Tracking Service-Oriented and Web-Oriented Architecture

SOA & WOA Magazine

Subscribe to SOA & WOA Magazine: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get SOA & WOA Magazine: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

SOA & WOA Authors: TJ Randall, Lori MacVittie, Andreas Grabner, Dynatrace Blog, Cynthia Dunlop

Related Topics: Java EE Journal, Apache Web Server Journal, SOA & WOA Magazine

J2EE Journal: Article

The 'United Nations' of the i-Technology World?


Related Links:
  • JDJ Exclusive Interview with David Skok

    Anyone in the i-technology world engaged in developing, deploying, integrating, or managing software applications knows Borland Software Corporation - BORL as it's known on the NASDAQ - to be the company that above all aims to let clients deploy online applications that are compatible with different platforms.

    This gives Borland's president and CEO, Dale Fuller, a unique vantage point from which to comment not just on Java or .NET, but on all manner of current technologies. Fuller also recognizes, along with JDJ, that these days "Everyone from the CIO through to the developer is in the business of software."

    Accordingly, this month's JDJ "Question & Answer" session gave Fuller an opportunity, from his corporate world headquarters in Scotts Valley, CA, to deal with a range of issues from ASPs and Web services to SOA, Linux, application life-cycle management, "invisible middleware," and Borland's future role as IT's nearest equivalent to the United Nations - serving all those in the world business of software, no matter what particular brand they may owe allegiance to.

    JDJ: We have the basic building blocks and standards in place for Web services, and people are using them. Where do you think we will go from here?
    Dale Fuller: We'll need tools that will allow us to put the Web services standards and building blocks to use in a higher-level way. Web services lend themselves to a more business process-oriented approach toward applications and we are going to require development tools that work in this more process-oriented way. This means tools that can visualize the solution at a higher level - think UML models, but UML models that serve as more than just a blueprint, and instead become the basis for the application itself. The software industry must go beyond SDKs and toolkits and toward a higher-level way of building and integra-ting applications.


    JDJ: Whatever happened to confuse the meaning of Web services? Do you think that it's because Microsoft focused at the beginning on Web services from the consumer, individual point of view rather than from the back-end IT point of view?
    DF: Looking at the origins of the Web services story, you'll see that the "marketing" side didn't embrace the terminology that IT and developers were accustomed to in distributed computing.

    The terms that we all understood at the time were things like remote invocation, request brokers, distributed objects, call backs, functions, and APIs. When Web services was introduced, there was a significant emphasis on the "services" aspect of this new technology opportunity. This made Web services appear to be something more magical and oriented, rather than a simple and practical technology for building networked and distributed applications.

    This probably had to do with the timing of the emergence of the technology; remember at the time the SOAP specification emerged, the hot prospect of the day was application service providers (ASPs). I think many in the industry saw a natural fit and wanted Web services to ride the coattails of that wave.

    JDJ: Does Borland see a role for Linux within Web services?
    DF: Wherever there is a Web server there is a role to host Web services, and there are certainly a lot of Apache Web servers running on Linux.

    JDJ:'s CEO Marc Benioff prefers not to use the term "Web services" at all but refers to the distributed application architectures that we've all been pursuing for years as "client/service." Do you agree? Should we be talking about "client/service" architectures now?
    DF: To the extent that "Web services" describes just one of the possible benefits of the technology, the title can be seen as a misnomer. It is certainly a good idea to talk about the other benefits of Web services such as "client/service," "peer to peer," and "business to business." However, over the last five years the term Web services has become pretty entrenched in IT vocabulary for a simple and fundamental idea, the practical applications of which are well beyond a single architecture.

    JDJ: How do we straighten things out so that from here on technology buyers can understand it all?
    DF: If the software industry were to talk about Web services as the simple messaging protocol it was initially designed to be, more organizations would begin to see the potential benefits for using Web services in their businesses. This could very well be a case where actually talking about technology could simplify the mystery of the higher-level marketing that has dominated the technology since its inception.

    JDJ: What about the buzzword of 2004, SOAs - do you agree with those who say that enterprise adoption of service-oriented architecture will take many years? If so, then in the meantime what should programmers be doing?
    DF: SOA is happening as we speak and IT organizations are quickly realizing that the business advantages of SOA are simply too large to be ignored. The need to quickly respond to changing market and organization conditions and to maximize business agility is the key drive behind service architectures. Quality, reliability, and maintainability are main concerns for CIOs today as they look at their software systems. However, shifting to SOA requires greater architectural discipline across the entire organization and throughout the application life cycle. To be successful at it, IT groups cannot work in isolation; strong team collaboration is required to define, implement, and manage SOA. SOA also incurs greater performance overhead, a risk for systems that have strictly enforced response times. This requires closer performance monitoring and a secure, scalable, distributed computing environment.

    JDJ: We heard reliable rumors at JDJ when "Tiger" was unleashed earlier this year that it went through unanimously, but that three of the heavyweight companies among the 14 members on the JSR committee concerned were complaining that there's not enough CORBA support. Does Borland have a position on CORBA support in Java?
    DF: The CORBA ORB capabilities offered within J2SE provide basic support for such technologies as the Internet Inter-ORB Protocol (IIOP), the communication foundation with CORBA. While J2EE uses Remote Method Invocations as its communication foundation, the EJB specifications within J2EE define that it is IIOP that must be used in order to access EJBs.

    This hasn't changed even with the latest J2SE 1.5 release. As organizations implement CORBA servers throughout their enterprise, they must rely on enterprise-grade CORBA ORBs that provide extensive CORBA services beyond those offered within J2SE/J2EE. Some of these services include security, transactions, notification, and messaging as well as support for other enterprise grade qualities of service such as failover, load balancing, and clustering.

    JDJ: Still on Java, what's your position on the JCP - is it the right way of doing things?
    DF: Borland is committed to the Java Community Process and to supporting Java standards. We believe that the JCP was the right model for a specific time in the evolution of Java technology. Sun maintained "executive" control as a lever to keep the technology evolution moving quickly. However, as Java technology matures, it could be time for Sun and the JCP to look at an increasingly participative, democratic model that is more reflective of open standards.

    JDJ: Would/could anyone else safely be the custodian of Java? The open source community, for example?
    DF: We may have reached a stage in the evolution of Java technology where it is much less vital to have a "custodian" as we have had in the past.

    JDJ: While on the subject of open source, what overall effect on Java do you foresee from the compelling economics of Linux?
    DF: Linux is driving the adoption of low cost, open source servers in the marketplace. Java is viewed as the de facto application platform for the building and deploying of applications on top of that operating system. You could see Linux driving increased demand for J2EE/JSP/ servlet-based applications.

    JDJ: Do you think J2EE is too complex? If so, what's the best way forward?
    DF: It isn't a question of J2EE being too complex. J2EE is a layer between the complexity of network-based utilities, such as DB, directory, security, and transaction processing.

    What's necessary is the ability to provide the developer with a better sense of "context" as to where he or she is in the process of building new applications that plug into an existing infrastructure or in the process of deconstructing and rebuilding existing applications.

    JDJ: Why do you think Sun's Jonathan Schwartz said to JDJ in February: "Middleware is history"? Is middleware in fact just beginning? Or is Schwartz right in saying that end-to-end "systems" will supplant it?
    DF: Borland has always had a strong relationship with Sun, but on this point, we'll have to lean toward the IBM billboards that dot Highway 101 in northern California. Their latest slogan is: "Middleware is everywhere. Can you see it?"

    Sun provides an integrated "Java Enterprise System" that offers all of the basic components of shared services, from directory and identity management to Web services, e-mail, and clustering. While Borland does offer an application life-cycle management solution for Java, we prefer to see ourselves as the Switzerland of software development - providing our customers with the freedom to choose from a variety of vendors' technologies depending on what works best in their organization's IT environment.

    JDJ: As a major business leader, what have you found to be the most compelling aspect of the technology space? And what's the least attractive aspect?
    DF: Most compelling: the fact that who we define as a software person today is drastically changing to recognize that everyone from the CIO through to the developer is in the business of software. The fact that the pace of change and innovation is forcing globalization and making the world smaller by bridging geographic and cultural divides. The fact that companies like Borland who helped drive the PC evolution are still making software pervasive in all that we do today - such as the use of our technologies in projects as broad as St. Judes Children's Hospital and the NASA Mars Rover Mission.

    Least compelling: when a vendor's self interest gets in the way of freedom and innovation.

    JDJ: After such great success with the notion that Borland is "the developer's Switzerland," will there ever be a need for you to choose - once and for all - between J2EE and .NET, for example?
    DF: The trend is definitely toward greater heterogeneity. The analyst groups predict that moving forward it will be a two-horse race between J2EE and .NET. Borland would have to agree, because at the end of the day, it is about having the right technology for the problem at hand. Borland exists to ensure that our customers have that freedom of choice.

    JDJ: If ever you abandoned the "Swiss" metaphor, what other country could Borland usefully emulate and invoke to characterize its unique role?
    DF: I would have to say that we wouldn't be a country at all - we'd have to be seen more like the United Nations. We don't make the laws governing software development, but we do set out to help resolve conflicts between programming languages and break down the barriers between people and technology across the software development paradigm. We see ourselves as the end user's champion - one of the last bastions of independence in the industry, playing a key role in helping to formulate the policies and standards that affect the industry.

    Where does Borland stand in the i-technology world?
    Borland has solutions available today that provide management, control, and visibility to implement SOA across the entire organization. Our application life-cycle management strategy is designed to reduce risk and uncertainty, helping to ensure that developers are not working in isolation, but are quickly and reliably building solutions in line with business needs. Our goal is to make standards easier to use while avoiding vendor lock-in. And that's what's made us #1 in Java development.

    JBuilder has always found ways to make Java and J2EE easier to use. We continue to do that with JBuilder X. JBuilder X offers a visual struts designer that takes advanced Web development in Java to the next level of productivity. As new Java standards emerge, we will continue to make them easier to use, and we've taken a leadership role in extending that productivity throughout the entire application life cycle with solutions such as Borland Enterprise Studio for Java. This combines JBuilder development technology with our Together solution for modeling, Optimizeit for performance management, CaliberRM for requirements management, and StarTeam for change management, as well as Borland Enterprise Server and JDataStore for deployment.

    And products such as Borland Enterprise Server; Borland Deployment Op-Center; and Janeva for .NET, J2EE, and CORBA interoperability were designed to enable architectural heterogeneity.

    Related Links:
  • JDJ Exclusive Interview with David Skok
  • More Stories By Jeremy Geelan

    Jeremy Geelan is Chairman & CEO of the 21st Century Internet Group, Inc. and an Executive Academy Member of the International Academy of Digital Arts & Sciences. Formerly he was President & COO at Cloud Expo, Inc. and Conference Chair of the worldwide Cloud Expo series. He appears regularly at conferences and trade shows, speaking to technology audiences across six continents. You can follow him on twitter: @jg21.

    Comments (1) View Comments

    Share your thoughts on this story.

    Add your comment
    You must be signed in to add a comment. Sign-in | Register

    In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.

    Most Recent Comments
    Boaz Goldszmidt 04/12/04 03:31:08 PM EDT

    Fuller gives a glossy enough interview, but he really hasn''t answered anything in more than a superficial manner. Worse, he makes comments about processes he barely understands.

    First, it is interesting that he touts IBM''s "middleware is everywhere" line. Since he has essentially cut BORL''s middleware products off at the knees, he has no other choice but to believe in his understanding of that quote. The Borland Enterprise Server and VisiBroker are essentially at an EOL stage, with no innovation or features going into the platforms. BORL''s middleware story is integration with BEA, IBM, JBoss, etc. Yet BORL does not have a coherent SOA strategy, and an at-least-7-year history of being unable to successfully partner with other vendors, so it depends upon a motley array of ''middleware'' to pull it''s ill-assorted product line together.

    Another issue is that BORL is no longer the leader in Java development space. JetBrains has eaten BORL''s lunch in the Java IDE space. The various open source frameworks and their packaged offerings are all more feature rich and better integrated than BORL''s tools. BORL pursues MSFT''s philosophy of integration between products. The problem with this is that it requires solid presence and infrastructure. BORL is a small fish with neither of those qualities.

    Most surprisingly, Fuller clains that Sun and the JCP need "to look at an increasingly participative, democratic model that is more reflective of open standards." How so? The JCP already provides a highly participative and democratic model which is reflective of open standards. Like any democratic society, interested parties need to participate and that takes work. Since BORL began divesting itself from standards-based architecture and design, starting in 2003, it''s participation in JCP intiatives has been minimal. Same can be said for it''s participation in the OMG. BORL is no longer in the standards-based business, and no longer cares about these standards, so Fuller has no one to blame other than himself if he feels the JCP is not participative or democratic enough.

    This is particularly troubling for BORL as it is trying to compete in a space for which neither Fuller nor his management team have any core competencies. The market is pushing for a greater level of standards support and compliance, and smoother integration, not less. BORL has everything to lose by moving backward and resisting the current. There is no logical reason for BORL to do this. A year ago it still had the resources to pursue the market strategy; yet most of these resources have since been squandered and cut. When Fuller takes a multi-million dollar pay raise for 2003 at the expense of common sense, it makes things look even worse for BORL.