[
Lists Home |
Date Index |
Thread Index
]
On Tue, 3 May 2005 8:43 am, Bullard, Claude L (Len) wrote:
> to put tags around it or more smilies. OTOH, black boxes are a pretty good
> place to hide bad processes. Simply hooking up SOAs won't change that.
> The idea that Sarbanes-Oxley is amenable to SOA is interesting. If ever
> there was an architectural challenge, that is one.
I admit, I am a simpleton and missed this one.
heck, I only got my first dvd player last year...
> 2) To reply to David(2): Vector models are informative. There is a lot
> we can do without XML such as is being done with your gmail. Discussions
> of the models can open up innovation. It's always worth discussing.
Still, we're only just scratching the surface with what can be done with xml.
Getting the ideas and selling them is the easy part. Writing them is a little
harder. Debugging them harder again. Production testing is time-consuming.
I really like deBonos illustration of innovation, that it is a non-symmetrical
process. In that, the way back to the original idea can be a very short
(straight) distance, when the path to the innovation is more like getting
lost in English countryside lanes.
And whilst xml itself is very simple, it's application in any given set of
businesses can be very complex. Given the high number of business rules that
need to be understood for any system to work.
The perennial "I just want a simple system that..." that inevitably leads to
an enormous amount of complexity. But what can you do.
It is the design of the universe that the most system simple of systems,
inevitably end up to be the most complex on closer examination.
and we haven't even started to discuss "String-theory" yet...
[runing for cover]
David
--
Computergrid : The ones with the most connections win.
|
