Patent Grant
7051071
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
The following applications are cross-referenced and incorporated herein by reference:
U.S. application Ser. No. 09/788,150 entitled “OPEN MARKET COLLABORATION SYSTEM FOR ENTERPRISE WIDE ELECTRONIC COMMERCE,” by Rocky Stewart, Timo Metsaportti, and Pal Takacsi-Nagy, filed Feb. 16, 2001.
U.S. application Ser. No. 09/785,687 entitled “MESSAGE ROUTING SYSTEM FOR ENTERPRISE WIDE ELECTRONIC COLLABORATION,” by Rocky Stewart, Pal Takacsi-Nagy, Timo Metsaportti, and Michael Hyndman, filed Feb. 16, 2001.
U.S. application Ser. No. 09/785,872 entitled “PLUGGABLE HUB SYSTEM FOR ENTERPRISE WIDE ELECTRONIC COLLABORATION,” by Rocky Stewart, Pal Takacsi-Nagy, Pascal Hoebanx, Sanjay Dalal, and David Wiser, filed Feb. 16, 2001.
U.S. application Ser. No. 09/785,693 entitled “CONVERSATION MANAGEMENT SYSTEM FOR ENTERPRISE WIDE ELECTRONIC COLLABORATION,” by Rocky Stewart, Pal Takacsi-Nagy, Timo Metsaportti, Sanjay Dalal and Pascal Hoebanx, filed Feb. 16, 2001.
The present invention relates generally to the transaction of business-to-business commerce using the Web.
To prosper in the Internet economy, a company must be able to streamline its business process, eliminate unnecessary cost, and react quickly to competitive pressures. To support this mandate, information technology (IT) organizations must have the ability to rapidly consolidate systems from newly created or incorporated divisions and, in many cases, to exchange mission-critical information with outsourced partners as a virtual enterprise. New Internet applications must be linked with each enterprise's core technologies to deliver accurate information, solid operational performance, and a satisfactory customer purchase experience. And, whether companies do business on the Web or not, customers expect all businesses to run on Internet time—a rapid pace that calls for seamless, high-performance business operations that can be accomplished only when applications are integrated with each other to create the zero latency enterprise.
The first business issue is the integration of critical business information and logic that reside in enterprise production and legacy systems. Multiple IT systems that are implemented as silos or standalone systems, each addressing particular business issues, should be integrated across the enterprise.
Traditionally, islands of individual applications that need to share data have been integrated in an ad hoc manner, using homegrown integration and, more recently, Enterprise Application Integration (EAI) tool sets. These hard-coded and point-to-point solutions attempted to address the problem but created an IT nightmare of innumerable spaghetti-like connections between applications. EAI solutions that cobble together a combination of point-to-point interfaces, procedure calls, file transfers, and “e-mail-like” messaging to deliver enterprise-wide data transformation and routing are notoriously high risk.
Further, multiple point-to-point and ad hoc solutions are inherently difficult to deploy and maintain because they lack a centralized management and support for real-time change management. These features are desirable to the management of business processes and the goal of addressing the e-commerce integration environment.
The second business issue is the need to incorporate Web-based, e-commerce applications. The goal of getting on the Web, however, was not a simple one; only new businesses starting from scratch had the luxury of building from the ground up. With substantial IT infrastructures and investments in place, most businesses were faced with having to leverage current information systems and merge them into the e-commerce age. Businesses that are not able to integrate customer-facing Web applications with enterprise systems find that they simply cannot support today's competitive market and customer demands. This is because data is locked away in disparate systems that do not speak the same language, much less speak to each other.
The third business issue is taking full advantage of integrated e-commerce, including the integration of business processes with trading partners. E-Commerce enables companies to enjoy significant savings and increase revenue opportunities by (1) improving service to customers, (2) engaging in just-in-time production, (3) finding more competitive providers of operational goods and production materials, and (4) being able to create new markets quickly. These two objectives—reducing cost and increasing revenue—have always been driving business forces. However, to date, there has been no solution that has provided the key enablers for e-business. Electronic data interchange (EDI) emerged in the 1990s as a standard for exchanging data between companies. EDI is a highly structured, expensive, and time-consuming technology. Packaged solutions have also emerged that claim to solve business-to-business (B2B) dilemmas; however they also require heavy investment in proprietary application technology across all business partners, entail significant configuration expertise, and deliver a hard-coded, pre-packaged process that all parties must adopt. Packaged solutions and EDI deliver very little competitive advantage, are not customizable for differing business processes, and are expensive, time-consuming options.
Most people are familiar with the traditional EDI batch mode type of B2B, and with the more recent electronic store fronts of dot-coms, otherwise referred to as basic commerce. While both EDI and basic commerce provide companies with business advantages that were not previously available—the high costs and technical constraints of implementing EDI and the minimal transaction functionality of basic commerce have limited their adoption as long-term e-business strategies. Community commerce, the third type of B2B e-commerce, features the same characteristics of basic commerce—the simple matching of buyers and sellers online—with perhaps some added functionality, such as auction and search, and the sharing of common information with vertical or community groups. However, these three types of commerce share one common limitation—they do not involve real-time, dynamic conversations between businesses. The sharing of information is in the form of simple point-to-point messages that are not really conversations. While these types of commerce may be sufficient to do things like sharing simple inventory information with suppliers, or posting product catalogs for buyers, these types of commerce have, to date, done little more than simply take information and automate its distribution to participants.
Existing B2B models fall short in meeting modern day requirements. From EDI and outsourced procurement to supply chain management systems and first-generation online exchanges, companies are feverishly searching out technology that will enable them to quickly, efficiently, and reliably conduct business with trading partners over the Internet. Unfortunately, companies are discovering that most of these traditional B2B technologies meet only a fraction of their e-business IT requirements. A more comprehensive, powerful and flexible solution is required to successfully meet the challenges of collaborative B2B commerce today.
While EDI offers an advantage over paper-based business transactions, it has traditionally been restricted to large companies with substantial IT budgets and technical resources. EDI typically requires as much as 9–12 months to implement and can be quite costly. Finally, EDI is not real-time or process-driven. Data is batch transmitted, and transactions can be sent out of sequence without system oversight. In some cases, data may not be received at all—creating consternation and confusion on all sides.
Unlike EDI, new B2B solutions are designed to leverage the Internet and technologies such as Java, in order to lower the barriers to entry for trading partners of any size, reduce deployment cycles and cost, and power managed, real-time communication between partners. Recognizing that EDI will continue to be used by many of the world's largest companies because of familiarity and confidence in legacy systems, new Internet-based e-business platforms must also co-exist comfortably with EDI legacy systems. Nonetheless, as the real and opportunity costs of using EDI grow, companies are investing IT resources in Internet-based systems that make more financial and strategic sense.
Another commonly found form of B2B interaction is the use of outsourced e-procurement applications. Web-based purchasing of maintenance, repair, and operation (MRO) supplies—the goods required to run a company but not the materials used in the direct manufacture of a product or the provision of a service—is becoming more and more common every day. To deliver process efficiencies and cost savings, many companies have outsourced their MRO provisioning to third parties because these goods are non-strategic, low-cost items that can be intermittently purchased in bulk from a pre-established set of suppliers.
However, from a technology perspective, outsourced e-procurement is much like EDI in that it increases efficiency over paper-based processes but is also monolithic, inflexible, and requires all participants to adopt a single proprietary application. This does not enable customers to create competitive advantage by streamlining their mainstream purchasing process or doing business any differently than direct competitors. These serious limitations, combined with the fact that MRO vendors typically own all the data associated with using their exchange, mean that most companies are reluctant to consider e-procurement vendors for their core or strategic enterprise functions. In fact, some participants in MRO exchanges may consider e-procurement vendors as potential downstream competition and are careful to limit the scope of business conducted on these sites.
Given these factors, while many companies may continue using MRO procurement for a portion of their e-business portfolio, they must still acquire strategic technology that will encompass MRO exchanges but support larger and more strategic e-business objectives.
Some organizations have implemented B2B solutions as a first-generation online exchanges for e-markets and supply chain management. An e-market maker can be defined as an organization that develops a B2B, Internet-based, e-marketplace of buyers and sellers within a particular industry, geographic region or affinity group. The list of e-market makers is growing larger every day.
Although it is indisputable that e-marketplaces in general will power much of e-business tomorrow, supporting a broad variety of collaboration, purchasing sales and trading activities, the majority of them are not delivering on the B2B promise today. A recent research study found that of 600 surveyed exchanges, only 10 provided application integration. A primary goal of B2B and online marketplaces is to reduce human intervention. This goal cannot be achieved unless systems are integrated from end-to-end.
