The InfoBus is a compact Java API which allows cooperating applets or Beans, on a Web page or in any other Java application, to communicate data to one another. The InfoBus architecture enables Beans to be categorized as "data providers" and "data consumers". Data providers are Beans whose primary function is to access data from their native store, such as a DBMS, spreadsheet, flat file, Lotus Notes database etc., and to offer data onto the InfoBus.
Data consumers retrieve data from the bus, for analysis or visual display. This segregation of provider from consumer is extremely powerful in that it enables applications to be independent of their data; for example, a charting Bean need not understand SQL or JDBC in order to access DBMS data.
Of course, a Bean can be both a consumer and provider: for example a spreadsheet may accept data from a DBMS and provide data to a chart Bean.
How does the InfoBus relate to JavaBeans?
The InfoBus specification extends JavaBeans by providing a set of enhanced interfaces to share and exchange dynamic data.
Does the InfoBus compete with JavaBeans?
On the contrary, InfoBus extends the power of JavaBeans to a new range of more dynamic applications. InfoBus is fully compatible with JavaBeans, and Lotus enthusiastically supports JavaBeans as the component standard for Java. Wherever possible, InfoBus uses the existing mechanisms of JavaBeans.
Will the general public have access to the InfoBus API's?
Yes, under the usual terms of the JDK license. The specification and technology preview are available in our products archive.
Is InfoBus easy to use?
Using InfoBus aware components, such as the ESuite components from Lotus, users can easily construct powerful data driven applications. No programming or scripting is required: simple parameters are used to establish connections to databases and to select data for processing. For developers creating InfoBus components, InfoBus offers a straightforward API compatible in style with other features of JavaBeans.
Is the InfoBus client side only?
Yes. InfoBus is typically used to communicate among Beans at the client, and it can also be useful for sharing information among components at a single server site. The initial version of InfoBus is not distributed, and is therefore not intended for transmission of data between clients and servers.
A variety of Java communication services, including JDBC, CORBA, and RMI can be used by InfoBus components for access to distributed data. For example, the ESuite Data Access component is a Bean that connects any JDBC compliant data source to the InfoBus. Once the data is on the bus, it can easily be imported by any InfoBus Data Consumer, such as a spreadsheet, charting component, word processor or data analysis tool.
JavaBeans has mechanisms like "bound properties" for data transfer between components. Why is the InfoBus necessary?
JavaBeans mechanisms, such as bound properties, are very useful in cases where the type of data to be communicated can be hard-coded into the source and target components. The InfoBus adds additional features required for more dynamic data interchange. These can be cases where communication is driven by the content of the data and where the nature of the data to be exchanged is determined at runtime.
Can both Java applets and JavaBeans components use the InfoBus?
What is the relation between the InfoBus and RMI?
The InfoBus architecture addresses Beans talking to one in a single JVM not across multiple JVMs; while RMI (Remote Method Invocation) is intended for communication across JVMs (different Java Virtual Machines across the network). As for IIOP, one can envision a JavaBeans component that uses RMI to talk to something on another JVM and then publishes the data on the InfoBus.
Additionally, RMI could be used to allow components in different security classes to communicate within the same JVM.
*As used on this web site, the terms "Java virtual machine" or "JVM" mean a virtual machine for the Java platform.