Steve Jones wrote:
> This is my point, not that REST is rubbish or SOAP rocks, but that the
> implementation of the execution context is of NO VALUE at all to the
> business, and having multiple approaches and multiple arguments
> reduces the value available, increases the complexity and reduces the
> focus away from solving the NEXT problem onto re-solving the current
> one.

This is a very powerful statement from my perspective.  I'd like for there to 
be 
one programming language, one transport protocol, one application interface to 
the network and a lot of other non-variation on a theme.

For decades now computer scientist have cast the same old paradigms into new 
representations, and then tried to breathe life into them with proud statements 
of purity and flexibility or reliability.  Then, we've gone through years of 
fixes, upgrades/enhancements that have not really made a big differenence.

For me, Java is the one difference.  Because of the "Java Platform", we have 
standard APIs that I don't have to build into my distribution.  The Java 
language spec provides some powerful things.  It lacks many things that I'd 
enjoy using.  But, because of the overall power of the platform, I don't worry 
about all those things, because they aren't keeping me from solving the 
problems 
I want/need to solve.

The big issue for me is that once I am using Java and talking to other Java 
applications, why would I want to use anything but RMI programming?  In the 
past, I did create abstraction layers to allow me to switch from RMI to 
something else.  But now days, with JERI in place, I don't worry about that.  I 
just leave the exercise of providing the proper invocation layer to those who 
want to talk to some other mechanism.  They can write the code and plug it in 
by 
changing the configuration file to use the appropriate bits.

If I program to just use HTTP, I'm stuck in a different rut.  If I design my 
systems to be restful, that doesn't preclude the use of RMI, at all...

I'm with Steve on the "Where's the value" thing, and how do we make the "users 
experience" best.  As Jim Waldo has said time and time again.  Wire protocols 
shouldn't matter any more.  I'd add that in the end HTTP in order to be used 
arbitrarily as a replacement for another application protocol would generally 
only appear to be a wire protocol in use.

Gregg Wonderly






------------------------ Yahoo! Groups Sponsor --------------------~--> 
Yahoo! Groups gets a make over. See the new email design.
http://us.click.yahoo.com/XISQkA/lOaOAA/yQLSAA/NhFolB/TM
--------------------------------------------------------------------~-> 

 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/service-orientated-architecture/

<*> To unsubscribe from this group, send an email to:
    [EMAIL PROTECTED]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 



Reply via email to