ObjectWeb Consortium
Search ObjectWeb Mail Archive: 

Advanced Search - Powered by Google


Mail Archive Home | deployment List | May 2004 Index

<--  Date Index  --> <--  Thread Index  -->

Re: [fractal] Re: [architecture] Fractal packaging & deployment, bis


Eric Bruneton wrote:

I hope it will be possible to create a deployment infrastructure for deploying Fractal-based applications that can also deploy arbitrary (Java and non Java) applications. This is not contradictory: it just imply that arbitrary applications can be seen as Fractal-based applications (and this should be possible since the Fractal model is modular, extensible, and not tied to Java; for example, plain old Java objects are compliant with Fractal level 0).


Of course, this forces people to view their arbitrary Java applications in Fractal terms just to deploy it. It doesn't sound like this would be a successful approach. I understand your goal, but to me it seems like you are reversing the order of things. You want to create a specific framework for deploying a specific component model and then figure out how this can work for abritrary applications.

My approach is to think about an arbitrary, low-level (Java) deployment framework and then figure out how these concepts can support component deployment. This is why I had originally pushed using OSGi as a low-level framework...and why I am now trying to formulate an even better OSGi-like (with respect to deployment) framework.

Or, in other terms, that existing packaging formats can be seen as Fractal packages (and, in particular, that their dependencies can be seen as Fractal "containment and binding" dependencies - hence this mail http://www.objectweb.org/wws/arc/fractal/2004-05/msg00005.html).


Still, I don't think this is a good approach for defining re-usable middleware. It is more of an approach for designing Fractal middleware, which is certainly a reasonable option. For me, though, I want to be able to support Fractal or any other component model that comes along.

The approach it sounds like you are proposing, as I understand it, is somewhat equivalent to creating an app server that defines a mapping of EJB beans to Fractal and then telling people who want to do EJB development to view their beans in terms of Fractal. They aren't interested in Fractal, they are interested in EJB. The same is true for someone wanting to deploy their application.

My impression is that Fractal is supposed to be the model underlying the middleware created by ObjectWeb, not a concept mapping language for everything.

I understand that there is a desire to put Fractal underneath everything else (i.e., just above the Java layer), and this approach might work if everything were based on Fractal, but this will probably not become a reality any time soon. As such, trying to push Fractal so far down the stack is probably a bad idea since it will be too large of a leap for many developers who are not using Fractal.

However I think we will first concentrate on a deployment infrastructure for Java based applications that are also compliant with Fractal at level 3. We will then look at arbitrary Java applications, and then, perhaps, at non Java applications.

So, this is the real issue for me. If your goal is to define Fractal middleware, then your path is probably sufficient. This is less of an interest to me, but perhaps we can meet in the middle. :-)

Note that these are just personal opinions about the goals and the strategy for "Fractal packaging". I would like to have other opinions about what the goals and the strategy should be.

Of course, same goes for me.

-> richard



<--  Date Index  --> <--  Thread Index  -->

Reply via email to:

Powered by MHonArc.

Copyright © 1999-2005, ObjectWeb Consortium | contact | webmaster.