E-Markets without back-end integration to a company's enterprise systems significantly reduces the return on investment fore-market makers and participants, and limits that company's opportunities to satisfy customers in new and unique ways. Supply chain transparency, which occurs when e-market makers have increased visibility into a supplier's product availability and delivery information, enables B2C companies to better meet customer demands in real-time and represents a significant source of competitive advantage. The differentiating factor between two supply chains often boils down to which one is better at managing the information float—the time between when data is captured in one place and when it becomes available and actionable elsewhere.
As e-markets continue to proliferate—interconnecting with each other and growing into tomorrow's interconnected e-business “powergrid”—they promise to super-charge the B2B network effect. E-businesses need to respond quickly to market demand, and as such look to build not just their own supply chains but also those of their customers and suppliers. These benefits will flow through entire industry segments as information improves, order processing becomes easier, and all parties can begin to rely on getting what they need to meet customer demands in a timely fashion.
The need to move both operational and customer information at high speed from one end of a distributed enterprise to the other, as well as beyond its firewalls, creates an e-commerce environment that can be adequately addressed only with an end-to-end integration architecture.
E-Business is about competition, and competition is about service speed and quality. Company size and scope are no longer the determining factors for marketplace success, because even very small companies can generate a significant Internet presence. Capturing new business opportunities using the Internet requires business strategies to be developed and implemented in Internet time, and the triumph of speed over size is becoming more apparent every day. While the key drivers for B2B are speed and competition, collaboration between trading partners is the key to winning the game. Truly changing the rules requires real-time information exchange, transactional integrity, flexibility to adapt to changing market conditions, and integrated trading relationships between business partners. The stakes are high, and can make or break a company. The rules change daily, as every e-business executive can attest. These are the real e-business challenges. Meeting these challenges requires a different kind of B2B technology that can meet the real demands of collaborative commerce.
As disclosed herein the invention helps to define an open, standards-based, and scalable software platform for dynamic business-to-business collaboration. The software platform may be embodied in various forms of server product, referred to herein as a collaboration server, or may be incorporated into larger scale enterprise collaboration server products referred to herein as collaboration systems. The several benefits provided by the invention include a comprehensive conversation management capability, an Internet based standard, and a pluggable business logic and protocol support.
A primary component of the collaboration system is the collaboration space (c-space). The c-space is an abstraction supporting a single business model, business message protocols, a secure message space, security policies, quality of service policies, and a registered set of business trading partners. The c-space contains message vocabularies, business process models, participant roles, and other e-market metadata that are essential to the creation, deployment, and ongoing maintenance of trading activities.
Using the invention, a c-space owner can create any number of concurrent c-spaces, each supporting any number of trading partners. Within a c-space, the invention provides asynchronous XML messaging capabilities to allow loosely coupled communication between trading partners.
The collaboration system further comprises a collaboration hub (c-hub). The c-hub is the execution engine of a c-space, allowing the c-space owner and trading partners to create, route, and manage messages within the trading environment. The c-hub delivers powerful messaging services that are essential to ensuring the security, quality, relevance, and success of conversations across the c-space.
Within the context of B2B e-commerce, well-defined and ordered sets of related messages are exchanged between trading partners. This collective set of messages or “conversation” involving two or more trading partners can often span days, weeks, or even months. As conversations and business processes are initiated, executed and completed, conversation management software tracks and manages these long-living conversations, ensures that they are completed, and orchestrates the overall process execution.
Each conversation has a unique context that enables users to manage multiple, concurrent conversations taking place in the c-space. The collaboration system uses such context information to help ensure that messages from one conversation do not get tangled up with messages from another. For example, an individual trading partner may be requesting proposals or negotiating prices with multiple vendors concurrently, and must maintain the integrity and security of each interaction.
In one embodiment, the invention comprises a collaboration system that allows for sharing of workflow information between collaboration participants, comprising a collaboration server having stored thereon an enterprise workflow, a workflow server having stored thereon a participant workflow and an integration mechanism for allowing said enterprise workflow and said participant workflow to interact with one another.
End-to-end integration is both an overall architecture and a series of IT solutions that address three key business issues: The integration of critical business information and logic that reside in enterprise production and legacy systems; the desire to incorporate Web-based e-commerce application; and the need to take advantage of integrated e-commerce, including the integration of business precesses with trading partners.
There is a general desire in the industry to develop new infrastructures for new B2B capabilities referred to as collaborative commerce. Collaborative commerce resembles an interactive, managed conference call of buyers and sellers. Real-time conversations among trading partners allow them to actively negotiate and execute business transactions, not just pass along information. Each and every B2B activity should be manageable as a seamless, integrated, and secure business event, regardless of how long this business transaction takes, or how many business parties or IT systems are involved—internally or externally.
The new economy has transformed traditional brick-and-mortar businesses into e-businesses, and this transformation directly affects the way companies collaborate with their customers as well as a multitude of trading partners. As such, e-business is of increasing importance to most companies from the corporation's business perspective, e-business collaboration among trading partners can help speed new products or services to market, improve customer service, reduce inventory and supply chain costs, and ultimately increase shareholder value. From a technology perspective, e-business extends the use of existing, proven IT systems and business processes to include external, disparate systems that are increasingly being accessed over the Internet. As a result, new groups of people, such as customers, suppliers, and distributors—who previously did not have access to these systems—are also directly exposed to the requirements and nuances of how these individual systems and business processes operate. But most importantly, access to these systems brings with it new access to information that was previously unavailable to all but insiders within an organization. With this information comes the ability to collaborate in ways and with partners that were previously unavailable to most parties.
E-Business is a real-time entity: interactive, customer driven, and highly competitive. The e-business winner will be the company—or set of companies—that can quickly and efficiently deliver products and services to partners, customers and end users with tight integration at all points along the value chain. Today's market leaders recognize that e-business is about creating a value proposition so compelling and a business experience so immediate and efficient that all partners and suppliers make doing business over the Internet a standard part of how they do business.
While the key drivers for B2B are speed and competition, collaboration between trading partners is the key to winning the game. Truly changing the rules requires real-time information exchange, transactional integrity, flexibility to adapt to changing market conditions, and integrated trading relationships between business partners. The stakes are high, and can make or break a company. The rules change daily, as every e-business executive can attest. These are the real e-business challenges. Meeting these challenges requires a different kind of B2B technology that can meet the real demands of collaborative commerce.
Good B2B infrastructure software should be readily configurable and rapidly deployable, while being part of a comprehensive and cohesive e-business solution. It should deliver time-to-market benefits now, with the ability to scale to meet future needs and interoperate with next generation applications. It should scale to handle any volume of B2B transactions, flawlessly and reliably, day-in and day-out, around the globe. It should enable a company to conceptually manage each and every collaborative commerce activity as a seamless, integrated and secure business event, making the firewall transparent to the business user while leveraging emerging Internet programming standards and security models to empower IT. Finally, this infrastructure should deliver new capabilities to dynamically add or change business partners on the fly, the power to model and manage partner relationships in all of their complexity, and the flexibility to enhance business processes to meet changing market conditions and customer demands.
These are the technology requirements for executing collaborative commerce strategies—and this is the area the present invention targets. As disclosed herein the invention defines an open, standards-based, and scalable software platform for dynamic business-to-business collaboration. The several benefits provided by the invention include:
1. Rapid Time-to-Market: IT execution is critical to companies that must extend their business to trading partners. Whether integrating the supply chain or launching new e-market exchanges, companies need to get into the B2B game today with technology that is readily configurable for rapid deployment—without the need for significant system re-writes, business process reengineering, or negotiation to install a common B2B partner application set. The solution should be easy and relatively inexpensive for partners to integrate in a non-proprietary fashion, with options for leveraging and extending existing investments in electronic data interchange (EDI), B2B applications, and emerging XML messaging standards such as RosettaNet. The system incorporating the invention is dynamic and simple to configure. Trading partners can be dynamically added and removed from the collaboration, helping to rapidly achieve the critical mass essential for marketplace initiatives—with the flexibility to make revisions as business models and partnerships evolve. Templates, easy-to-use graphical user interfaces designed for the business-level analyst, and configuration tools ensure that businesses can quickly design and execute mission-critical B2B processes.
2. Lowered Barriers-to-Entry: Trading partners can rapidly join and participate in an automated manner by simply downloading lightweight software which enables them to participate in trading activities. Additionally, the invention has been designed to provide support for multiple business protocols—to accommodate the diverse needs of multiple industries and diverse communities of trading partners.
