Ajax and rich clients will be having a bit of a coming out party at the Eclipse Foundation’s annual EclipseCon conference. Among the headlines, Eclipse’s Ajax Tools Project is hitting its first milestone, containing ‘most’ of the features that are expected for its 1.0 release, which now is planned for June.
The initial milestone, modestly dubbed 0.2 (obviously, the core open source folks aren’t terribly marketing-savvy when it comes to version numbers), supports four Ajax frameworks out of the virtual box. They include Dojo, which is arguably the best known, plus Rico, Zimbra, and another called Script.aculo.us.
The Ajax Tools Project is an attempt by the Eclipse Foundation to produce a core toolset that in turn can support some of the hundreds of Ajax frameworks that have sprung up from the woodwork over the years.
It’s a matter of timing, said Mike Milinkovich, executive director of the Eclipse foundation and also a member of the OpenAjax steering committee. He claimed that Eclipse has for now a 6-month head start on the OpenAjax effort.
It’s still too early to see what will come out of the OpenAjax workgroup, he said.
At this point, Nexaweb, Helmi, and Genuitec are supporting the emerging Eclipse Ajax Tooling Project in their products.
In a related development, Eclipse will announce the new Rich Ajax Platform project. Designed to complement its existing Rich Client Project, which offers a neutral framework that enables developers to use Eclipse tooling outside the web on Windows, the Mac, Linux, and several flavors of UNIX, RAP is supposed to enable Eclipse tools to work in a rich Ajax browser client.
According to Milinkovich, the eventual goal is to get a one-to-one mapping between the two rich Eclipse clients.
Our objective is to defer the deployment decision as far along the development process as possible, he said.
Today, the choice of the client platform for a software application is made at design time, or it is made even earlier as a basic architectural decision.
But the goal of harmonizing the Eclipse Rich and Ajax clients would be to make deployment platform a last minute decision. That would enable you to design the application independent of the target client, and decide at the last minute where it will deploy.
Ironically, while this was a goal of Java’s original Write Once, Run Anywhere promise, in reality, having platform-independent clients has never been much of a dream in the development world because it was never deemed feasible. That’s a bit ironic, given that separation of other tiers, such as data from logic and presentation, is now taken for granted.
Milinkovich would not hazard a guess as to when Eclipse rich clients would become client platform independent.