Brent Hailpern  Brent Hailpern photo         

contact information

Scientific Director, AI Horizons Network
Head of Computer Science, IBM Research - Distinguished Research Staff Member


OOPSLA 2001 Workshop on Object-Oriented Web Services

Supporting the Development, Deployment and Evolution of Web Services
Monday, 15 October 2001 - Tampa, Florida, USA

Brent Hailpern, Associate Director of Computer Science, IBM Research
Don Batory, Professor, University of Texas at Austin
Richard P. Gabriel, Distinguished Engineer, Sun Microsystems
Aditya Mathur, Professor, Purdue
Peri Tarr, Research Staff Member, IBM Research
Sanjiva Weerawarana, Manager, Component Systems, IBM Research

Abstract: XML has started a revolution on the web: from static formatted content to dynamic self-defining information with real semantics. The addition of semantics is only the first step to providing real applications, known as "web services", to Internet users. The first generation of web-service infrastructure is in development, allowing one web service to issue a request to another and to register/describe/find a service to use (e.g., SOAP, WSDL, UDDI).

Eventually, web services will become electronic utilities that are delivered to end users over the internet-representing a critical new application domain in electronic commerce. Like traditional utilities, such as electricity, web services will be metered and customers will pay for their use. The terms of use (called service level agreements or SLAs) of the services will include functionality, performance, and reliability. The need to dynamically monitor and control SLAs and to respond to changing needs and resources, imposes some challenging requirements on the design, development, deployment, and evolution of web services.

This workshop explored issues in, and new technologies and methodologies to support the engineering and deployment of, web services.

Workshop Presentations:

Download all workshop papers (zip file)

A Framework for Facilitating Dynamic e-Business Via Web Services: Rama Akkiraju, David Flaxer, Henry Chang, Tian Chao, Liang-Jie Zhang, Frederick Wu, and Jun-Jang Jeng, IBM T.J. Watson Research Center

Abstract: While Web services provide an essential building block for realizing dynamic e-business by facilitating standards-based program-to-program interaction, they do not address many business aspects of communication such as security, access control, business partner selection, service level agreement monitoring, and logging. In this paper, we propose a service-oriented framework that provides these much needed infrastructure services. Using the framework, service providers can prepare their business applications for publishing as Web services, and specify access controls on their services in the registry while service requestors can find suitable service providers and invoke their services dynamically. The framework ensures that all transactions are secure, monitored for service level agreements and logged for audit purposes. By providing business utility services and automation technology founded on Web services, we believe that, our framework takes Web services to the next level in supporting dynamic e-business.

Coordination Technologies for Web-Services: Luís Filipe Andrade (ATX Software S.A.) and José Luiz Fiadeiro (ATX Software S.A. and University of Lisbon)

Abstract: Based on the identification of some shortcomings of object-oriented methodol-ogy and technology to address the challenges of supporting the engineering and deployment of Web Services, we suggest that alternative approaches can be found in what we call coordination methodologies and technologies -- a set of modelling primitives, design principles, design patterns, and analysis techniques that we have been developing for supporting the construction and evolution of complex software systems that need to operate in very volatile and dynamic environments.

On the Integration of Heterogeneous Web Service Partners: Francisco Curbera, Ignacio Silva-Lepe and Sanjiva Weerawarana, IBM T.J. Watson Research Center

Abstract: A new distributed computing model for the Web is now being defined as a result of several ongoing standardization initiatives. This effort is driven by the need to enable open business to business (B2B) interactions on the Web, and is centered around the development of new vocabularies based on the Extensible Markup Language (XML) standard. The Web Services Description Language (WSDL), the interface definition language of the new computing stack, introduces a new generic model of application interactions. This position paper maintains that, while WSDL promotes the use of both messages and operations, it falls short of supporting the full integration of Web Service interactions, at least as it pertains to the timing and coupling aspects of such interactions. We further maintain that this integration can be better understood and approached from the point of view of a generalized framework for Web Service interactions.

Web Services Invocation Framework (WSIF): Matthew J. Duftler, Nirmal K. Mukhi, Aleksander Slominski and Sanjiva Weerawarana, IBM T.J. Watson Research Center

Abstract: This paper proposes a framework that allows the application-programmer to program against an abstract service description, in a protocol-independent manner. We call this framework the Web Services Invocation Framework (WSIF). WSIF supplies a simple API to invoke Web Services, no matter how or where the service is provided, so long as the service is described in WSDL. WSIF enables the user to move away from the usual Web
Services programming model of working directly with the SOAP APIs, and towards a model where the user interacts with abstract representations of the services. This allows the user to work with the same programming model regardless of how the service is implemented and accessed. The framework also allows new bindings to be dynamically added, and current bindings to be dynamically replaced.