3. Reliable, Layered Messaging Platform: Companies typically go-to-market rapidly with their B2B strategies, gaining an initial e-business foothold and then implementing the rest of the technology in phases. Although time pressures are intense, IT recognizes that it is high-risk and costly to invest in piecemeal technologies from multiple vendors that do not work together, cannot scale, and are not readily extensible. Individual e-business IT initiatives ideally fit into a single, comprehensive architecture that allows companies to get into the game quickly, yet be confident that they can successfully incorporate value-added business logic and build an integrated and high-performance B2B platform over time. To accomplish this, the invention provides a robust and reliable messaging platform designed specifically to meet the distributed, heterogeneous and loosely coupled nature of B2B e-commerce. Intelligent message routing and the ability to define the quality of service are an intrinsic part of the messaging platform. Additionally, a layered architecture provides a flexible means of adding new and emerging transport protocols to meet current and future requirements.
4. Comprehensive Conversation Management Capabilities: Complex B2B relationships require open, standards-based process technology that reliably defines and manages the execution of cross-enterprise processes. This process architecture must respect the autonomy of individual trading partners, while allowing the definition and management of common process agreements across trading partners. To be effective, this technology must be flexible enough to manage system-to-system processes as well as seamlessly incorporate humans and real-world interactions into the process. Unlike most B2C transactions which may involve a limited number of participants over a limited number of days (and, in some cases, hours), well-defined B2B interactions or “conversations” between trading partners may take weeks or even months to complete and may involve numerous participants. To meet these requirements, the invention may utilize a variety of sophisticated messaging protocols, including the extensible Open Collaboration Protocol (XOCP) which provides life cycle management of long-living conversations, and ensures the overall execution and completion of business processes.
5. Single Process Platform: While B2B technology must be rapidly deployable; it must also be powerful and flexible enough to support changing processes, evolving partner relationships, and new business models. It must support a variety of configurations ranging from centralized e-market administration to distributed partner-centric workflow integration. It must be straightforward and enable the user to add and change relationships without bringing the system down or compromising the integrity of transactions that are in progress. Embodiments of the invention provide end-to-end e-process management by interfacing with other third-party workflow products including the BEA WebLogic Process Integrator, a J2EE-compliant and XML-based process integration engine from BEA Systems Inc., San Jose, Calif. BEA WebLogic Process Integrator is one example of products that enable the effective, end-to-end management of system and application-based business processes, as well as traditional human-based workflow on both sides of a company's firewall.
6. No Need for Business Process Reengineering: The invention enables the creation of one or many virtual trading communities on the Internet, connecting trading partners in a conceptually free-standing environment with its own defined business processes. This unique design reduces the need for business process reengineering, enabling disparate trading partners to participate in public processes which take place within the e-market while allowing individual trading partners to define, implement and execute distributed, private business processes locally.
7. Internet Standards-Based: B2B is most valuable and effective when it is fully integrated from the back-end systems and extends out across the firewall to trading partners, regardless of the disparate IT systems and processes employed. Adhering to open industry standards such as Java, XML, and HTTP on nonproprietary systems optimizes partners' ability to integrate seamlessly and quickly. Standards-based technology also allows IT to leverage existing skill-sets rather than being forced to invest in something that is proprietary and cannot be replicated. To ensure rapid, efficient integration of partners, embodiments of the invention use a standards-based architecture leverages an HTTP communications protocol for information transportation over the Internet, as well as standards and concepts such as SSL, PKI and J2EE—reducing the need to dedicate internal IT resources to proprietary technologies.
8. Security: With trillions of dollars at stake, B2B commerce systems must support high transaction volumes and varying levels of traffic and throughput. Furthermore, this IT infrastructure must be 100 percent reliable and ensure the integrity and security of all interactions. Today's e-business not only stakes millions in B2B infrastructure and applications, but also puts its reputation with trading partners and customers on the line. Thus, the technology must have a well-designed architecture, and be compliant with the full range of emerging Internet security standards. The invention provides a trusted and secure platform for conducting B2B e-commerce, including built-in Internet security capabilities for data privacy, authorization, authentication, and an infrastructure to support non-repudiation.
9. Pluggable Business Logic and Protocol Support: Embodiments of the invention provide a plug-in architecture for dynamic and intelligent routing of messages. This enables market makers to design and implement business rules that meet their specific needs. Additionally, multiple business protocols such as RosettaNet and EDI can exist concurrently in the same deployment—reducing the need for all trading partners to standardize on one particular protocol. The invention also provides a flexible framework for rapidly implementing custom business logic that may be required to maintain a competitive advantage.
10. Proven, Scalable Technology: As B2B infrastructure vendors consolidate, it is increasingly critical to ensure that you will be fully supported by a strategic partner you can count on 24-hours a day, seven days a week, around the globe and well into the future. Embodiments of the invention can be designed to run on industry-leading transaction servers such as BEA WebLogic Server from BEA Systems, Inc., San Jose, Calif.
11. Interoperability: The invention uses an open and flexible architecture that can interoperate and leverage existing investments in EAI/B2B technologies as well as enterprise databases, Supply Chain Management (SCM), Customer Relationship Management (CRM), and mainframe systems. The invention also enables explanation of Common Object Request Broker Architecture (CORBA), Enterprise JavaBeans (EJB), Tuxedo, and COM+ applications. This inherent flexibility provides a highly scalable and adaptable platform for the development and deployment of future B2B initiatives.
Glossary of Terms
DUNS—Data Universal Numbering System. A distinctive nine-digit identification sequence that is an internationally recognized common company identifier in global electronic commerce transactions.
PIP—Partner Interface Process. A protocol provided by RosettaNet that specifies the business messages exchanged, their vocabulary, and how to validate them. A number of these protocols have been defined for various types of business transactions (e.g.: TransferShoppingCart, QueryOrderStatus, etc.)
RosettaNet—Depending on its use in context, RosettaNet is either of the following definitions: An independent, non-profit consortium that is defining standard electronic business interfaces; or the business and transport protocol specified by this group.
XOCP—XML Open Collaboration Protocol. A protocol used by the collaboration server that provides more complete life cycle and management support for business interactions. XOCP is described in more detail later in this document.
Collaboration System
Roughly described, the invention comprises a collaboration server, which may itself be a component of a larger collaboration system.
Atop the collaboration system sits a business process and workflow server 104. Examples of workflow servers that may be used with the invention include the WebLogic Process Integrator product from BEA Systems, Inc. The workflow server allows a company to define their business processes as a workflow. The workflow may take into account all aspects of a company's business from start to finish, including, for example, manufacturing processes, purchasing, sales, financial processes, etc. A key attribute of a workflow is that it be dynamic, flexible and configurable, i.e. that it may be readily modified to reflect a company's changing business model, to add new elements such as new product lines, and to account for new events as they occur. The workflow server also acts as an interface for a company's business application to plug into and utilize the workflow. Business application may include, for example, catalog management 106, dynamic pricing 108, order management 110, logistics 112, and payment applications 114.
In one embodiment of the invention, sandwiched between the application infrastructure server and the workflow server lies the collaboration server 116. The collaboration server is a key component in creating a collaboration system, and in bringing workflow management to the enterprise. The collaboration server allows separate companies, herein referred to as trading partners, to collaborate on complex tasks or projects. Such collaboration may extend, for example, to complex business projects, manufacturing processes, or logistical tasks encompassing many companies spread over many countries. As such, the collaboration server acts as part of a collaboration system to allow globally dispersed trading partners to share information and resources, and to work towards a common goal.
A key attribute of the collaboration server is the ability to interface with workflow processes running on a workflow server. To this end the collaboration server must be able to control the flow of information into and out of the workflows maintained by the workflow server. For collaboration among different companies the collaboration server must be able to do this for many workflows, and many workflow servers, and must be able to perform this task simultaneously, equally, and reliably. The collaboration server of the invention does this by means of a sophisticated routing and filtering mechanism. Messages from different trading partners, or in some instances the workflows of different trading partners, are filtered by the collaboration server, and routed to the appropriate recipients in a true collaborative fashion.
The development of the collaboration space is shown in
Collaboration System Architecture
In one embodiment the collaboration system may utilize many components, including:
Not all of these components need be installed centrally, on the same server, and indeed the c-space enabler is more likely to be installed at a remote trading partner location than at any central hub location. The c-space enabler, or simply the enabler, is however an important part of the enterprise-wide collaboration solution. Other components, such as the workflow integrator are optional, and while not necessary for participation in the collaboration network, greatly increase the potential benefits to any company or participant using a workflow to model their business processes.
Collaboration Hub (C-hub)
The c-hub is the centerpiece of the collaboration server. It is responsible for routing messages between various c-enabler components and for managing the lifecycle of the various conversations between the participating trading partners. The invention uses a c-hub which provides the following functionality:
The message flow through the c-hub begins with an incoming message 172 from a trading partner via proceeds to the hub router, and then proceeds down as an outgoing message 188 to the recipient trading partner or partners, as indicated by the arrows. The two sides of
The main characteristics and contractual obligations of each component are illustrated in the flowchart
The incoming message is read and routed to an appropriate decoder chain based on the message protocol (e.g.: XOCP, RosettaNet, cXML, etc.). In one embodiment, that defines c-spaces in terms of URL's, the URL on which the message is received serves to identify both the protocol being used and the destination c-space.
