Oracle Corp is set to announce its new open repository strategy Monday, something it’s been hinting broadly about over the last few months (CI No 3,444, 3,405). While Oracle has had a repository embedded within its Designer 2.1 modeling and generation tool for some time, on Monday it will preview version 7, the first to have its own, separate identity. The repository is intended to make code re-use easier for developers, help with the set-up and maintenance of complex data warehouses, and enable Oracle end users to more easily customize their applications. Based on top of Oracle 8.1, the Oracle Repository 7 should move into beta testing by the end of this year, and is expected to go into full production by the middle of next. The product is intended to support all of Oracle’s range of software development tools, including Designer, Developer, JDeveloper and future tools. It will also work with the Oracle Express OLAP, the Discoverer query and reporting tool and Warehouse Builder for data warehouses. And Oracle says the new version will open up the repository to third party tools as it implements various industry standards. Version 7 will be extended to support some of the newer object-oriented analysis and design standards that are emerging, the Universal Modeling Language in particular. The Java language and Java components will also be supported, to take advantage of Oracle’s Jdeveloper tool and the Java Virtual Machine support built in to Oracle 8.1. Version 7 will also include Java and Corba bindings, rather than the proprietary PL/SQL database programming language it supports today. For data warehousing, Oracle plans to add a common data warehouse model, and is one of the submitters of the proposed XML-based SMIT structured metadata interchange format for repositories to the Object Management Group. Other supporters include IBM Corp, Platinum Technology Inc, Rational Software Inc and Select Software Tools Ltd. For applications integration, Oracle plans to eventually add complete business models, design models and information models to support applications generated mostly or entirely from its modeling tools. And it will include support for full-scale software configuration management, a weakness of many competitive repository products, it claims. Built on top of Oracle 8.1, the system will be suitable for large scale set-ups including 100s of gigabytes of metadata serving systems in multiple locations, according to Oracle. That’s the kind of market segment where Microsoft will find it hard to follow with its SQL Server 7-based repositiory, says Oracle. Although partners aren’t due to be announced on Monday, Oracle claims to have a handful of household names ready to support its efforts, and says it isn’t aiming at the kind of shallow integration that its competitors have indulged in. Instead it will work with a small number of important players: on tight integration. Those names should emerge over the next few months. Most of the development has been done at Oracle’s UK labs in Reading, Berkshire.
Content from our partners