Validating Message Sequences in Web Services: Matthew Fuchs, Commerce One

Abstract: Web Services are a part of distributed collaborative processes among various parties. As such public specification of a Web Service requires information about the sequences of valid messages/methods a service can receive or will send. XLANG, WSFL, and ebXML's BPSS are attempts to specify this. One can view this, in part, as an extension of OO by viewing the set of valid sequences as specifying a language over the messages. It is important to consider which kinds of languages we will need. These languages have the additional complexity that they must model not just the set of acceptable invocations of a single participant, but of the interactions among the participants. We also need a type system covering both message parameters (traditional) and the valid sequences (less so).

Reliability of Composed Web Services: From Object Transactions to Web Transactions: Thomas Mikalsen, Isabelle Rouvellou and Stefan Tai, IBM T.J. Watson Research Center

Abstract: Multiple Web Services often need to be composed within some business process. Existing Web Service standards do not address reliability of such compositions. In this position paper, we argue that reliability can be achieved by adopting and extending existing advanced object transaction processing technology, like the OMG/J2EE Activity Service. We identify and discuss some important problems and research issues related to this approach.

Web Services: Why and How: Francisco Curbera, William A. Nagy and Sanjiva Weerawarana, IBM T.J. Watson Research Center

Abstract: Web services are a natural consequence of the evolution of the Web into an open medium which facilitates complex business and scientic application interactions. Web services are concerned with the problems of enabling systematic application-to-application interactions over the Web, and the integration of the existing network computer infrastructure into the Web. The key components of this work are a focus on interoperability, support for efficient application integration, and the creation of a uniform representation of applications within
heterogeneous distributed systems. The Web services model builds upon existing industry standardization efforts centered around the Extensible Markup Language (XML), and proposes a two pronged approach to deal with the interoperability requirements associated with heterogeneous systems: enable base interoperability using a small set of common protocols, and develop a uniform representation of network applications which are accessible using multiple communication protocols. The goal of Web services is to provide a flexible framework where universal interoperability does not preclude efficient integration.

Software Engineering for Web Services: A Focus on Separation of Concerns: Brent Hailpern and Peri Tarr, IBM T. J. Watson Research Center

Abstract: XML has started a revolution on the World Wide Web, moving it from static formatted content to dynamic, self-describing information with real semantics. The addition of semantics is only the first step to providing real applications, known as "web services," to Internet users. The first generation of web-service infrastructure is already under development, and it will allow one web service to issue a request to another and to register/describe/find a service to use (e.g., SOAP, WSDL, UDDI), thus creating interconnected sets of cooperating web service components.

eUtopia: A Framework for Managing Web Utility Services: Vijay Naik, Giovanni Pacifici, Asser Tantawi, AlaaYoussef (IBM T.J. Watson Research Center), Mangala Gowri, Ashish Kundu, and Pradeep Varma
(IBM India Research Lab)

Abstract: Today, the web services model defines roles (service provider, service broker, and service requestor) and their interactions (publish, discovery, and usage). The various interfaces are specified in a simple way that is based on ubiquitous protocols, language-independence, and standardized messaging. Such technical advantages, as well as a growing industrial support, have given rise to a proliferation of web services. However, most web services that are provided today are free and unmanaged. Nevertheless, it is envisioned that web services will play a key role in (dynamic) e-business. In this new business environment, services are expected to be dependable, secure, reliable, guaranteed, and profitable. We refer to a web service that satisfies such requirements as a web utility service (or utility, for short). Thus, the current web services model needs to be augmented with management functions such as usage metering, accounting, controlled access, dynamic resource allocation as well as service security, reliability and availability. The resulting utility model is realized in a web utility services platform (or utility platform, for short). This platform provides the necessary management functions to offer web services as utilities–that is, as web services that can be subscribed to, measured, delivered reliably and on demand. Such a platform will manage the various phases in the life cycle of a utility such as development, provisioning, and invocation. In this paper, we present eUtopia, a web utility services platform, where we focus on management aspects. Our platform uses and extends the existing web services technologies (SOAP, WSDL and UDDI) to transparently provide subscription, metering, authentication and SLA management functions. We define as web utility service (or in short utility) a web service that runs on our platform and that can be transparently metered, monitored and managed.