Any protocol-specific headers in the message are processed by the decoder and the appropriate business protocol handler is assigned. The sending trading partner is also identified. The sending trading partner is then enlisted in a conversation, and a reply is prepared to be returned to the sender.
The message is stored for subsequent retrieval.
The trading partners to whom this message should be routed are determined. Logic Plug-Ins can add or remove recipients from this list.
Final validation of the message recipients is performed, and the message is stored for delivery to the targeted trading partners.
A message destined for a single trading partner recipient is received and a decision is made on whether to send the message. Filter logic plug-ins may be used to add or remove this message from this recipient's receive mechanism.
Internal operations relating to quality of service issues, conversation management, etc. are performed if the message is still schedules for sending to a recipient.
The encoder performs any transportation of the message required to support the business protocol of the recipient.
The C-hub transport layer sends the message to the trading partner recipient.
For each module chain, the main responsibility is as indicated above. This contract must be fulfilled in order to ensure proper message processing and delivery. As long as the contract is fulfilled, other modules in the chain are relatively unencumbered in terms of the processing that they may be required to provide.
Collaboration Space (C-space)
A primary component of the invention is the c-space.
Using the invention, a c-space owner 214 can create any number of concurrent c-spaces, each supporting any number of trading partners. Within a c-space, the invention provides asynchronous XML messaging capabilities to allow loosely coupled communication between trading partners 216, 218. This model leverages routing and filter functionality that can be associated with messages in order to classify the set of trading partners that should receive a message, allowing partners and their interactions to be managed individually, based on their role or trading preferences within an e-market.
The c-hub is the execution engine of a c-space, allowing the c-space owner and trading partners to create, route, and manage messages within the trading environment. To facilitate the execution of business transactions across a disparate base of trading partners, one embodiment of the invention uses XML as its e-business messaging semantic.
Within, as well as outside of a c-space, XML offers tremendous advantages as a universal format for messages passed between trading partners because it provides a common syntax to structure information. However, XML by itself doesn't solve the interoperability problem, as collaborating entities must agree on the semantics of business protocols for this information. The invention is independent of business protocol message vocabulary, so it can support any standards-based or proprietary business protocol or vocabulary. As such the invention delivers the ability to support multiple protocols within the same c-space as well as to extend a c-space through supplemental protocol handlers or business logic plug-ins.
Within the context of B2B e-commerce, well-defined and ordered sets of related messages are exchanged between trading partners. This collective set of messages or “conversation” involving two or more trading partners can often span days, weeks, or even months. To manage these unique conversations, embodiments of the invention implement an extensible Open Collaboration Protocol (XOCP), which provides the ability to specify both the information (vocabulary) and business protocol (process flows) for a given conversation. As conversations and business processes are initiated, executed and completed, conversation management software tracks and manages these long-living conversations, ensures that they are completed, and orchestrates the overall process execution.
The invention allows the powerful business process management and workflow capabilities of workflow server products such as BEA WebLogic Process Integrator to provide the infrastructure to facilitate conversations within the c-space. The innovative architecture allows each trading partner to handle the implementation of their own business process and rules locally, while conforming to the rules of engagement (defined by the global information and business protocols) for a given c-space.
Features include:
Given the complexity, duration, and volume of business transactions required between trading partners conducting collaborative commerce, an effective solution must provide functionality to address these specific requirements. Unlike traditional solutions, the invention provides a unique approach to managing on-going interactions between individual trading partners as well as the context of each conversation within a c-space.
Conversation Management
Within the context of B2B e-commerce, well-defined and ordered sets of related messages are exchanged between trading partners. This collective set of messages or “conversation” involving two or more trading partners can often span days, weeks, or even months. To manage these unique conversations, embodiments of the intention provide a collaborate system that implements the extensible Open Collaboration Protocol (XOCP) described in further detail below, which provides the ability to specify both the information (vocabulary) and business protocol (process flows) for a given conversation. As conversations and business processes are initiated, executed and completed, Conversation management software tracks and manages these long-living conversations, ensures that they are completed, and orchestrates the overall process execution.
Within the collaborative system defined by the invention the powerful business process management and workflow capabilities of workflow products and workflow servers such as, for example, the BEA WebLogic Process Integrator provides the infrastructure to facilitate conversations within the c-space. Additionally, the invention's innovative architecture allows each trading partner to handle the implementation of their own business process and rules locally, while conforming to the rules of engagement (defined by the global information and business protocols) for a given c-space.
Conversation context is another important aspect of managing conversations within a given c-space. Each conversation has a unique context that enables users to manage multiple, concurrent conversations taking place in the c-space. The collaboration server uses context information to help ensure that messages from one conversation do not get tangled up with messages from another. For example, an individual trading partner may be requesting proposals or negotiating prices with multiple vendors concurrently, and must maintain the integrity and security of each interaction.
As shown in
A primary difference between the two means of conversation finishing illustrated in
URL Accessible C-spaces
C-spaces are a useful, but generally XOCP-specific, concept. Most existing standard business protocols like RosettaNet and cXML do not use the concept of c-spaces or partitioning. Many of these protocols are point-to-point oriented, and do not even have the concept of multi-cast. To increase accessibility and ease of integration, embodiments of the invention preserve the c-space concept for XOCP, and introduce it for non-XOCP business protocols. In order identify a particular c-space and a business protocol, each c-space/business-protocol combination may have a unique uniform resource locator(URL). A client can use this URL in order to access a particular c-space using a particular business protocol. This also allows a single c-space to support multiple business protocols by using multiple URLs.
A c-space might have more than one URL to help identify special additional processing that may be required to support multiple business protocols or to handle variations in business protocol implementations. For example, a c-space may have a RosettaNet orientation, but participants may be using RosettaNet implementations from different vendors. Variations in the different implementations can affect how the decoder needs to process the information for routing purposes. One way to handle such a situation is to have a single decoder try to apply some heuristics to determine from which protocol variation a message is being received, and then process the message accordingly.
An alternative solution is to have the c-hub provider assign a separate URL for clients using each of these variations within a single c-space. In this approach, each message can then be routed to the specific Decoder that knows exactly how to handle the type of message being received.
Through the use of different URLs, a single c-space can support multiple protocols. These protocols can be quite dissimilar (e.g.: RosettaNet and XOCP).
Collaboration Enabler
In accordance with one embodiment of the invention a Collaboration Enabler (c-enabler) is a lightweight, readily downloadable software that enables a trading partner to participate in a c-space. A trading partner must be identified and authorized by the c-space owner or administration in order to activate the c-enabler and actively participate in a c-space. One of the security protocols supported the system is the X.509 certificate, which can be authorized by the c-space administration and granted by a Certificate Authority. The trading partner can then be authorized by the c-space administration (based on conversation types, business documents, and roles) to participate in a particular c-space. Like the c-hub, the c-enabler can easily integrate with applications running on other application servers and reduces the barrier-to-entry for trading partners who want to participate in the c-space.
In one embodiment, the minimum configuration requirement to run the c-enabler maybe designed to run on a servlet engine, for example the BEA WebLogic Express servlet engine. For complex B2B scenarios, many organizations may benefit from the use of a robust process management tool such as BEA WebLogic Process Integrator, a workflow server and integration product described in detail below, which runs on BEA WebLogic Server.
Workflow Integration
Workflow products and servers may be integrated with embodiments of the invention to manage complex e-business processes between multiple trading partners. Together they provide a federated process model among trading partners, where market processes are defined globally but deployed in a distributed manner with local processes being modeled and executed at each trading partner's site. The invention provides the overall management and coordination of these distributed processes as part of the product's conversation management functionality.
Workflow server products such as BEA WebLogic Process Integrator from BEA Systems, Inc. provide the ability to graphically model partner-based business processes. This tool also provides graphical design functionality for open business processes such as RosettaNet Partner Interface Processes (PIPs), which govern processes across the global information technology supply chain. RosettaNet PIPs enable companies to deploy standards-based, interoperable applications for supply-chain functions and to optimize networks of suppliers, manufacturers, and distributors.
While collaborative commerce is driven by the desire to automate and streamline e-business processes, one can not completely eliminate the ability to involve humans in the execution of e-business processes. Unlike other B2B platforms, the invention's process capabilities provide the ability to define and direct e-business process exceptions to human users for resolution. The combination of a collaboration server and a workflow server in one collaboration system also delivers unparalleled flexibility to drive and integrate cross-enterprise collaborative processes, enterprise applications, and transactions on either side of the firewall using a single process technology.
