Current trends in IT and business computing indicate an evolution toward the dynamic exchange of business information, market intelligence, and commercial transactions. Technology standards such as XML have provided the common language needed to interoperate effectively across networks and systems.

In addition, the collection of technologies known as Web services (XML, SOAP, UDDI, WSDL, etc.) provides a complete platform for interbusiness service and data exchange that's rapidly becoming the modus operandi for e-businesses.

The goal of these efforts is twofold yet oddly diametric - on the one hand you have the need to de-couple business systems and infrastructures so they can change freely, and on the other you want to couple disparate networks and systems together on demand. Enterprises use a variety of different technologies to meet their needs and these have often prevented seamless ways of connecting their systems together. To address this, XML-based protocols and schemas have introduced a way to de-couple technical systems, allowing for dynamic exchange of business-specific, self-described data. Yet, the other extent of Web service technologies is to join businesses, not so much technologically but rather in terms of business processes or information.

Popular Web service platforms utilizing UDDI (Universal Description, Discovery, and Integration) permit for the dynamic searching and association of business entities described therein by WSDL (Web Services Markup Language) documents. Specifically, one business system can establish criteria to match entries in a chosen registry, which can lead to spontaneous business transactions with the described entity in an implementation-independent manner.

These capabilities will change the way enterprises interoperate (or enter-operate if you will). One possibility that can emerge from this is the aggregation of business information with the goal of establishing coherent market intelligence among members of a business community. Companies sharing a common interest or objective will find ways to band together and benefit from the ability to contribute to and participate in a common economic model that enhances their market viability as a whole within given vertical sectors. I use the term information syndication to refer to the aggregated publishing of business information as a commercial service.

Today, tangible and media commodities can be delivered across a value network via corporate syndication through various channels (see Figure 1). The presence of Web services provides the necessary technology platform for a variety of disparate companies to create a virtual platform with the aim of delivering business information, data, and market intelligence as a combined service. This article explores the idea of information syndication using Web services.

The Importance of Web Services
Traditionally, enterprise extranets have involved managing vast software, hardware, and human resources. Business extranets are costly to develop, operate, and maintain. Common facilities among extranet partners were often duplicated between their systems, increasing the complexities of the partnership network. Furthermore, security concerns regarding controlled access to information within corporate network boundaries typically meant sensitive (and nonsensitive) infor-mation couldn't be shared or transacted since they often existed in the same place. It's difficult to assign segregated policy concerns in a fine-grained way to corporate data, let alone provide external access control to resources within the corporate intranet and continually manage them.

These issues have stymied businesses' ability to come together dynamically in ways that allow for a rich flow of information into and out of corporate networks and repositories. One of the attractions of the Web services platform is the way it hides the details behind the Web services protocols (or interfaces; see Figure 2). This effectively allows companies to publish only the services they intend to offer and to encapsulate acceptable security policies within those services regardless of how a particular user accesses the system (since Web services define this already).

Because XML is the protocol metalanguage used to describe messages used in Web service conversations, businesses can provide their own semantic meanings to self-described data arriving or retrieved from a variety of sources. This de-coupling of data formatting from semantics enables businesses to share common ontologies of data, which can then be turned into information useful for that particular business.

Information Service Providers
In fact, information service companies have been doing this for many years. They've used a variety of means to capture volumes of raw data from various sources, which is then deposited into a database where it can be sorted and formatted for content. Additionally, information subscription services, such as Reuters and AP, online content service providers like the Wall Street Journal and Bloomberg, and database service providers such as Lexxus/Nexxus all capitalize on gathering data, converting it to information, and formatting it as content for particular audiences. Often the base data is the same, but enhanced or tailored to particular subscribers.

This is the general model employed by an information syndicate, yet the information is gathered from the syndicate network members and typically distributed via proxy (which we'll discuss later). In traditional media syndication, you often have a single source with multiple distributors. Our information syndication model is the inverse of the traditional approach and therefore I refer to it as the inverse syndication model (see Figure 3).

Data Warehousing
Data warehousing technologies attempt to bind disaggregated data into common, semantically coherent sets for search and re-distribution. They often do this by normalizing a dictionary of metadata associated with the raw data, which is then converted into searchable, often usable, information. This is useful for a number of reasons, including qualified searches, more accurate results, and interoperable information.

Different sectors will promote specific metadata most applicable to their data (or industry), and similar companies will seek out common, binding metadata to make their information useful and usable within their community. (Fortunately, this has been going on for a few years now as companies and sectors define rich XML schemas to process intercompany XML-based messages.)

Information service companies, sometimes known as "data marts," will promote these data-unifying efforts and in turn publish their databases to numerous subscribers. In addition to this, information syndicates will seek to promote an economic common interest between the member companies or the market as a whole. In this way, it acts as an avenue to revenue for the data providers who seek to hone their offerings for the advancement of the alliance.

A separate revenue model can be employed on behalf of the syndicate as profits can be re-distributed in various ways back to the member providers (in the form of dividends, incentives, and the like). This economic justification is important to lure members into a particular syndicate network, which then, in turn, improves the attractiveness of the system - achieving a natural feedback and scaling effect.

Information Webs
All businesses produce information, yet most don't maintain ways of publishing that information, converting it to valuable content or monetizing it. This stems from some basic economic assessments of the businesses' core competencies. The fact is, many businesses, especially those outside of IT, do not have the capital or expertise to set up complex service provisioning networks, nor the business model to justify housing the technical responsibilities or weathering a separate value network exclusively. Nonetheless, they continue to gather volumes of raw data that can be mined for valued information. The trick is extracting it in a safe and secure manner, and providing an economic justification at the same time.

