Re: How many XML gurus does it take to change a light bulb?

[Francis Norton <francis@redrice.com>]

Stuart Naylor wrote:
> 
...
> 
> Without any arguments to what is potentially a better method than
> another which initiatives seem to be the winning horses. My criterion
> for what is a winner is purely based on commercial usage at this moment.
> 
> My main area of interest is in commercial transactions of sensitive data
> and I have been looking at EbXML and various others.
> 
> Also there seems to be two approaches of large xml document interaction
> V XML RPC such as SOAP on small XML packets.
> 
I can't give you a comparative analysis but I'm pretty well committed to
web services, by which I mean the XML Schema / SOAP / WSDL / UDDI
technologies structure. Commercially, we have done SOAP over SSL, with
certificates. I used to think that WSDL was unnecessary cruft, but now
I'm using .NET beta 2 I've started to realise that it is the interface
spec, and if you believe in interface-centric development (and I do)
then WSDL is probably pivotal, the place to start the automagical code
generation. 

Since we already use XML interfaces between our application components,
I'm right at home with the literal document usage of SOAP.

Reservations? Performance isn't exactly lightning right now, but I'm
just using beta software on an overloaded laptop. I'm pretty sure
they'll fix this. Also I haven't tested interoperability yet, though
again I'm pretty confident that this is subject to a strong pattern of
forces that will lead to a satisfactory outcome. (Thouhg I did twitch
when I read that "The " (certification) "tests used by UDDI operators
are not published, but instead are part of the overall contractual
relationship binding UDDI operators.") 

Overall, I believe that web services will be the next major internet
application

Francis.