Integration Server
At run time, shown in
The workflow server process engine 340 may thus communicate directly with a worklist client, shown in
Business Protocol and Logic Plug-Ins
Evolving business models on the Internet require the flexibility to measure business performance in many different ways. Business Logic Plug-Ins allow the c-space owner to add unique functionality to a c-space such as auditing, billing, or invoking complex business rules. For example, the c-space owner may have a business model that requires them to bill or audit based on the content, volume, number of recipients, etc. of messages, which have been mediated by a c-space.
Additionally, while the Internet has provided companies with the ability to do business with nearly anyone that they want to—this might not always be desired. Using the Business Logic Plugins and complex business rules, filtering functionality can be used to enable individual trading partners to decide whether or not they want to receive a particular message (or perhaps any message) from a particular trading partner. For example, a trading partner may not want to receive a “request for proposal” message for fewer than 5,000 units. In this case, a filter would be used to reject any request for 4,999 or fewer units. Conversely, routing functionality can be used by a c-space owner to add to or delete trading partners from the recipient list for a particular message. This could be used to tailor the list to a relevant, desired set of recipients—or expand it in order to increase the chance of finding an appropriate “buyer” or “seller”.
Logic Plug-ins
Logic Plug-Ins is the term used to describe the individual components or modules in the chains shown in the preceding c-hub architectural overview. Each chain, as a whole, has an obligation to satisfy its respective contract, described earlier. However, as long as the contract is satisfied, the other modules in the chain are free to do additional processing.
Logic Plug-Ins can be “chained” so that, after one Logic Plug-In is finished running, the next sequential Logic Plug-In of this type will normally be activated. Each successive Logic Plug-In will be able to access the changes, if any, to modified message information.
The following list illustrates the types of additional processing (sometimes referred to as rules) that can be performed by modules in a chain:
Publish-Side and Receive-Side filters (described in further detail below) are special route modification Logic Plug-Ins placed in the Router and Filter chains, respectively. Either the c-hub owner or the trading partner or both can specify these filters. The Publish-Side filter is used by the sending trading partner or the c-hub owner to specify a list of target trading partner recipients for the message being sent. This filter, a router Logic Plug-In, may add to, remove from, or not change the current list of recipients. This list of recipients is still subject to validation by the c-hub Router. The Receive-Side filter is used by the receiving trading partner or the c-hub owner to determine whether or not the message should be sent to the specified recipient. It must provide a “yes” or “no” decision as to whether to send the document to the specified trading partner recipient.
Business Protocol Plug-Ins
Someone who writes a business protocol plug-in needs to implement the full suite of Decoder, Router, Filter, and Encoder Logic Plug-ins in the c-hub. This is required in order to provide the message processing capabilities necessary to pass a message from a sender, through the c-hub, to zero or more recipients. In addition, a conversation manager is required to handle the basic message processing capabilities needed to route a message. A business protocol plug-in developer should provide:
The c-hub stores all the information it needs to manage a c-space in a c-hub repository.
The fields used in one embodiment of the invention include the following:
//c-hub/cspace/business-protocol:
Each c-space supports one or more identified business protocols. The name is a logical name for a protocol defined in the Business Protocol Def.
//c-hub/cspace/business-protocol@url:
Each business protocol must specify a URL to be used to connect to the decoder for that c-space.
//c-hub/business-protocol-def:
The definition of a business protocol that is used in one or more c-spaces.
//c-hub/business-protocol-def/decoder:
This specifies the chain of decoders to be used when processing a message. These are logical names for Logic Plug-Ins defined elsewhere. At least one of these is responsible for fulfilling the responsibility for this type of Logic Plug-In and this business protocol. The Logic Plug-Ins in the chain will be processed in the specified sequence.
//c-hub/business-protocol-def/router:
This specifies the router chain, which is analogous to the decoder chain described previously.
//c-hub/business-protocol-def/filter:
This specifies the filter chain, which is analogous to the decoder chain described previously.
//c-hub/business-protocol-def/encoder:
This specifies the encoder chain, which is analogous to the decoder chain described previously.
//c-hub/business-protocol-def/java-class:
The Java class specifying the implementation of this particular protocol.
//c-hub/business-protocol-def/init-params:
When the business protocol is loaded and initialized, its init( . . . ) method will be invoked with a hash table of the name-value pairs specified in the following elements.
//c-hub/business-protocol-def/init-params/name:
The “name” part of a name-value initialization pair.
//c-hub/business-protocol-def/init-params/value:
The “value” part of a name-value initialization pair.
//c-hub/user-exit-def:
This provides a concrete definition for a Logic Plug-In specified by a logical name.
//c-hub/user-exit-def/document:
Zero or more DTDs used by the Logic Plug-In. These are the logical names for DTDs defined elsewhere in the Collaboration Server repository as Documents. The main decoder Logic Plug-In will frequently require one or more DTDs for non-XOCP protocols.
//c-hub/user-exit-def/java-class:
The Java class specifying the implementation of this particular Logic Plug-In.
//c-hub/user-exit-def/init-params:
When the Logic Plug-In is loaded and initialized, its init( . . . ) method will be invoked with a hash table of the name-value pairs specified in the following elements.
//c-hub/user-exit-def/init-params/name:
The “name” part of a name-value pair.
//c-hub/user-exit-def/init-params/value:
The “value” part of a name-value pair.
//c-hub/trading-partner-protocol:
Additional trading partner information needed for non-XOCP business protocol support.
//c-hub/trading-partner-protocol /external-url:
An external URL to where messages should be routed (business message and replies).
//c-hub/trading-partner-protocol /business-identifier:
A business identifier to be used by an external trading partner, distinct from the internal The Collaboration Server trading partner name.
//c-hub/trading-partner-protocol /business-identifier/type:
The type of the business identifier. For example, for RosettaNet, a “DUNS” number is used. The type names are under control of the c-hub administrator and need to be synchronized with the related business protocol plug-in.
//c-hub/trading-partner-protocol /business-identifier/value:
The value of the business identifier, correlated with the business type described earlier. For example, a “DUNS” identifier is a 9-digit number.
Deploying the Collaboration System
The invention provides a comprehensive infrastructure that meets the rigorous technology standards for B2B e-commerce, but keeps the barrier to B2B low for you and your business partners. The following section outlines how a company can rapidly create a c-space and transform traditional trading partners into co-collaborators, using the invention.
The privacy, security, and verification-of-receipt capabilities of B2B technologies must enable users to maintain proprietary business practices, meet ongoing business demands and maintain an advantage over their competitors. To support these requirements the c-hub can be configured to support a comprehensive set of Internet security features, including:
Privacy: SSL support for ensuring privacy of data over the Internet.
Authentication: Mutual authentication, where trading partners and the c-hub authenticate each other before allowing participation in dynamic collaboration activities. Authentication is based on X.509 certificates obtained from a trusted, third-party Certificate Authority.
Authorization: Access Control Lists and a role-based conversation participation mechanism to ensure authorized access to e-business collaborations and the corresponding messages that exist within in a c-space. Trading partners are authorized for specific roles within a c-space which, along with filtering features, govern the types of messages they can publish and receive.
Nonrepudiation: In some cases, business rules require that a receiver of a business document or message has legally sufficient and persistent proof about the receipt, the content, and the sender of the document—similar to that of a registered letter or notary public in the off-line world. The invention the required infrastructure support for digital signatures, digital receipts, and secure auditing.
Embodiments of the invention also include an administration console, that which allows the c-space owner or designated administrator to configure and manage services and c-spaces using a Web browser. Common administrative functions include managing trading partners (adding or removing trading partners, granting and revoking access to conversations, etc.), configuring messaging services, monitoring on-going conversations, browsing system status (interactions, message delivery, and logs), and generating activity reports.
The administration console offers dynamic management capabilities, which enable the re-configuration of system services on-the-fly, and the addition of new trading partners without disrupting the system. In addition to the Web-based user interface, system management functions are available via administration messages—this allows for automated monitoring and error recovery, and for integration to third-party system management tools.
XPATH Filtering
The c-hub routes business protocol messages that are sent between trading partners. The c-hub supports different business protocols, which can be generally classified as XOCP protocols or external protocols:
XML Open Collaboration Protocol (XOCP) is one of the default protocol used by the invention. This protocol is sent and received by trading partners using the c-enabler API. It provides support for conversation life cycle tracking and the management of business interactions by a mediating c-hub.
External protocols are business protocols that may be sent by other products and conform to other conventions or standards (such as a RosettaNet message from a Vitoria trading partner). The c-hub determines the business protocol to be processed based on the URL that is specified when a connection is initiated from a trading partner to the c-hub transport layer.
