More importantly, if we don't really know what SOA is, how do we know 
that it's a good idea?  Or that it's actually better than any other 
approach?

One of the common factors in the various industry definitions of SOA I 
see is a long list of unsubstantiated claims for the benefits such as 
how it will make your business more agile, reduce costs and so on.

But what I am still missing is a proper reasoned argument as to exactly 
*how* SOA (and only SOA) realises these business benefits.  This is, in 
my view perhaps the single most important attribute of any architecture 
definition.

In my work in this area, I struggle to see how what many people call 
SOA actually delivers the stated business benefits.  Conversely, if you 
look at how really to achieve those benefits, you end up with a very 
different kind of architecture!

-Mike.

--- In [email protected], "Gervas 
Douglas" <[EMAIL PROTECTED]> wrote:
>
> No one anywhere in the known universe has yet come up with a
> definition of SOA which commands widespread acceptance.  Perhaps it is
> time we had another crack at it.
> 
> Over to you ladies and gents...
> 
> Gervas
>


Reply via email to