From E-Processes to E-Networks: an E-Service-oriented approach: Giacomo Piccinelli (Hewlett-Packard Laboratories) and Eric Stammers (Hewlett-Packard E-Process Operation)

Abstract: Looking at the recent history of information systems, two complementary trends seem to emerge. On the one side, systems become more modular. The shift is from tight integration to loosely coupled components. On the other side, the distance between business models and information technology (IT) is shortening. Aggressive business models impose new requirements on IT. At the same time, operational capabilities made available by IT drive the definition of new business models. Web services are the most noticeable outcome of the first trend. E-services play a similar role for the second trend. In this paper, we concentrate on e-services and their relationships with the type of business processes used by e-businesses. We refer to this type of processes as e-processes.
We first introduce e-services, and outline the interconnections with web services. Focusing on the
composition aspects of e-processes and e-services, we then discuss potential and requirements of processlevel cooperation for composite e-service solutions. We refer to these composite solutions as e-networks. The discussion is based on our work on the DySCo (Dynamic e-Service Composer) framework for enetwork management. A brief description of the framework and the associated platform is included, together with an analysis of some of the issues related to service level agreements (SLAs) in e-networks.

The need for SLA in web services: Frank Prothmann

Abstract: The current standards for web service leave out much of the stuff we've already done for other application that we expose to the public. Some requirements are new. This paper gives a short introduction in the field of SLA's and why we have a need for this in web service environment.

Requirements for Implementing Web Services Beyond XML Standards: Alok Srivastava, Oracle Corporation

Abstract: Web Services have re-introduced to the world the idea that software can be very useful if it did not look and feel so complicated to the end users. In order to bring web services in the same domain as utilities, the systems around the web services need to work the same way as utilities do. XML and XML based standards like SOAP, WSDL and UDDI provide a valuable means to connect web service producers to the consumers. The actual delivery and automation of the entire process comes after the connections have been made. In this paper, a number of requirements for such an environment have been discussed through a variety of use cases and scenarios. The goal of this paper is to bring forward several issues that need to be addressed in order to bring web services to a large-scale deployment and help start active discussions in this area.

Credential Sheets for Components: Judith A. Stafford, Software Engineering Institute

Abstract: If one is to predict properties of a component assembly before actually assembling the components, it is necessary to make explicit any component property that can impact the behavior of the assembly. Likewise, if one is to reason about the impact of replacing one component with another in an existing, perhaps running, system then one must have access to information about the properties of potential replacement components. There are several problems associated with predicting the behavior of component assemblies. Identifying which properties are important to make explicit, and determining how to measure these properties are addressed in other research efforts. In this position paper we addressed the specification of the component properties. We suggest the use of XML schema for defining credential sheets in which properties are described so that automated tools can be used to predict the behavior of web­based applications.

On the Management of Compositions of Web Services: Vladimir Tosic, Bernard Pagurek, Babak Esfandiari, Kruti Patel, Carleton University

Abstract: In this position paper, we present our work on Web services with multiple classes of service, called service offerings, and on management and dynamic (i.e., runtime) adaptation of their compositions. We explain the motivation for Web services with multiple service offerings, present some management and dynamic adaptation algorithms based on the manipulation of provided service offerings, and discuss
some of the issues with the corresponding management infrastructure, called DAMSC (Dynamically Adaptable and Manageable Service Components), that we are developing. We also briefly present our work on WSOL (Web Service Offerings Language) – an extension of WSDL (Web Services Description Language) that enables specification of various types of constraints— including functional, nonfunctional (QoS – Quality of Service), authorization policies, etc.— and specification of Web services with multiple service offerings. At the end, we summarize some of the challenges for future research in the area of management and dynamic adaptation of compositions of Web services.

ELPIF: An E-Logistics Processes Integration Framework Based on Web Services: Liang-Jie Zhang, Pooja Yadav, Henry Chang, Rama Akkiraju, Tian Chao, David Flaxer, Jun-Jang Jeng, IBM T.J. Watson Research Center

Abstract: In recognition of the increasing importance of globalization and the resulting need for greater, faster and more flexible communications, a framework is required to allow any company to establish itself in no time or make optimum use of their legacy applications and run efficiently with minimal cost input. This paper presents such a framework called ELPIF for e-logistics processes integration based on Web Services via incorporating (1) common alliance layer (2) adaptation layer; and (3) dynamic data binding mechanism. This framework can be adopted as a new service delivery model which uses a design pattern and solution templates. The interaction between the e-logistics processes and business process manager that orchestrates e-logistics processes in an e-business solution will be described in this paper. A transportation planning in the purchase order management process of a B2B solution is used as an example to illustrate the usage of ELPIF by encapsulating United Parcel Service (UPS) on-line XML Tools as Web Services.

This web site hosted by IBM Research
Page contact: Brent Hailpern