The Web services platform, including ebXML (electronic business XML), provides the needed pieces to automate service monetization and transactions over the Internet. This provides a scalable, volume-driven commerce model that can balance the costs of establishing the infrastructure. However, it's important to understand the effective demand for the services and information provided by an individual company. From the view of a single company, the economic picture may be difficult to see clearly, yet when the demand is multiplied through the establishment of a strong partnership network, a critical mass can be achieved. This is what information alliances and syndication offer in this context.

Using the Web service approach, companies within (or across) vertical sectors can establish interconnected service networks permitting them to exchange information useful to their markets. However, it may be necessary for members to establish a consortium of interest to carry the group policies and manage the technical aspects of delivering the combined information services on behalf of the individual members. In this case, the need for proxy representation becomes clear.

Proxy Repositories
It will be advantageous for the syndicate alliance to appropriate a proxy service provider to manage the warehouse repository and the policies governing commercialization and distribution of its content. It can do this a number of ways, one of which is to set up and sponsor a separate company (or delegate to one) to handle the duties. By outsourcing this role, the businesses can overcome the economic and technical burdens that would not otherwise mix with their core com-petencies or revenue models. The proxy service provider might toll transactions against the syndicate and not require any direct service fees or payments. Many interesting business models can be employed here; however, that discussion falls beyond the scope of this particular article.

A proxy service provider will host access to an information syndicate and expose a Web service front-end allowing subscribers to execute complex queries, receive periodic transmissions, and access other useful ways of extracting information from the syndicate members without having to compromise their identity. It'll also provide clusters of repositories containing strategic information obtained from the syndicate members. This externalization of information protects the provider's intranet and databases from anonymous access and shields the technical platform details of the provider's infrastructure from users of the proxy service provider. The use of Web services as an exposure mechanism allows subscribing entities such as individuals or other computing systems to transact and acquire the information automatically.

Publishing Models
There are two primary methods of interfacing with the proxy warehouse as a provider. One involves publishing or proactively pushing information from within your corporate networks directly into the proxy warehouse via a Web service interface (maintained by the proxy service), as shown in Figure 4. This gives you the comfort and security of deciding what information is targeted and how this actually takes place internally - as well as how often. The other method relies on the proxy warehouse to poll your Web service interface periodically and decide what information to retrieve (see Figure 5). Even then, the individual company can restrict the scope of access from the proxy warehouse. The added benefit of the latter permits the business to simply publish the appropriate compliant Web service front-end that the proxy warehouse will use to contact it.

No additional internal maintenance or management need be done routinely to ensure participation in the network. However, this is a more passive approach as you may not be precisely aware in advance of who can be accessing your services, or how access is achieved.

The proxy service provider can also broker requests for information directly into the syndicate members in a dynamic, on-demand basis. In this situation, information seekers access the syndicate network, as a whole, through the proxy service, much as Internet search engines work today. All system interaction occurs via Web service interfaces (or common Web front-ends) and the appropriate access permissions and policies will be in place within and behind the proxy service provider. In fact, its duty is to enforce the concerns and policies of its sponsors.

A Business Scenario
Here's a practical business scenario. There are many different auto service and repair stations throughout any given metropolitan region. If you're looking for a particular tire for your vehicle, depending on which service center you go to, they may or may not have it. Fortunately, most service center computer systems can search other locations and check inventory elsewhere, but only within their own company. Now, suppose a band of service centers run by different owners know they're constantly referring business to one another when supplies are low, yet they reap no reward for this. The owners come together and decide to execute a referral reward system. Whenever some automotive part, in this case a tire, is sold as a referral from an alliance member, a finder's fee is awarded. Now the problem is that they have no way of automating this process since their software systems are all different. They want to form an information syndicate where they can publish to their proxy warehouse only the number of tires they have on hand (maybe they agree on the referral award program only for tires and don't want to expose other inventory details).

Using the approaches we've discussed, each alliance member would publish tire inventory data into a proxy warehouse where other members could subscribe to it and pull it into their software systems. Maybe their policy also allows anyone to subscribe to that information for a fee. The information is updated automatically by each service center's systems using Web services as the glue. Their systems aren't directly connected and therefore are protected from direct access.

Consider the independent outsider, for example, an industry analyst researching the number of tires sold or housed by service centers in a particular region. The analyst too could subscribe to the information through the proxy service provider and get up-to-date and accurate information. Nothing special would be needed as the information could be obtained or delivered in any number of ways, such as common Web content, e-mails, pages, etc. Many other kinds of businesses can benefit from this approach as well. Financial, health care, and market research organizations naturally transact volumes of data to conduct business, but the manner in which they do so doesn't permit for the levels of separation and automation we've presented here, or is more exclusive to the individual parties where greater economic interest might exist.

Conclusion
The dawn of Web service technologies is here, and before sundown companies will be racing to identify how to benefit from the platform and what it has to offer. Ideally, we see an increase in the exchange of data via Web services as it provides an automated way to tap into business information. For many companies, the issue is what to offer and expose, and why. Information syndication can provide a demand-driven market for corporate data gathered and provisioned by large alliance networks. The syndication of information will provide dynamic ways to publish and transact otherwise dormant data for the sake of market intelligence and profitability. It remains to be seen how the critical masses can be achieved and whether the revenue model of information syndication is sustainable beyond certain high-volume verticals. We can expect to find out in the coming years.

Author Bio:
Darren Govoni is the founder and CTO of Metadapt Design Systems, Inc. Darren currently works for Cacheon, Inc, whose technology is based on technology he developed at Metadapt. dgovoni@metadapt.com