On 11/07/06, Jan Algermissen <[EMAIL PROTECTED]> wrote:
>
>
> On Jul 11, 2006, at 5:56 PM, Steve Jones wrote:
>
> > What someone cares
> > about is that they can turn the lightbulb on and off,
>
>
> If it is a lighbulb app, *that* really should not be an issue.
Eh? That is the only issue. There is a lightswitch a couple of feet
from me now, I use it, my wife uses it, and my four year old daughter
users it.
I'm 100% certain my daughter doesn't know how it works. This is a
brilliant example of good interface design.
>
> What I'd really care about (as developer, project lead, upper
> management, or shareholder) is the amount of work to be done when it
> comes to integration or evolution (which it will - I hope we agree on
> that at least). And that is all the REST proponents are saying: by
> constraining your architecture to the REST style you gain better
> changeability.
But, unfortunately, with very little evidence to back this up. Most
people would say that having an interface that clearly describes what
something _does_ and is simple to navigate and interact with is a
pretty effective approach. And that the interfaces are different for
different things (a switch being a fairly rubbish interface for
steering a car for instance).
My case would be that better flexibility comes from being better
aligned to how the requirements are going to come in and require
change. This means understanding how the business operates, the
technology is secondary.
REST is okay, SOAP is okay, but if you don't understand the problem
domain properly then they are both going to deliver rubbish.
I'll stop now as I've seen this bit of the conversation before.
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/
