Hi Steve, Here are some answers for you:
> Overall, I'm impressed with the progress and glad that a > near-dead project has been revitalized. I'd also like to > find out more about your OSGi-based deployment design. Joel Hawkins from our team could speak in more detail about the OSGi deployment code itself, but in general, we have a programming model that is consistent across J2EE/OSGi, so users don't have to recompile code or change their deployment descriptor when moving from one platform to another. Each platform we support (Axis2, OSGi, etc) has a platform-specific module that handles the required deployment/semantics and allows the user to ignore the underlying platform (for the most part ;) ). > As I am currently trying to get some WSRF interop between > Muse 1.x and another stack working, I have some questions: > > 1. How have you tested WSRF/WSN interoperability? > 2. are there any normative tests for WSRF/WSN? If not, > when you have an interop problem, how do you decide > who gets the blame? 1. Interop testing happens in a number of ways. First, schema validation of incoming/outgoing messages. But more importantly (since a number of the WS-* specs are rather liberal in their schemas), we have tested with other WS-* client/browser tools (IBM has made at least three that I know of); testing with other stacks happens as we discover the stacks. 2. There are no normative tests - the TCs do not publish any kind of tests like Sun does for Java specifications. So, I will go with the traditional programmer answer: "it depends". Since it's an open community, I think the only reasonable answer is that we would work with whoever was responsible for the other stack and try to determine which part(s) of our implemenation(s) was incorrect. Dan Steve Loughran <[EMAIL PROTECTED]> wrote on 09/20/2006 08:46:26 AM: > Daniel Jemiolo wrote: > > Hello PMC Members, > > > > The Muse team is hard at work making web site and documentation updates > > for the current build, with a release planned for Friday, the 22nd[1]. I > > wanted to take this opportunity to pause and send an update to the entire > > PMC team and see if anyone had any questions about the state of the > > project. I put a quick update in the September Board Report, but I know > > everyone is busy and wanted to make sure this info was out there. > > > > Since our first milestone at the end of June, we have had a second > > milestone (August), frozen the code (September), and have refactored the > > web site and documentation to account for new versions. We have been > > fortunate to receive a lot of bug reports, ideas for improvement, and > > (best of all) patches for both. > > > > The IBM programmers that donated the current code base have continued to > > work on bugs related to the core engine and code generation tools. > > Programmers from Compuware have contributed a large set of code for > > OSGi-based deployment and some refactoring of the core engine. Our Cisco > > contributors helped us recreate our build with Maven so that we had less > > scripts and less steps; they have submitted a number of bug reports/fixes > > as well, but the help in setting up a build/test framework was most > > important (and unexpected). Finally, the Eclipse TPTP and Corona > > programmers have integrated Muse into their projects, giving it even more > > exercise before our freeze and release. > > > > When you add all of these contributions together, we have a stable > > codebase that offers an implementation of the recently-ratified WSRF 1.2, > > WSDM 1.1, and WSN 1.3 specs for Axis2 and OSGi. Our plan for the rest of > > the year is to focus on bug reports and the mailing list for our release. > > Overall, I'm impressed with the progress and glad that a near-dead > project has been revitalized. I'd also like to find out more about your > OSGi-based deployment design. > > As I am currently trying to get some WSRF interop between Muse 1.x and > another stack working, I have some questions: > > 1. How have you tested WSRF/WSN interoperability? > 2. are there any normative tests for WSRF/WSN? If not, when you have an > interop problem, how do you decide who gets the blame? > > -Steve > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
