Standards Stack Aspects

Standards Stack Aspects

Whereas message-oriented protocols follow a very structured, regimented layer scheme, document-oriented specifications have no layering structure to them. Rather, these specifications can be applied to any level of the message-passing stack as well as the community vocabularies in the level above. As such, rather than having layers, the document-oriented specification’s portion of the XML standards stack merely has “aspects” that are applied when they are needed.

In general, there are at least four major categories of document-oriented specification:

   Presentation specifications. These specifications detail how XML should be presented or modified in presentation for usability.

   Security specifications. These specifications provide a level of protection of XML information.

   Query specifications. These specifications assist in locating XML resources.

   Semantics specifications. These specifications help to apply meaning and context to XML documents.

Presentation Aspect

One of the major document-oriented specifications aspects involves those specifications that define how XML documents should be presented to the user. These presentation aspects help to transform any XML document on any layer of the XML standards stack into a form that can be visually understood and processed by humans. The goal of speci-fications in this stack is not to focus on machine processing of XML documents but rather on the human factor in using XML. Presentation specifications include the format-ting of documents for display as well as the addition of graphical, multimedia, and tim-ing elements.

Usability and information portability also are major factors addressed by presentation aspect specifications. Usability specifications focus on making information easier to use and access by users of all types. These specifications aim to meet the needs of those who are physically handicapped as well as to help to make information generally more acces-sible to all. Form technology and text-to-speech specifications are good examples of usability-focused presentation aspect specifications. Information portability specifications aim to make the information contained within more accessible to different devices, form factors, and systems. With the increased usage of cell phones, PDAs, and memory-con-strained devices, various presentation aspect specifications have been created to enable the widest distribution of content as possible.

Major presentation aspect specifications include XHTML, XForms, and Scalable Vector Graphics (SVG), among others.

Security Aspect

With the increased distribution of content and sensitive data comes the need to protect that information. However, security is a catch-all word that actually embodies many different concepts around protecting information, all of which can be applied to XML documents.

The first level of security is the protection of information from prying eyes. Encryption specifications help to alleviate these concerns by masking XML data and preventing it from being used, viewed, or processed by parties that are not privy to the information.

Encryption specifications make use of widely available techniques for protecting data, including advanced private-key protection mechanisms. Therefore, the strongest of protection technology can be used and applied to XML data.

Authentication provides another level of security to XML documents. Even though you may be able to decrypt an XML document, it is important for an application to verify that you are who you say you are. Numerous compromises in security occur when unau-thorized users abscond data and make use of keys they should never have had access to. A variety of authentication specifications have been created to address this need.

A further level of security is provided by authorization and permission specifications that attempt to identify which resources a valid user has access to. These authorization speci-fications indicate the specific resources, information, or other digital assets a user can use and the restrictions on that use. These specifications include not only the assignment of user controls but also controls on the content to be exchanged to the user to prevent unauthorized duplication and use. Known as Digital Rights Management (DRM) specifications, these limits on the use of intellectual property are becoming increasingly popular in this era of the digital asset.

A final level of security is applied by privacy specifications that aim to make sure that those who are entitled to information don’t intentionally or inadvertently spread the information to parties who aren’t entitled to it. Increasingly, users are worried that their personal and private information will be shared with parties they have no intention of sharing their information with. Privacy specifications, and especially the Platform for Privacy Preferences (P3P), are aimed at giving users control of how their

data will be used, shared, and stored. In this manner, all aspects of information security can be ensured.

Query Aspect

Another universal need for information, especially the richly structured information contained with XML, involves the ability to locate and make proper use of data. As is necessary in most database and data storage systems, the ability to query information is as important as the ability to store and represent that data. Query specifications are responsible for retrieving information and tagging it for proper identification and return.

A number of major XML specifications exist that help in the tagging of XML documents with metadata needed to assist in their proper retrieval. In addition, many proposals and specifications have been created to specify a language for the global query of these documents.

Semantics Aspect

Because XML allows users the ability to create any vocabulary and structure of their choosing, the main challenge is in synchronizing these vocabularies with other, incom-patible representations. In addition, it has become important for machines to understand not only the literal encoding of documents but the intent and context of the human who created them. There is nothing that prevents different organizations from calling the same data element different things. Also, there is nothing that prevents these very same users from using the same name to mean entirely different things. For example, the meaning of the word title denotes different things to those in the publishing and insurance industries. Another major problem is the fact that different languages and cultures have different names for the same item. It is important for our representation of information to cross these conceptual, semantic, and language boundaries.

A major initiative called the Semantic Web is squarely focused on addressing these prob-lems and producing specifications that add a contextual, or semantic, layer to the way we represent information in XML. The semantics aspect applies these specifications to all levels and layers of the XML standards stack. The immediate application of these speci-fications is to simplify and enable users to make better, more relevant searches for con-tent. Many search engine responses to user inquiries result in large amounts of irrelevant information. To a machine, the information may seem relevant, but to a user the context of those responses is entirely inadequate. Besides, we should be able to search for a term in any language of our choice and have the results still be relevant to us even if it is pre-sented in a different language.

A larger and more ambitious implementation of semantic aspect specifications is for machines and systems to make intelligent guesses as to our intent for the use of informa-tion and to retrieve data sources in an “educated” manner. In this vein, the Semantic Web approaches the goals of artificial intelligence as much as it solves needs for the XML user community. As semantic aspect specifications are developed, they will no doubt be applied to all levels of the XML standards stack.