In the XOCP protocol the routing criteria are specified in a separate header of the message protocol. The sender can specify an XPATH router as a component of the message. This XPATH router is a string expression that specifies the intended recipients of the message in a very flexible manner using the XPATH syntax. The c-hub evaluates this XPATH expression after it receives an XOCP message from the sender. Thus, the XOCP protocol enables dynamic routing of the message to target recipients.
The c-hub maintains contextual information about interactions between trading partners (also known as conversation coordination). The XOCP protocol carries the information that enables the c-hub to monitor and maintain the status of the conversation and its participants. Before the c-hub retransmits an XOCP message to a set of target trading partners, it evaluates the XPATH expression against the set of trading partners configured in the repository to determine the routing of the message.
The routing functions of the c-hub may be extended to support additional XPATH expressions defined in the repository, as well as user-defined extended properties that can be referenced to control the routing of messages.
XPATH expressions for routing messages to trading partners is a special feature of the XOCP business protocol. When an application publishes a message using an external protocol (such as RosettaNet), the target recipient is explicitly encoded by the sender in the content of the message. Unlike the XOCP protocol, there is no concept of a mediating c-hub that can dynamically route the message to target recipients—both the sender and the target are completely specified by the external protocol.
The repository provides a pre-defined set of entity definitions (and associated properties) that support conversations between trading partners in a collaboration system using the invention. However, applications need to be able to extend these definitions so that application-specific information can be modeled and stored in the repository.
More specifically, the trading partner information in the repository is extended to support application-defined properties that can be associated with a trading partner. C-hub routing and filtering features can make use of these properties. These extended properties are organized by uniquely named extended-property-sets using the c-hub administration console and associated with one or more trading partners.
The extended property sets are modeled in the repository so that they can be retrieved as subtrees within an XML document. These XML subtrees appear in the generated message-context XML document sent to the XPATH routers and filters. The XPATH expressions for the routers and filters can reference these extended properties. The root elements of each extended property set associated with a given trading partner will be inserted as the last children of the <trading-partner> element node.
XPATH Routers
C-enabler XPATH routers 402—These are XPATH routers that are sent by an c-enabler application when publishing a message. The Conversation.publish( ) method allows a c-enabler application to specify an XPATH router expression when sending a message. This router contains an XPATH expression that selects a subset of <trading-partner> nodes from the message-context XML document generated by the repository. These XPATH router expressions can be extended so that c-enabler applications can select a set of trading partners by including references to application-specific extended properties.
Trading partner XPATH routers 412—These are routers that are specified by a c-hub administrator in the repository and associated with a sending trading partner. These routers are similar to the c-enabler XPATH routers that are sent with a message, in that they use XPATH expressions to select a subset of target <trading-partner> nodes. For every message sent by a trading partner using an XOCP conversation, the associated XPATH routers for that trading partner are evaluated against a message-context XML document generated by the c-hub for that message. The generated message-context XML document contains information about the sending trading partner, the conversation context, the message context, and a sequence of subtrees containing the XML parts of the business document.
C-hub XPATH routers 410—These are routers that are specified by a c-hub administrator in the repository and associated with all incoming messages in an XOCP protocol stack. These routers are similar to the trading partner XPATH routers that are sent with a message, but they are evaluated after any c-enabler or trading partner XPATH routers.
A c-hub administrator can specify multiple XPATH routers for the c-hub and for each trading partner using the c-hub administration console. These XPATH routers are stored in the repository. The c-hub administrator specifies the XPATH expression for each router, whether the router should reference the content in the application message parts, and the order in which the XPATH routers should be evaluated for each trading partner and for the c-hub.
The message-context XML document that the XPATH expression is evaluated against can be generated with or without expanding the child nodes of the <message-part> element. For large application messages expanding the <message-part> could have a significant performance impact. If an XPATH expression does not reference the content of a message-part then the administrator can specify that the router does not need the message-part to be included.
Each XPATH router expression in the ordered sequence can be configured to replace previous routers (that is, it overrides any routing done by a previous router in the sequence), or it can be configured to append a subset of trading partners to the previous router results. The c-hub XPATH routers are evaluated after all of the sending trading partner XPATH routers.
XPATH Filters
Trading partner XPATH filters 416—These are filters that are defined by an administrator and associated with a trading partner. Using the c-hub administration console, the administrator defines a filter with an XPATH expression and associates it with a trading partner. When the c-hub routes a message to a trading partner with the XOCP protocol, the XPATH filter is used to examine the message context and determine whether to send the message to the trading partner or not. The XPATH expression is evaluated against a message-context XML document generated by the XOCP filter module. It contains information extracted from the message context and the repository. If the c-hub evaluation of the XPATH expression returns false, then the c-hub does not route the message to the target trading partner. Otherwise, the c-hub processes the message as usual.
C-hub XPATH filters 418—These are filters that are defined by an administrator and are associated globally with the c-hub for all XOCP messages. Using the c-hub administration console, an administrator defines these filters with an XPATH expression. When the c-hub routes a message to a trading partner in an XOCP protocol c-space, the XPATH filter is used to examine the message context and determine whether to send the message to the trading partner or not. The XPATH expression is evaluated against a message-context XML document generated by the XOCP filter module. It contains information extracted from the message context and the repository. If the c-hub evaluation of the XPATH expression returns false, then the c-hub does not route the message to the target trading partner. Otherwise, the c-hub processes the message as usual.
An administrator can specify multiple XPATH filters for each trading partner and for the c-hub. The c-hub administrator uses the c-hub administration console to specify the order in which the filters should be evaluated for each trading partner and for the c-hub. The filters for the c-hub are evaluated after the filters for a target trading partner. Each filter has an XPATH expression and an indication as to whether to expand the <message-part> element of the message-context XML document.
Each XPATH expression in the configured sequence must evaluate to a Boolean true or false result. If the c-hub evaluation of an XPATH expression evaluates to false, then the c-hub does not send the message to the target trading partner and no further filtering is performed. Otherwise, the c-hub continues to process the XPATH filters until either a subsequent filter in the sequence returns false, or until all filters return true.
The message-context XML document generated for a c-hub or trading partner XPATH filter is almost identical to the document generated for an XPATH router. The only differences are that the context attribute on the root element is identified as “hub-filter” or “trading-partner-filter”, and the set of target <trading-partner> nodes is always just a single occurrence.
Bridges, Gateways and Proxies
Since the invention is designed to be as flexible as possible in facilitating enterprise-wide e-commerce collaboration, the collaboration system may support a number of complex scenarios that extend the reach of the collaborative workflow beyond that described earlier. These scenarios include the use of bridges, gateways, and proxies.
RosettaNet Protocol Router
A RosettaNet router, using the business protocol plug-in mechanism, allows RosettaNet messages to be routed by the Collaboration Server hub between any two participating RosettaNet clients. The clients supported include “native RosettaNet” clients like Vitoria and webMethods.
An example of a RosettaNet router is shown in
The RosettaNet router supports the Vitoria and webMethods implementations of external RosettaNet clients. In order to perform proper routing, the RosettaNet router needs information from the Preamble and Service Header XML components of the RosettaNet object in order to conform to the RosettaNet specifications. This is explained later in the Message Processing section. The out-of-the-box RosettaNet router is intended to comply with the published RosettaNet standard (as detailed in the RosettaNet specification at www.rosettanet.org, herein incorporated by reference) and is not biased towards any one vendor.
When an RNO (RosettaNet message) is received by The Collaboration Server and sent to the decoder portion 404 of the business protocol for processing the message the following happens:
The RNO format is not changed:
The digital signature is not validated. Validation of the digital signature is expected to be done by the receiving trading partner.
Read-only access is required to the document in order to retrieve the information needed for routing purposes. The generic c-hub RosettaNet router does not modify the document. However, customer-provided Logic Plug-Ins or business protocol plug-ins might do so.
All of the XML components are processed: the Preamble, the Service Header, and the Service Content.
Because of the above processing, the following information must be valid for the routing to proceed:
RosettaNet Implementation Framework (RNIF) version
RNO content length
MIME multipart structure, including:
Content-Type
Boundaries
Valid XML for the Preamble
Valid XML for the Service Header
Well-formed XML for the Service Content
RosettaNet Object Routing and Conversation Information
Routing and conversation information is constructed from parsing the RNO's XML parts. The c-space is determined based on the URL used to access the c-hub. The remaining RNO fields must be valid in form and in how they are used:
//Preamble/VersionIdentifier: The RNIF version number. The only RNIF version that The Collaboration Server will support is “1.1”.
//ServiceHeader/ProcessControl/ProcessIdentity/GlobalProcess IndicatorCode: The PIP identifier.
