<!ATTLIST subtopic number CDATA #REQUIRED>
<!ELEMENT subtopic.title (#PCDATA)>
<!ELEMENT subtopic.section (section.title, section.subsection+)>
<!ATTLIST subtopic.section number CDATA #REQUIRED>
<!ATTLIST subtopic.section keyword-one CDATA #REQUIRED>
<!ELEMENT section.title (#PCDATA)>
<!ELEMENT section.subsection (subsection.title, subsection.paragraph+)>
<!ATTLIST section.subsection number CDATA #REQUIRED>
<!ELEMENT subsection.title (#PCDATA)>
<!ELEMENT subsection.paragraph (paragraph.sentence+)>
<!ATTLIST subsection.paragraph number CDATA #REQUIRED>
<!ELEMENT paragraph.sentence (#PCDATA)>
<!ATTLIST paragraph.sentence number CDATA #REQUIRED>


The proposed DTD defines the hierarchy described above and illustrated in Figures 3 and 4. The XML tag requiring that at least one keyword be included in each section is in bold. DTD's provide a definition of the required and permitted grammar for a document collection. A particular instance of a document is compared to the DTD to ensure that it is a valid version. DTD's perform an important function where document integrity and uniformity is important (Fisher 2004). While document definition can also be provided through the means of XML schema's, DTD's are particularly suited for textual data since they have a simpler syntax, are typically shorter and allow for small, efficient text parsers (Carlson 2001) (Liebmann 2001).

CONCLUSIONS AND FUTURE RESEARCH

Research strongly suggests that the explicit representation and utilization of document structure can result in more effective document communication and navigation. Document structure can also drive both greater precision and greater recall in information retrieval from document collections. XML has been shown to be an effective means of representing document structure, enriching the structure through the application of domain-specific mark-up tags (to add meaning to structure) and enabling computer manipulation of documents to facilitate information search and retrieval tasks.

This work suggests that the codification of the AAL will benefit from the incorporation of a carefully considered formal XML structure. such a structure can support improved search capabilities for the AAL. More specifically, the search process can be made easier by supplementing user-entered keyword searches with additional related terms that can be derived from mark-up. A clearer determination of the meaning (beyond simple word counts) of portions of the AAL can be described through markup tags enabling both increased recall and precision of user searches.

First steps toward refining the codification structure are presented here. Additional refinements will likely be necessary as the migration from U.S. GAAP to the IFRS begins. The globalization of financial accounting standards introduces additional language problems including differences in regional dialects in English (e.g., stock options vs. share options) and regional differences in spelling (e.g., recognize vs. recognise). such differences will exacerbate the problems, previously discussed, associated with keyword searching. Current controversy surrounding the derivatives market may result in new accounting literature providing for new accounting treatment that, in turn, introduces new accounting terminology and/or new accounting concepts. There may be multiple resulting versions of the AAL over time. It is critical that research address how best to in corporate changes to the AAL and to archive versions of the AAL.

Ultimately the success of the codification project will be determined by the users of the AAL. That is, will users find the codification easier to navigate, more understandable and will user searches of the AAL be more effective? Future academic research, that addresses how the codified text is used and by whom, should be performed. Research into how the technological specifications of document structure can better assist in facilitating the use of the AAL also should be conducted. For example, academic research into appropriate 'keywords' that would best serve as indicators of the meaning and application of the various sections of the AAL would be valuable to the FASB as it continues the codification project. Finally, research should explore the development of an accounting ontology. Ontologies explicitly describe the vocabulary for a domain along with definitions of classes, relations, functions and constraints (shah et al. 2002). Such an ontology could be used to describe categories the documents comprising the AAL fall into, which could further be used to describe relationships between the documents. Whether the current FASB Conceptual Framework Project will either serve as or facilitate the development of an accounting ontology remains an open question.

REFERENCES

Abiteboul, S, Buneman P., & Sucio S. (2000). Data on the Web. San Francisco, CA: Morgan Kaufman Publishers.

Baeza-Yates, R., & Riberiot-Neto B. (1999). Modern Information Retrieval. New York, NY: ACM Press-Addision Wesley.

Baker, W. H. (2001). HATs: A design procedure for routine business documents. Business Communication Quarterly. 64:65-75.

Banerjee, s., Krishamurthy, v., Krishnaprasad, M., & Murthy, R. (2000). The XML Enabled Data Management System. Proceedings of the 16th International Conference on Data Engineering. Washington, D.C., USA: IEEE Computer Society. 561-568.

