Recognizing Digital Genre

Much like the distinctive profiles of downtown Manhattan and Toronto that provide valuable cues about the identity of those cityscapes, the physical landscape of a document also contains distinctive, salient features that inform users about a document's identity. These physical traits shape seemingly amorphous bodies of information into cohesive units that make up a functional whole. For example, the undulating pattern of bold word and definition signal a dictionary while the juxtapositioning of date, address and salutation denote a letter. These patterns define a genre's form and, for the reader, trigger document recognition. Genre, derived primarily from literacy studies, explores the classification of texts. We can say that any digital document is an instance of one class of genre such as report, bibliography and dictionary and that each class is distinguished by a set of attributes. Some attributes relate to a document's content – the meaning of the document as expressed in words, phrases and sentences – and some relate to its form, as manifested in the visual appearance of the document such as formatting and layout. These attributes (both form and content) evolved in accordance with the customs and practices of a discourse community as a useful means of representing information to be communicated. For example, the size of the paper, columns and headlines together with the newspaper articles signal that a document is a newspaper. Some classes have almost universal recognition such as a letter or memo while others such as a bibliography or financial report are limited to specific user communities. The ability to recognize documents – almost instantaneously – enhances a user's interaction with documents and makes document use more efficient. Typically in the interaction with documents people rely on signaling devices to help with the reading process as well as with the comprehension of the document. Many of these signaling devices are dictated by the document genre. It has been suggested that the introduction of the typewriter and file storage led to the standardization of office documents and thus the emergence of new genre in business. Common textual features enabled office workers to distinguish readily among different types of business documents and typists subsequently became gatekeepers for correct formatting of each class of genre. [See Yates & Orlikowski for a detailed discussion.] Practices such as these have disintegrated on the Web; document format is constrained by the limitations of HTML and how it is rendered and suffers from the every-person-a-designer syndrome. Creating Web documents is a cookie-cutter affair as documents of differing types are formatted with essentially the same structure, eliminating or disguising those visual cues that help people to make sense of the content and requiring additional effort to interpret the document. To better understand the role of document structure in document recognition, we devised a set of studies to test various aspects of the concept. Applying the concept of genre as a foundation or theoretical construct to the study of digital documents is popular device used to acquire a better understanding of the development and evolution of digital documents. [See Crowston and Williams as an example.] In our case, the use of genre was a convenient and fruitful device that enabled the separation of form from content. In a series of studies, we used the separation of form and content to create different version of documents. We defined three versions: Content version: all structure, layout and formatting were deleted, leaving only the words, phrases and sentences. In this case the document resembled a single running paragraph of text. Form version: all characters were converted to 'X' or 'x' and all digits were converted to '9,' leaving the document's structure in tact, but devoid of understandable content. Full version: with both content and form in tact. Over a series of three studies, people examined multiple versions of a document selected from a set of 24 unique genre that are used within academic discourse communities. Participants came from academic circles with the exception of one study that also used members of the general public. A Curriculum Vitae (column2) is formatted in column 1 as a Letier and in column 3 as a Dictionary. [For details of these studies, examine Toms et al.] In general, people identified some documents by form, some by content but, not unexpectedly, most by the full version. Overall the form version was recognized the least. But, in a surprising twist people took twice as long to recognize the content and full versions as they did the form version. The form version exhibited a regular pattern that was logical and consistent with a person's mental model of that class of genre; the groupings of the strings of the characters "x" and "9" seemed to represent a recognizable shape. That the form version did not outperform the content version was not the whole point of this exercise; the content version provided self-evident semantic clues such as "minutes of meeting," while the form version required the matching of visual cues with a corresponding representation stored in long-term memory. In addition, the presence of a document in a different medium (print vs. digital) made no difference to the ability to recognize the document. Surprisingly, discourse community also had no impact on recognition. People from non-academic worlds were equally adept at recognition. Because non-academic people had spent some time in their past in an academic setting, they clearly had experienced slow or no degradation of the genre concept. Over the course of the studies only half the documents were recognizable in the form version. However, people tended to recognize the form version of a document if that document type had been used frequently and/or recently. As a final test of the contribution that form makes to document recognition, we mixed and matched content and form within a single document. For example, the content of a curriculum vita was reformatted to look like a letter, a dictionary and so on (see Figure 1). People focused on the form or structure of the document first and were more likely to call a bibliography-formatted-like-a-dictionary (illustrated in the third column in Figure 1), a dictionary rather than a bibliography. Because the form takes on a distinctive visual appearance, document form essentially represents the shape of a document. That shape is likely two-dimensional since people did not seem to need the multi-dimensional qualities present in the print world to distinguish a shape. Ultimately, the unique shape triggers a user's mental model of that class of genre. In interpreting the shape, a user develops a set of expectations about the document without first having to read the semantic content. It seems that form of a document is a powerful metaphorical construct in document recognition and one that should be exploited in document creation in order to enhance document use. While a class of genre has a particular format, all instances of one class of genre are not necessarily identical; the journal article with variations by both form and content is a good example of this. Documents, however, contain a minimum set of attributes that define the class and likely a threshold that distinguishes one class from another. Clear from our work is the notion that form is tightly coupled with the class of genre. Footnotes for example, will never be displayed in 16pt bold font. The current move to XML (and its family of standards) for the creation and manipulation of Web documents will impose standards on document structure. However, the loose connection between style and content still has the potential of ignoring the relationship between content and its form and thus reducing our ability to instantaneously recognize a document by form. The effective use of digital documents depends on a person's ability to recognize the structure and purpose of a document – its class or genre. To be useful documents must conform to the regular and logical pattern of elements (the shape of information) expressed by a discourse community. The visual structure manifested by the form, e.g., layout and formatting, suggests a shape of information – a shape that is universally recognizable within particular discourse communities, such that it may act as an interface metaphor with the potential to facilitate user document interactions. Understanding digital genre will impact the chaotic, "anything goes" world of the Web document presentation. As more extensive use of the Web is made by e-commerce, the need for standard document types will be of commercial expediency and force the standardization of digital genre. Findings to date suggest that capitalizing on the concept of genre and exploiting its expression as shape will aid document recognition and facilitate user-document interaction. The author wishes to thank OCLC for funding the research underpinning this paper through an OCLC Library and Information Science Research Grant. Elaine G. Toms is associate professor in the Faculty of Information Studies at University of Toronto, 140 St. George St., Toronto, ON M5S 1A1. She can be reached by mail or by telephone at 416/978-7802 or by e-mail at toms@fis.utoronto.ca

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.