Businesses are embracing an XML future to prepare for changing times.

XML has just celebrated its tenth birthday, yet while that period has seen a rapid expansion of XML's technical capabilities, its transition from the technical to the business realm is still in its youth. There is little doubt that technologists have been proved right in their evangelising of XML as "the future", however there are still many lessons to be learned from businesses so that XML can fulfil its full potential and slot smoothly into our digital futures.

Today's changing times put ever more pressure on businesses to be able to respond quickly, to respond well, and to respond profitably. Increasingly companies rely upon fast, efficient movement of data, whether in e-commerce exchanges with others or when processing data through their own internal workflows. Whenever data is exchanged, however, it is vital that it is both consistent and accurate.

No longer just the subject of backroom technical activity, data is becoming a hot topic for boardroom discussion. Increasingly with dramatic examples of identity theft and lost data in the news, it has become a priority in many companies. In an information age, data is a valuable company asset and its use (and abuse) must be strictly monitored and controlled. Looking ahead it will not be enough simply "not to fail"--smart companies which optimise use of their data will be afforded opportunities at the expense of less savvy competitors.

Adoption of e-commerce

All organisations that hold sensitive personal data must be vigilant for identity theft, while industries such as the pharmaceutical or aerospace need to be very clear on the validity of their data where the consequences of error can be catastrophic.

Other industries, such as the retail, financial services, advertising and petrochemical businesses are fast adopting e-commerce as the preferred way to conduct business transactions. Orders, invoices, supplier and product information messages all rely upon the integrity of their underlying data to ensure smooth exchange. If the data is poor, it will always take more time to resolve the issues and consequently will be more expensive.

XML--fast becoming an orthodoxy

Clearly the way that data is stored within an organisation is important and is significant for data governance. Data must be agile, accessible and useful for the long term. XML has established itself as the most popular format for high volume data storage, and has over recent years grown from being adopted by many content providers and publishers, to become the universal language underpinning the Web and modern application software. Even commodity applications like Microsoft[R] Office[TM] today store their files in an XML format.

XML initiated the revolution when it was published in 1998 and it popularised structured mark up and introduced the concept of well-formed documents. It is an incredibly simple, well-documented, straightforward vendor-independent data format, and has lowered the barriers to entry--any tool that can read text files can display an XML document.

Longevity for data is an issue. It is a fact that reading older Word processing files might already be problematic (it is even difficult to find machines with floppy drives today!). Adopting a universal standard for storing data, ensuring that data is stored in the correct way and being able to audit exactly what data you have, is what companies strive for. Not to mention protecting that data for its long term use.

Flexible and portable

XML is considered today the most portable and flexible document format since the ASCII file, supporting all human languages. Indeed it has been so widely adopted that it is even a standard for storing the data in TV remote controls. It has been standardised by the World Wide Web Consortium (W3C) as a format for computer documents.

XML is flexible enough to be customised for domains as diverse as web sites, e-commerce and voice mail systems. However, while it represents many opportunities for organisations, it is not a panacea and there has, inevitably, also been considerable hype around XML. XML is not a programming language, network transport protocol or database. On its own is has no intelligence and no meaning. It is a format for data, and as such, to ensure business benefit is derived from any data, there is a need for validation of the meaning of that data, and for managing the correctness of an organisation's entire XML data holding--in other words, XML data governance.

XML Data Governance

Data Governance ensures that the correct schemas, systems, personnel, procedures, practices and feedback are in place for well-managed XML data. IT governance is developing in many areas already, but is really in its infancy for XML processing.

For commercial publishers, a group who have been at the forefront of XML adoption, data quality is immediately relative to the value of the service being provided. For many publishers it is important to verify the quality of any intellectual content coming into, and flowing out from, the organisation. It is a known fact that bad data fails more often than programmatic communication between systems.

There are products on the market today that address the issue of the technical state of data for digital content. Such products can validate data and minimise the subsequent problems that may occur through exchange. Validating the accuracy of data also provides a publisher's customers with an assurance of quality on approved material being supplied.

Publishers value

Cambridge University Press (CUP), for example, has taken this validation one step further. The organisation has developed an application that acts as a 'data firewall'. CUP has managed to reduce their participation in the supply chain process by developing validation rules that are trusted enough to provide assurances that content passes a quality threshold. At this point, CUP routes documents directly from their suppliers to their onward destinations.

This application has been developed over time, but has clear business benefits to all parties. Suppliers can address issues early in the production process and the publisher gets timely throughput of content than would not otherwise have been possible. Faster, better and cheaper--this is the mantra of the successful business.

Griffin Brown Digital Publishing's particular product, XML-Probe has been used to implement some public online services to promote the use for modern validation. These have to date focused on publishing and include ONIX for Books--an e-commerce data transmission for the publishing industry, whereby potential users can get a feel for the type of checks that can be done using such validation technologies.

Unique brands

However, industry level support is not enough for real-world applications. In reality each content producer, each industry, will have different standards, which represent that organisation's unique brand offering. For a publisher these may, for example, be reflected in certain house style and editorial practices; for a research organisation it may be reflected in the richness of the data it can provide.

To accommodate a company's unique data sets, a validation product must be able to be developed with different rule sets. The validation tool then highlights violation of these rules, according to pre-set constraints. This may be the way in which a date is written, for example, or an author's name recorded. This might be done at different stages of content production document creation, review, classification and publication. The next stage might then be for the tool to communicate automatically with a cleansing tool.

Data--the most valuable asset

There is no doubt that in the digital age, content is king. To this, it might be added that "and quality is queen"--since, in practice, the value of any data is defined by its usefulness. The confidence of users always relies on quality assurance.

For organisations that rely upon data to support their core business, it is important that they store data in a high quality, agile format. Those publishers who were slow to move from print publishing and embrace the web have quickly seen their businesses assailed by more nimble competitors. Content providers that looked to the future saw that digital content must be adaptable to be supplied online, and today data must be made available to hand-held devices such as mobile phones and BlackBerrys.

Similarly businesses that rely upon data access for engineers, field workers and mobile employees, need to ensure that their data is stored in a universal format that is easily understood and read by a wide range of people and devices. Choosing a non-proprietary technology such as XML is the future. Forward-looking companies will also already be assessing the risks of low data quality and proactively addressing issues of data governance and quality. www.griffinbrown.co.uk

Alex Brown, Griffin Brown Digital Publishing