Belkaoui, A. (1980). The Interprofessional Linguistic Communication of Accounting Concepts: An Experiment in sociolinguistics. Journal of Accounting Research. 18(2), 362-374.

Bench-Capon, T.J.M., & Visser, P. R. S. (1997). Ontologies in Legal Information Systems; The Need for Explicit Specifications of Domain Conceptualisations. In the Proceedings of the Sixth International Conference on Artificial Intelligence and Law. Melbourne, Australia: ACM Press. 132-141.

Bovee, M., Kogan A., Nelson, K., Srivastava, S. P., & Vasarhelyi, M. (2005). Financial Reporting and Auditing Agent with Net Knowledge (FRAANK) and eXtensible Business Reporting Language (XBRL). Journal of Information Systems. 19(1) 19-41.

Burke, J. (1997). Report on Standards Overload. The CPA Journal. 66(3) 11.

Burnett, R. E. (2005). Technical Communication. Boston: Thomson Wadsworth.

Campbell, N. (2006). Communicating Visually: Incorporating Document Design in Writing Tasks. Business Communication Quarterly. 69:399-403.

Chien, S. Y., Tsotras, V. J., & Zaniolo, C. (2001). XML Document Versioning. ACM SIGMOD Record Special Issue: Special Section on Advanced XML Data Processing. New York, New York: ACM Press. 46-53.

Chien, S., Tsotras, V. J., Zaniolo, C., & Zhang, D. (2006). Supporting Complex Queries on Multiversion XML Documents. ACM Transactions on Internet Technology. New York, New York: ACM Press. 6(1) 53-84.

Couture, Barbara. (1992). Categorizing professional discourse: Engineering, administrative, and technical/professional writing. Journal of Business and Technical Communication. 6:5-37.

Debreceny, R. & Gray, G. L.. (2001). The production and use of semantically rich accounting reports on the Internet: XML and XBRL. International Journal of Accounting Information Systems. (2): 47-74.

Eunson, B. (2005). Communicating in the 21st Century. Milton, Australia: Wiley and sons Australia.

FASB. (2007). FASB Codification and Retrieval Project structure of subtopics. Accessed on June 10, 2007 from http://72.3.243.42/project/C-Structure of subtopics.pdf.

Fisher, I. E. (2007). A Prototype System for the Temporal Reconstruction of Financial Accounting Standards. International Journal of Accounting Information Systems. 8(3) 139-164.

Fisher, I. E. (2004). On the Structure of Financial Accounting standards to support Digital Representation, storage and Retrieval. Journal of Emerging Technologies in Accounting. 1(1) 1-21.

Flower, L. & Hayes, J. (1980). The cognition of discovery: Defining a rhetorical problem. CCC. 38:47-55.

Gangolly, J. S., Hedley, T. P., & Wong, C.T. (1991). Semantic Knowledge Bases for Financial Accounting Standards. Expert Systems with Applications, 3, 117-128.

Garnsey, M. R. (2006). Automatic classification of accounting concepts. Journal of Emerging Technologies in Accounting. 3(1).

Garnsey, M. R., O'Neil, J. & Stokes, L. (2006). FARS Database Searching: Providing Potential Search Terms to Students. Paper presented at the 2006 AAA Midyear Meeting of the Information Systems/Artificial Intelligence and Emerging Technologies sections held in Scottsdale, AZ.

Garnsey, M. R., & Hotaling, A. (2005). Improving Identification of Search Terms in the FARS Database. Paper presented at the 2005 Accounting Information Systems Educator Conference, Breckenridge, Co.

Geerts, G. & White, C. J.. (2004). SportsStuff.com: A Case Study on XML Technologies, e-Business Process and Accounting Information Systems. Journal of Information Systems. 18(2) 12-17.

Goldberg, L. (1965). An Inquiry into the Nature of Accounting. Sarasota, FL: AAA.

Guo, L., Shao, F., Botev, C., & Shanmugasundaram, J. (2003). XRANK: ranked keyword search over XML documents. Proceedings of the 2003 ACM SIGMOD International Conference on Management of Data. New York, New York: ACM Press. 16-27.

Haried, A.A. (1972). The semantic Dimension of Financial Statements. Journal of Accounting Research. 10(2), 376-391.

