Just a quick couple of points:

There are a LOT of changes happening in web services
threse days, including that SwA (Soap with
Attachments) is formally replaced with MTOM at the
same time that WS-I 1.1 came out. Ditto that DIME is
also nogo in SOAP 1.2 and WS-I 1.1. 
Plus, last fall, the XML protocol group published a
binary capability called XOP. 
JBoss is working on replacing Apache Axis with their
own stack. They have a good wiki on the general
subject, esp. since they are working toward J2EE 5.0:
http://www.jboss.org/wiki/Wiki.jsp?page=JBossWS
Apache Axis 1.2 is much more robust than 1.1, and 1.2
was specifically targeted to deliver WS-I 1.0
conformance. 
Axis 1.2 supports wrapped/doc/literal, which seems to
improve one's chances to get interoperability with
.Net. (of course, you may not care.) It certainly gets
you out of the traditional java -> XML serialization
problems that appear in (e.g.) jax-rpc. including that
wrapped/doc/literal is a much smaller footprint over
the wire than fully specified java package.class
naming conventions.

fyi - b

--- Chris Duck <[EMAIL PROTECTED]> wrote:
> I have been doing some research on SOAP
> implementations and came across 
> this link
>
http://www.sosnoski.com/presents/cleansoap/results.html
> with 
> some speed comparisons.  It does not include Apache
> SOAP, but does 
> include Axis (in several configurations), and I have
> read that Axis is 
> significantly faster than Apache SOAP so anything
> should be an 
> improvement over what we've got right now.
> 
> Anyway according to this, JAX-RPC outperforms Axis
> (doc/lit) for small 
> to medium sized responses.  I am not sure where SJ
> would be classified 
> on the "density" scale, but thought it would be
> worth a look.
> 
> Chris
> 
> Robert MacGrogan wrote:
> > Marc, you make a good point about having to work
> against multiple servers. If we have a good SOAP
> > API, then, you're right, we shouldn't have to
> change it. The problem is, the API is not
> > necessarily all that good. And sometimes improving
> the functionality of existing features requires
> > modifying the existing API.
> > 
> > One way limit the impact of backward compatability
> would be to have the client be backward
> > compatible but not the server. If you get the
> latest client, you'll be able to work against any
> > existing server. Or we could freeze the most
> significant API's, connect, get project object, get
> > file object, get file, check in, and check out. Or
> at least required future versions of the server
> > to always support the current API for those
> features. 
> > 
> > You make a good case, Marc. I'll definitely
> consider what you're saying.
> > 
> > By the way, my first work on the 2.2 release will
> focus on speeding up SJ. I'm going to change SJ
> > to use SAX to read XML files instead of DOM. This
> should have a pretty big impact, I think. And
> > I'm going to switch SJ to use the latest version
> of Apache Axis. One change between Apache SOAP
> > and Apache Axis is that AS uses DOM and AA uses
> SAX. Using SAX for all SOAP calls should speed up
> > SJ significantly.
> > 
> > --Rob
> > 
> 
> 
>
-------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT
> Products from real users.
> Discover which products truly live up to the hype.
> Start reading now.
>
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> SourceJammer-devel mailing list
> [email protected]
>
https://lists.sourceforge.net/lists/listinfo/sourcejammer-devel
> 


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
SourceJammer-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sourcejammer-devel

Reply via email to