This typically has a value like “3A4” and it is used as the conversation name. It must be entered as such in the repository.
//ServiceHeader/ProcessControl/ProcessIdentity/VersionIdentifier: The version of the PIP specification used. This typically has a value like “1.1” and should not to be confused with the RNIF version number. It is used as the version number of this conversation. It must be entered as such in the repository.
//ServiceHeader/ProcessControl/TransactionControl/PartnerRoleRoute/fromRole/PartnerRoleDescription/GlobalPartnerRoleClassificationcode: The “from role” or the role of the sender. This typically has a value like “Buyer” and it is used as the role of the sender in the conversation. It must be entered as such in the repository.
//ServiceHeader/ProcessControl/TransactionControl/PartnerRoleRoute/toRole/PartnerRoleDescription/GlobalPartnerRoleClassificationcode: The “to role” or the role of the recipient. This typically has a value like “Seller” and it is used as the role of the recipient in the conversation. It must be entered as such in the repository.
//ServiceHeader/ProcessControl/TransactionControl/[ActionControl|SignalControl]/PartnerRoute/fromPartner/PartnerDescription/BusinessDescription/GlobalBusinessIdentifier: The business identifier of the sender of this message. For the unique business identifier, RosettaNet specifies use of the 9-digit DUNS number. It must be entered as such in the repository. The business identifier is used to identify the sending trading partner in the conversation.
//ServiceHeader/ProcessControl/TransactionControl/[ActionControl|SignalControl]/PartnerRoute/toPartner/PartnerDescription/BusinessDescription/GlobalBusinessIdentifier: The business identifier of the recipient of this message. It is similar to the “fromPartner” business identifier. The business identifier identifies the recipient trading partner in the conversation.
//ServiceHeader/ProcessControl/ProcessIdentity/InstanceIdentifier: A supposedly unique alphanumeric identifier for this business process, though RosettaNet also states that the initiating partner's business identifier (DUNS number) should be used. These two numbers are combined to form the conversation id, but are only used for monitoring and auditing purposes, not for routing purposes.
//ServiceHeader/ProcessControl/ProcessIdentity/initiatingPartner/GlobalBusinessIdentifier: The initiating partner's business identifier (DUNS number). As stated earlier, this is used to construct the conversation id.
//ServiceContent/thisDocumentIdentifier/ProprietaryDocumentIdentifier: The document identifier. This is the initiator.
//ServiceContent/thisMessageIdentifier/ProprietaryMessageIdentifier: This document identifier. This is a follow-on document in the same dialog.
//ServiceContent/receivedDocumentIdentifier/ProprietaryDocument Identifier: The initiating document identifier.
When decoding the RosettaNet message, an additional check can be made that the DUNS identifier of the sender, as retrieved from the message, matches that of the certificate. Since the c-hub is acting as a trusted intermediary, this will validate that the sender is who is claimed before the message is passed on to the recipient. If the match fails, the message will be rejected by the c-hub.
The collaboration server processing layers and business protocols neither generate nor validate any digital signature associated with the message because this is an issue among the various trading partner clients, whereas the c-hub is simply relaying messages. The c-hub only validates based on the SSL certificate.
If a particular business protocol is passing messages that have a digital signature, then the operations that a Logic Plug-In can perform on those message will generally be restricted to read-only unless the Logic Plug-In is allowed to re-sign messages on behalf of the original sender (or has access to the original sender's certificate/private key). However, nothing prevents a hub-provider from introducing Logic Plug-Ins that can validate or generate digital signatures.
Replying to the RosettaNet Sender
RosettaNet is a point-to-point protocol, and not prepared to interact with an intermediate hub. There are two main ways to approach the hub issue with RosettaNet—Synchronous and Asynchronous Message Transfer.
Synchronous Message Transfer
In this approach, the sender delivers a message to the c-hub, and the message passes all the way through the c-hub to the recipient, who returns the HTTP status code. The transport layer then returns that status to the layer initiating the outbound request, which passes it back through the various c-hub layers and eventually to the sender.
Asynchronous Message Transfer
This approach can also be referred to as a “delegated responsibility” model. The c-hub already needs to do about the same level of initial validation that the recipient would need to do. Enough of the lengths and first two XML components need to be checked that the message should make it to the upper protocol-processing layer of the recipient. In this case, it should be acceptable for the c-hub to send back the OK/BAD response after its initial processing. All responses may be logged.
An advantage to this approach is that the message persistence capabilities of the c-hub allow delivery to a recipient that is temporarily unavailable. Many of the problems that may be encountered by the RosettaNet recipient would already have been caught and reported by the c-hub. If the RosettaNet client starts processing the message and encounters errors at the business rule level (message content), a separate out-of-band error reporting mechanism is already provided by RosettaNet, and this can be routed through the c-hub like any other RosettaNet message.
If the c-hub thinks a message is not valid that the recipient might accept, then the hub-provider will have to determine which party is non-compliant, the sender, the c-hub, or the recipient, and take appropriate action. For example, if it was the c-hub, the RosettaNet router could be modified for different behavior (presumably being less strict in compliance checking).
If the recipient rejects a message that the c-hub accepted, again the hub-provider will probably have to get involved to determine the non-compliant party. However, the recipient would generally respond to the c-hub with a BAD REQUEST message, which the c-hub can log and pass back a RosetttaNet “General Exception” RosettaNet PIP to the initiator with the information.
XML Processing
The RosettaNet router needs important pieces of information from the Preamble and Service Header. This will be parsed with a validating XML parser. An error here will be logged and returned to the sender as a rejected message.
To properly function, the RosettaNet router does not need to fully parse and validate the Service Content portion. The Service Content is the business-specific portion of the message and is more appropriately handled by the receiving client. The information obtained here is for monitoring purposes. Furthermore, to validate the Service Content would require storing over 100 additional RosettaNet PIP DTDs, not including variations that might enter due to trading partner specific customization. The Service Content will be parsed as a “well-formed” document in order to retrieve the few non-essential fields required. If this part is not well-formed, however, the message will be logged and rejected by the c-hub.
For all of the XML parts, specifying an “XPath” expression for the field can retrieve the value for that field. For the Service Content part, the root node should be specified as “ServiceContent”, as shown earlier. It will be replace by the appropriate message-specific root node when evaluating the XPath expression.
The Logic Plug-In mechanism does not preclude the customer from doing their own XML parsing and processing of the Service Content, however. This would allow for filtering or monitoring based on the content of the Service Content portion of the message. The customer will be responsible for the parsing, however, in order to do the filtering.
The RosettaNet router requires certain fields to be entered in the repository, typically via the administration console. The required additional information includes:
DTD to be able to parse the RosettaNet Preamble
DTD to be able to parse the RosettaNet Service Header
DUNS for the trading partners participating in RosettaNet conversations. The “Trading Partner Protocol” repository item mentioned earlier is where this information is stored, and can be entered via the administration console.
Appropriate conversation and subscription information for the trading partners. The conversation information must match the conversation information specified in the RNOs being exchanged, as described earlier.
UML Configuring of C-spaces
Extensible Markup Language (XML) is rapidly becoming a key enabler for EAI and B2B e-commerce. In most EAI and B2B scenarios, the greatest challenge is creating a dynamic free flow of information and knowledge between applications and enterprises. XML provides an open and flexible message format, or common dialect, for structuring information flow between heterogeneous systems.
XML is a metadata language, a language used to define other languages and a universal standard for structuring data. XML is intrinsically extensible and self-describing, and thus supports an extremely flexible and dynamic business environment. XML easily enables the transfer of information across the Internet and between organizations, allowing them to communicate and conduct Dynamic B2B. Dynamic B2B requires technical underpinnings that adapt to changing e-business strategies and can be extended across multiple enterprises, business partners, and diverse applications.
In one embodiment of the invention, using Rational Rose, ac-space can be constructed using the unified modeling language (UML).
The Message Type System is a polymorphic hierarchy of message types. A message type is an abstraction of information that will be shared by transactors (e.g. ORDER, CUSTOMER, PRODUCT etc.). All message types share some common behavior, such as how the encapsulated information (XML) can be manipulated. Therefore the type system implements basic manipulation capabilities (create, read, update, delete) on the base level. Communication Adapter is a notion that abstracts a wire-protocol, such as HTTP, SOAP etc. When a transactor wants to communicate with somebody over a network connection, it needs to instantiate an appropriate adapter object and then pass it's reference to the message object. The content of the message object then becomes a payload of the network message and the adapter takes care of the communication protocol (headers, exceptions, timeouts etc.).