Harrison, J. & Freeman, R. E. (1998). Stakeholders, social responsibility and performance. Academy of Management Journal. 42, 479-485.

Kon, H. & Hoey, M. (2005). Leveraging collective knowledge. Proceedings of the 14th ACM International Conference on Information and Knowledge Management. New York, New York: ACM Press. 560-567.

Larsen, B., Tombros, A. & Malik, s. (2006). Proceedings of the SIGIR '06. New York, New York: ACM Press. 663-664.

Lauritsen, M. (1993). Knowing Documents. Proceedings of the Fourth International Conference on Artificial Intelligence and the Law. New York, New York: ACM Press. 184-191.

Lee, Y.K., Yoon, S. J., Yoon, K., & Berra, P. B. (1996). Index structures for documents. Proceedings of the First ACM International Conference on Digital Libraries. 91-99. Bethesda, MD: ACM.

Marian, A., Abiteboul, S., Cobena, G. & Mignet, L. (2001). Change-centric Management of versions in an XML Warehouse. Proceedings of the Twenty-seventh Very Large Database Conference. San Francisco, CA: Morgan Kaufman. 581-590.

McEwen, R.A., Hoey, Y., & Brozovsky, J. (2006). The FASB's Codification Project: A Critical step Toward simplification. Accounting Horizons. 20(4).

Moens, M.-F. (2005). Combining structured and unstructured information in a retrieval model for accessing legislation. Proceedings of the International Conference on Artificial Intelligence and the Law. New York, New York: ACM Press. 141-145.

Norvag, K. (2002). Temporal Query operators in XML Databases. Proceedings of the 2002 ACM Symposium on Applied Computing. New York, New York: ACM Press. 402-406.

Redish, J. C. (2000). What is information design? Technical Communication. 47:163-166.

Riahi-Belkaoui, A. (1995) The Linguistic Shape of Accounting. Westport, CT: Quorum Books.

Routen, T. & Bench-Capon, T. J. M. (1991). Hierarchical formalizations. International Journal of Man-Machine Studies. 35:69-93.

Saliminen, A. & Tompa, F. W. (2001). Requirements for XML document databases. Proceedings of the 2001 ACM Symposium on Document Engineering. New York, New York: ACM Press. 85-94.

schriver, K. A. (1997). Dynamics in document design: Creating texts for readers. New York: John Wiley.

Seeger, M. W. & Ulmer, R. R. (2003). Explaining Enron: Communication and Responsible Leadership. Management Communication Quarterly. 17:58-84.

Shah, U., Finin, T., Joshi, A., Cost, R. S. & Mayfield, J. (2002). Information retrieval on the semantic web. Proceedings of CIKM '02. New York, NY:ACMPress. 461-468.

Smart, K. L., Whiting, M. E., & DeTienne, K. B. (2001). Assessing the need for printed and online Documentation: A study of customer preference and use. Journal of Business Communication. 38:285-314.

Sumita, K., Ono, K., & Seiji, M. (1993). Document Structure extraction for interactive document retrieval systems. Proceedings of the 11th Annual International Conference on Systems Documentation. 301-310. Amsterdam, The Netherlands: ACM Press.

Svenonius, E. (1992). Ranganathan and Classification Science. Libre. 42(3). 176-183.

Wilkinson, R. (1994). Effective retrieval of structured documents. Proceedings of the 17th Annual International ACM-SIGIR Conference on Research and Development in Information Retrieval. 311-317. Dublin, Ireland: Springer-Verlag.

Wong, R. K. & Lam, N. (2002). Managing and Querying Multi Version XML Data with Update Logging. Proceedings of the 2002 ACM Symposium on Document Engineering. 74-81. New York, New York: ACM Press.

Yi, J. & sundaresan, N. (2000). A classifier for semi-structured documents. Proceedings of the 6th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 340-344. Boston, MA: ACM Press.

Ingrid E. Fisher, Ph.D., C.P.A.

The University at Albany

Department of Accounting and Law

i.fisher@albany. edu

Ruth Ann McEwen, Ph.D., C.P.A.

Suffolk university
COPYRIGHT 2009 Innovation Congress
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Fisher, Ingrid E.; McEwen, Ruth Ann
Publication:Issues in Innovation
Date:Mar 22, 2009
Words:6813
Previous Article:Intolerance of ambiguity effects on evaluations with the balanced scorecard.
Next Article:Toward developing a social network site-based model for knowledge sharing in the travel industry.
Topics:

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters