Standards can drive revolutionary changes in technology: consider the impact that SQL has had on the database market, or consider that the World Wide Web was launched by the combination of HTML, HTTP, URL, and SSL. My belief is that protocol standards (XML, Web services) and programming standards (Java and the. NET alternatives, XML Query, etc. ) will have a similarly profound impact on integration.
Integration encompasses a broad range of information technology (IT) needs: Enterprise application integration (EAI): Directly interconnecting two or more business applications over intranets (that is, behind corporate firewalls) Business-to-business integration (B 2 BI): Directly interconnecting the business applications of one company with those of a business partner across the Internet or a virtual private network User interface (UI) integration: Aggregating and personalizing UI across applications so that the human user experiences a single unified "Web flow" (e. g. , Java page flows within a portal), bringing together those disparate back-end systems Data integration: Aggregating data queried from disparate applications and data stores in order to define a unified view of a business object -- for example, creating a unified "customer" schema by combining data from a customer relationship management (CRM) system, an Enterprise Resource Planning (ERP) system, and so on Modeling Using Java and Xml for Web Applications Today Today, the integration market remains largely fragmented across proprietary solutions, none of which scale to meet the needs of our Web-oriented world. It is also important that proprietary integration technologies generally require the user to run a particular vendor's software on all participating nodes. Convincing a worldwide company and all of its business partners to run the same proprietary integration software just will not fly.
At the same time, integration remains very expensive -- by various measures consuming 70+% of discretionary IT funding, and remains sufficiently complex that organizations tend to only integrate applications as it becomes a business imperative. Clearly, the industry must do better. We are convinced the answer lies in extending the Web from the UI platform it is now (browser to data / application ) to become an integration platform. Today, most new applications are "Web ready" out of the box -- i.
e. , designed to support a Web browser front end. Our claim is that future applications will also be "Web integration ready" out of the box -- that is, ready to plug into this emerging "back plane" of eXtensible Markup Language (XML) and Web services that will enable them to seamlessly interconnect with other Web-based applications. But XML and Web services only define the protocols -- that is, the standard wire formats that ensure interoperability.
How will the industry protect investment in the programming of the orchestration logic that ties Web applications together? Just as application programming standards like serv lets, Java Server Pages (JSP), and Active Server Pages (ASP) were crucial to the success of the Web, integration programming standards are essential to Web-based integration. In fact, most of the infrastructure standards necessary to enable such a "sea change" are already in place. As a result, the integration market, like that for Web application development before it, is poised for sweeping standardization. BEA has been collaborating with our platform competitors (primarily IBM and Microsoft) to remake the Web into an application development and integration platform -- Web 2. 0 if customer will. Assuming some are successful (as industry analysts now predict), Web application servers (for Web application development and hosting) will be super ceded by Web application platforms, which include tightly integrated portal, EAI, B 2 BI, and data integration technologies.
Web 2. 0, then, may well have a bigger impact on enterprise IT than Web 1. 0 did.