On Mar 22, 2006, at 11:53 PM, ant elder wrote:
Sure ok, but I think I have a slightly different and less main stream
perspective on whats important/interesting to do :-) So for this
release I'd
rather fit in with what others want which is why i'm asking for
specific
suggestions.
O.K. but I think it is important for people to express what they
think is important, even if it may be less mainstream.
That said, I started thinking about this a bit more and wanted to
take a step back. While Tuscany will (and should) be more than what
the SCA and SDO specifications dictate, one of the fundamental goals
of the project as a whole is to provide an open source implementation
of them (probably more accurately, a subset right now). In terms of
the Java SCA runtime, for JavaOne, I think it is important we focus
on that.
A lot of the WS suggestions so far - pure doc vs doc-wrapped, wsdl
and xsd
imports, interop etc - are probably more data binding and SDO
related. So
far I've been trying to avoid get to much into how SDO works, so I
like the
XSLT componentType suggestion and along with that I'd like E4X
going with
the JavaScript componentType (was the main reason for doing
JavaScript).
I think it would be good - i.e. a p1 - to improve the JavaScript
component to the point it demonstrates a deeper extensibility
story. XSLT is a "nice-to-have" but if the choice is between having
depth versus breadth in extensibility, I prefer we stick with depth,
which means focusing on JavaScript.
Related to that is the thread about "Data flow on the wire", as
without that
its not possible to avoid the data binding.
Yes
Axis2 has already been shown to interop well, so if our data
binding is
working in theory we should as well. Of course we should test that,
but I'm
less convinced its something thats really important to get done for
the very
first milestone release. People will come look at Tuscany for its
potential,
not to try to actually do some real work.
In the context of JavaOne, my experience has been people will gauge
potential by what it can do in May. I think we are going to have to
demonstrate some use-cases than cannot be achieved easily with
existing technologies or we will fall victim to the argument that
"Tuscany says it will do that some-cool-thing someday, but with
technology X you can do some halfway-cool-thing now". That's why I'd
like to get a set of use cases that first cover the basics (WS,
interop) and then show value-add (multiple bindings, extensibility,
possibly async).
Also, perhaps we should assume things like interop will not work and
we need to factor in significant time in our calculations? My
experience has been that things work well in theory but when the
practicalities of interop hits all hell breaks loose ;-) I'd really
like for our web services story to be rock-solid or at least not
based on Jello.
I'd really like async going and to have that well integrated with
the WS
binding, but I think its probably better to leave that till after this
release to avoid the instability its development is likely to cause.
I think we need to have this as a background process. My reasoning is
that this is likely to cause a refactor for extensibility and we want
to get that out of the way so as not to break people later. I think
the importance of having a more solid extensibility API is worth the
risk this introduces now. What do you think?
I think the item "refine and simplify our extensibility story" is
important
to get done real soon. If we're going out asking people to start
contributing new stuff but then keep breaking what they're trying
to do it
may make them give up.
Yes, agree this is one of the most important things.
Having a release will give us a binary download, and to get to
there it will
help make us sort out things like the project and distribution
structure,
how to contribute new things and where to put the doc and samples for
contributions, and that type of thing. I think thats an important
reason for
having a release.
I think these are all Good Things.
Other than that I think it would be good to find ways to
show how Tuscany plans to be more than just another platform for
Java web
services.
Yes, I agree with that. We may have different reasons why we agree.
My reasoning is based on SCA, which is intended to be more than
"another platform for Java web services". Tuscany should be this
primarily because of SCA.
I think the best way to demonstrate this is through extensibility
inline with SCA around impl types and bindings. Practically, I would
suggest this be a JMS binding since JMS is something a lot of Java
people use, particularly in comparison with others, and is the next
binding type targeted by the spec. This shows SOA != web services.
Jim
...ant
On 3/23/06, Jim Marino <[EMAIL PROTECTED]> wrote:
Ant,
Do you want to outline some things you think are important too? I
didn't list anything specific b/c I don't have any firm opinions.
Jim
On Mar 22, 2006, at 3:31 PM, ant elder wrote:
Are there any _specific_ items people could suggest to improve
the WS
support for this upcoming release? The next steps thread from a
while back
was pretty quiet.
Axis2 already supports most things and interop's well so its 'just'
a mater
of integration with Tuscany. Attachments? raw doc style? headers?
rpc/encoded? WS-X? Whats important to you?
...ant
On 3/22/06, Jim Marino <[EMAIL PROTECTED]> wrote:
That's great. Perhaps we could look at that as part of the Axis
"cleanup work" which needs to be done?
Jim
On Mar 22, 2006, at 1:59 PM, Davanum Srinivas wrote:
Jim,
Axis2+Sandesha has already been thru WS-Addressing Interop and
WS-RM
Interop using both SOAP 1.1 and SOAP 1.2 (specifically with
Indigo/WCF).
thanks,
dims
On 3/22/06, Daniel Kulp <[EMAIL PROTECTED]> wrote:
Jim,
4. Additional bindings to Axis through integration with Celtix,
particularly ws and JMS
- Basic integration (p1)
- Ability to retarget (p2)
- Dan, would Celtix get us Indigo interop?
Good question regarding the Indigo stuff. We do have Indigo
interop on
the roadmap, but I don't know the complete timetable or what that
really
means (since I know nothing about Indigo).
That said, post milestone 5 of Celtix (next week), we have
someone
that
will be doing some interop testing using Celtix with Indigo to
test the
WS-RM, WS-A and SOAP (1.1 only for now) interoperability.
Obviously,
until the results of those tests come in, I don't know how far
away we
are.
--
J. Daniel Kulp
Principal Engineer
IONA
P: 781-902-8727 C: 508-380-7194
[EMAIL PROTECTED]
--
Davanum Srinivas : http://wso2.com/blogs/
