RE: [emergency] Groups - ICS-201-draft0.2.xsd uploaded

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

I am saying don't normatively associate any spec with any other spec 
unless you have to.  That's all.  Don't let me confuse you because 
I don't have time to look into ICS.  I was responding to Rick 
being uncomfortable with specifying XForms.

XForms is an explicit application language.  There are alternatives 
to implementing XForms so you should be uncomfortable with that choice.

Is the process specification meant to be executable or 
only an abstraction of a process that can be implemented 
differently?  Do you *need* a process specification or is 
it informative material provided to help other implementors?

Dare to do less in cases where doing more forces you to 
make choices for the implementors that they can better 
make for themselves.  Do more if the chances are good 
that without the extra work, the specification can't 
be implemented interoperably at all.  Be very certain 
about interoperation:  systems interoperate; data is 
portable.  So if you spec interoperations normatively 
you are designing the system, not the data.

len

-----Original Message-----
From: R. Allen Wyke [mailto:emergency-tc@earthlink.net]
Sent: Thursday, April 01, 2004 1:33 PM
To: Bullard, Claude L (Len)
Cc: emergency@lists.oasis-open.org
Subject: Re: [emergency] Groups - ICS-201-draft0.2.xsd uploaded


Ok, so are you agreeing or disagreeing with what I proposed? It sounds 
like you agree that we should not attempt to define an XSD (schema) for 
ICS 201, but then you mention preparing a process spec and not address 
the front end (GUI), which is what I *think* we can do with XForms. Or 
are you saying that even touching this with a 10 foot pole (ie: XForms 
too) is way to close?

Just wanting to make sure I am not misunderstanding. Personally, I 
could go either way. I just know that I have a comfort issue with the 
XSD. Jury is out on the XForms idea - it was just a thought that I felt 
was better than defining an XSD.

On Apr 1, 2004, at 2:25 PM, Bullard, Claude L (Len) wrote:

> I agree with Rick.  Don't open liaisons or set dependencies
> on other specifications and standards unless absolutely necessary,
> meaning, your specification can't be used without them.
>
> 1.  Your specification will be tied to the evolution of the
>     other specification, and
>
> 2.  It might lose.
>
> len