Attendees: Lawrence Mandel, John Kaputin

Minutes

1. Open Action Items - Lawrence Mandel

2008-01-08 - John will fill in the January incubator board report for 
Woden.
2008-01-22
John: I missed Jan. I'll fill in the Feb. report. Feb. is when Woden is 
scheduled to report.
Closed.


2007-12-18 - John will propose a new M8 date on the dev list.
2008-01-08
John: I haven't had a look yet.
Jeremy: It would be good to get M8 out before the Axis2 1.4 release. No 
date yet.
2008-01-22
John: I've estimated a new date of Feb. 20. This allows for 16-17 days of 
effort. Complete development on Feb. 15. I've put this revised date on the 
web site.
Closed.


2007-12-18 - Lawrence will create a tools page on the Woden site with the 
XSLT and converter.
2008-01-08
Lawrence: No progress.
John: Will the permalink for the xslt go on this page.
Lawrence: Yes. We may require build modifications to build the converter 
and package it separately.
2008-01-22
Lawrence: No progress.


2. Milestone 8 (M8) Status - John Kaputin

John: Woden-47 details a requirement to access the extension properties in 
the extension model without having to use a statically typed API. 
Lawrence: This seems like a significant change.
[action] John will summarize this change and post to the dev list before 
merging the change into trunk.


3. Development Discussion - All

Lawrence: Is there a dependency matrix for WSDL 2.0 assertions? Have you 
seen one?
John: No. I think this needs to be written down somewhere. 
Lawrence: I think this can be presented in a table with two columns: 
assertion id and dependencies.
John: A tree is another representation that would be useful.
Lawrence: I'll see if I can find some time to work on this matrix.

John: How should dependencies be specified? XML file? 
Lawrence: We can use an XML file or just call methods on the extension 
registrar and pass in the dependencies with each assertion class.

John: How should the runtime order of assertions be determined? Should 
extenders specify the runtime order?
Lawrence: I think it's preferable for Woden to determine the runtime order 
based on the specified dependencies. This removes the burden from 
extenders and means Woden can handle extensions from multiple extenders 
without adding a burden on the consumer to mediate and create the runtime 
order.

John: For the assertion ids we may run into conflicts with simple 
identifiers if different extenders use the same identifiers. How about 
qualifying the identifiers with a namespace?
Lawrence: That's a possibility but I don't think we should add this 
complexity until we see that there is a problem. For the WTP WSDL 
validator there are very few projects that provide custom assertions. 
Let's cross this bridge when we have to. We should specify a concrete 
example of defining ids in the documentation.

John: I've had another look at the API issue that's open about merging the 
component and element APIs into a single API and I think this is a good 
idea. As the Woden API evolved I've added more methods to the element API 
to make it useful. The element API now looks more like the component API. 
Section 1.3 in the spec defines the reason to have the component model. 
The spec defines a component model for validation. If we don't have 
clients who will be able to use the component model alone I don't think 
it's useful. I'll continue to look into merging the APIs.


4. Other Business - All

Lawrence: I'm on vacation next week. I propose that we cancel next week's 
call.

*Note: The Woden status call for 2008-01-29 is cancelled.


Thanks,

Lawrence Mandel


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to