Message factories can be used to create message objects out of incoming network messages. Factories are specific to wire-protocols, not to message types. So, for example HTTPRequestFactory can handle any message that comes over HTTP. If a new message type is added, the existing factory can be leveraged. If support is added for a new wire-protocol, a corresponding factory would be implemented as well.
This embodiment implements a simple workflow engine, which is based on the idea of DFA (deterministic finite automata). The engine is driven by a XML-based flow-language. The schema (DTD) of the language looks that of Listing 2.
The workflow model is de-centralized, so each transactor executes it's own instance of the engine. The implementation manages multiple contexts simultaneously, so a transactor can be involved in many (but separate) business transactions at the same time. XML is also used for storing the state of a workflow context. This allows this embodiment to re-load the state of a context in a different system. This design has some interesting consequences, for example the system should scale nicely in the WLS cluster environment.
The design process is started by modeling business transaction using Use Case diagram. For the example illustrated in
The first step is to break the Use Case into actions, and add those to an activity diagram.
The next step is to take the roles from a use case diagram, map those roles to swim lanes of an activity diagram and assign the activities to corresponding swim lanes.
The next step is to define messages that will be delivered and processed by activities.
Configuration files (publish/subscribe), message types etc., can then be generated from those diagrams. It is also possible to model each transactor's DFA by using state chart diagrams and to generate the necessary workflow definitions out of those:
Actions
Message Types
Flows
Referring now to
The invention helps to define an open market platform system for conducting business-to-business commerce or the Web. The collaboration system defined herein includes many novel and industrially applicable features, some of which are listed here.
Mainframe Connectivity
There are tens of thousands of IBM CICS applications hosted on the IBM MVS and VSE operating systems, and deployed on mainframes in every country around the world. Hundreds of the world's largest databases are accessed through IMS on IBM's MVS S/390 platforms. Unlike most EAI vendors, the collaboration system provides the greatest, most robust connectivity to mainframe applications. The collaboration system provides connectivity to the following: TCP, SNA and OSI TP.
Adapter Integration for Major Packaged Applications
The collaboration system products minimize programming, which greatly reduces the development effort required to connect applications. The emphasis is instead on configuration, which not only saves programming costs, but also enables developers to focus on building new e-business applications that will significantly improve their company's competitive edge.
Support for XA-Transactions
The collaboration system provides transparent support for XA-transactions. Message-based EAI products define a transactional unit of work to mean only the guaranteed delivery of a message. However, if a unit of work between integrated applications translates into multiple reads and writes, or if the integrated applications are heterogeneous, there is a strong risk of corrupt or inaccurate data, which can only be avoided by complex programming. This risk is eliminated if applications support XA-transactions and are integrated using the collaboration system because the collaboration system includes a full standards-based transaction manager capability.
E-Transaction
An E-Transaction embodiment of the invention enables businesses to jointly and successfully accelerate e-commerce strategies in real time, empowering dynamic business-to-business transactions.
The invention enables the participation by a business in the process of conducting e-commerce on the Internet with other business collaborators. This approach is significantly different from back-end business transaction systems. First, the “transactions” are of a business nature and are long running. Indeed, a business transaction may require weeks, months, or even years to complete. The state for such transactions must be kept persistently so as not to lose it during various system failure modes. Thus, a single business transaction can cause one or many back-end database transactions to occur. Traditional business-to-business e-commerce required businesses to agree on a business model for the relationship, a uniform vocabulary, and a messaging technology. With the present invention, that is no longer necessary—the platform is independent of any one business model or vocabulary and uses XML as the standard.
Zero-Latency and End-to-End Integration
The present invention provides an enterprise-class solution in the market that provides zero-latency, end-to-end integration. From the mainframe to the Web, the collaboration system provides a proven, scalable, and pre-integrated server-based environment for integrating an enterprise's packaged and proprietary enterprise applications, whether they reside on the back-office, front office, or on the Web. The collaboration system performance excels in the most demanding, distributed 24×7 environments, and offers a choice of messaging paradigms and pre-packaged application adapters and custom Adapter Development Kit to best solve the business problem at hand. The collaboration system platform migration to XML-based messaging keeps the integration solutions at the forefront of technology.
Other features, aspects and objects of the invention can be obtained from a review of the figures and the claims. It is to be understood that other embodiments of the invention can be developed and fall within the spirit and scope of the invention and claims.
The foregoing description of preferred embodiments of the present invention has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations will be apparent to the practitioner skilled in the art. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling others skilled in the art to understand the invention for various embodiments and with various modifications that are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalence.
This application claims priority from U.S. provisional patent applications “OPEN MARKET PLATFORM SYSTEM FOR CONDUCTING WEB BASED BUSINESS”, Application No. 60/183,067, filed Feb. 16, 2000, incorporated herein by reference and “OPEN MARKET PLATFORM SYSTEM FOR CONDUCTING WEB BASED BUSINESS”, Application No. 60/258,804, filed Dec. 29, 2000, incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5321841 | East et al. | Jun 1994 | A |
| 5446842 | Schaeffer et al. | Aug 1995 | A |
| 5463555 | Ward et al. | Oct 1995 | A |
| 5630069 | Flores | May 1997 | A |
| 5634127 | Cloud et al. | May 1997 | A |
| 5704017 | Heckerman | Dec 1997 | A |
| 5748975 | Van De Vanter | May 1998 | A |
| 5835769 | Jervis et al. | Nov 1998 | A |
| 5836014 | Faiman, Jr. | Nov 1998 | A |
| 5913061 | Gupta | Jun 1999 | A |
| 5926798 | Carter | Jul 1999 | A |
| 5930512 | Boden et al. | Jul 1999 | A |
| 5931900 | Notani | Aug 1999 | A |
| 5941945 | Aditham | Aug 1999 | A |
| 5960404 | Chaar et al. | Sep 1999 | A |
| 5961593 | Gabber et al. | Oct 1999 | A |
| 5983214 | Lang et al. | Nov 1999 | A |
| 5995945 | Notani | Nov 1999 | A |
| 6018716 | Denardo et al. | Jan 2000 | A |
| 6044217 | Brealey et al. | Mar 2000 | A |
| 6052684 | Du | Apr 2000 | A |
| 6064980 | Jacobi et al. | May 2000 | A |
| 6067623 | Blakley, III et al. | May 2000 | A |
| 6073109 | Flores et al. | Jun 2000 | A |
| 6073138 | de l'Etraz et al. | Jun 2000 | A |
| 6119149 | Notani | Sep 2000 | A |
| 6122666 | Beurket et al. | Sep 2000 | A |
| 6157915 | Bhaskaran et al. | Dec 2000 | A |
| 6212548 | DeSimone et al. | Apr 2001 | B1 |
| 6226675 | Meltzer et al. | May 2001 | B1 |
| 6230287 | Pinard et al. | May 2001 | B1 |
| 6338064 | Ault et al. | Jan 2002 | B1 |
| 6353923 | Bogel et al. | Mar 2002 | B1 |
| 6397192 | Notani et al. | May 2002 | B1 |
| 6401111 | Dan et al. | Jun 2002 | B1 |
| 6594693 | Borwankar | Jul 2003 | B1 |
| 6604198 | Beckman et al. | Aug 2003 | B1 |
| 6732237 | Jacobs et al. | May 2004 | B1 |
| 6795967 | Evans et al. | Sep 2004 | B1 |
| 6802000 | Greene et al. | Oct 2004 | B1 |
| 6804686 | Stone et al. | Oct 2004 | B1 |
| 20010037367 | Iyer | Nov 2001 | A1 |
| 20020016759 | Macready et al. | Feb 2002 | A1 |
| 20020078365 | Burnett et al. | Jun 2002 | A1 |
| 20020165936 | Alston et al. | Nov 2002 | A1 |
| 20030018665 | Dovin et al. | Jan 2003 | A1 |
| 20030041198 | Exton et al. | Feb 2003 | A1 |
| 20030110117 | Saidenberg et al. | Jun 2003 | A1 |
| 20030131050 | Vincent | Jul 2003 | A1 |
| 20040015548 | Lee | Jan 2004 | A1 |
| 20040034687 | Friedman | Feb 2004 | A1 |
| 20040078435 | Dunbar et al. | Apr 2004 | A1 |
| 20040078455 | Eide et al. | Apr 2004 | A1 |
| 20040133660 | Jumghuber et al. | Jul 2004 | A1 |
| 20040136511 | Schwartz et al. | Jul 2004 | A1 |
| 20040145608 | Fay et al. | Jul 2004 | A1 |
| 20040158610 | Davis et al. | Aug 2004 | A1 |
| 20040225995 | Marvin et al. | Nov 2004 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20020010741 A1 | Jan 2002 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60258804 | Dec 2000 | US | |
| 60183067 | Feb 2000 | US |
