On Thu, Sep 11, 2008 at 9:29 AM, ant elder <[EMAIL PROTECTED]> wrote:
> > > On Wed, Sep 10, 2008 at 2:46 PM, Simon Laws <[EMAIL PROTECTED]>wrote: > >> >> >> On Wed, Sep 10, 2008 at 2:30 PM, ant elder <[EMAIL PROTECTED]> wrote: >> >>> >>> >>> On Wed, Sep 10, 2008 at 2:24 PM, Giorgio Zoppi <[EMAIL PROTECTED]>wrote: >>> >>>> 2008/9/10 Simon Laws <[EMAIL PROTECTED]>: >>>> > >>>> > >>>> > On Wed, Sep 10, 2008 at 8:32 AM, ant elder <[EMAIL PROTECTED]> >>>> wrote: >>>> >> >>>> >> >>>> >> On Thu, Sep 4, 2008 at 7:53 AM, ant elder <[EMAIL PROTECTED]> >>>> wrote: >>>> >>> >>>> >>> While implementing part of the webapp support in the JEE spec i've >>>> made a >>>> >>> JMS version of binding.sca, its still a bit rough around the edges >>>> but its >>>> >>> got to the stage where it does work. >>>> >>> One feature i think is cool is that by default it uses UDP multicast >>>> to >>>> >>> discover nodes, so you can just start up nodes and they >>>> automatically find >>>> >>> each other and become part of the domain. >>>> >>> >>>> Back from the Way of Saint. James (Camino de Santiago). >>>> I'm in processing to start a new binding for Peer2Peer using Overlay >>>> Weaver (http://overlayweaver.sourceforge.net) >>>> with and going on with my demo. >>>> Ciao, >>>> Giorgio. >>>> >>> >>> That sounds great Giorgio and would complement this JMS based binding.sca >>> way of creating a dynamic SCA domain. If we can get the underlying Tuscany >>> runtime to work well in a dynamic way then we should be able to swap out the >>> sca binding to be jms or peer-to-peer based depending on whats suitable for >>> the environment. >>> >>> ...ant >>> >> >> Yes, that sounds like a good approach. People may have many different >> requirements in terms if the environments in which they want to run and if >> we have investigated at least one other way of configuring nodes/services >> then we have a better view of what it takes to do it. >> >> Simon >> > > I'm starting to think we need to be able to support several different > binding.sca impls concurrently. Right now its a hard coded choice to use the > local or remote binding.sca and that caused TUSCANY-2578 to get > binding.sca.jms working. We've now four different binding.sca impls, Giorgio > is now starting on a new Peer2Peer one, you can easily envisage different > nodes in a distributed domain being connected together in different ways and > so using each of those binding.sca impls as required. We also really have > this already with the implementation.web and implementation.widget use of > jsonrpc - ideally you wouldn't need to specify <binding.jsonrpc> when using > those but leave the binding to default to <binding.sca> and have the runtime > workout that it needs to use the jsonrpc protocol to talk to the browser > client. > > ...ant > > This is probably related to policy, i.e. the policy specifies the QoS requirements of services and references and binding.sca chooses the implementation that supports these requirements. Simon
