Re: [ciq] OLD [new requirements]

[Max Voskob <max.voskob@paradise.net.nz>]

David,

Again, it's an old argument.
We have to be realistic - people are not going to embrace CAM just because
it's a good idea and there is an open source implementation. There are many
factors affecting this:
1. they don't understand it
2. they have never heard about it
3. they are conservative
4. they require wider industry support to adopt the standard

Same thing applies to RDFS and OWL.

Take for example NZ E-Gov with their flavour of xNAL. They had a very well
defined requirement - "we wanna keep it simple". If I said "CAM" there they
would walk away. There are others who would readily embrace it. There is no
one size fits all.

So, the idea is - let's keep it simple and modular. Everyone will simply
pick the level of complexity that suits the task.
What if we stage it this way:

1. Main stream - agree on the basic structures and define them in XSD.
2. add additional features such as CAM and RDFS

David, why don't you grab the requirements document and make a few changes
as you see it? :-)

Cheers,
Max







----- Original Message ----- 
From: "David RR Webber" <david@drrw.info>
To: "Max Voskob" <max.voskob@paradise.net.nz>; <ciq@lists.oasis-open.org>
Sent: Tuesday, 3 February 2004 10:19
Subject: Re: [ciq] OLD [new requirements]


> Max,
>
> OK.
>
> Actually I'm seeing that a CAM template is just the same
> as an XSD in terms of implementation here.
>
> Since the CAM processor is open source now - people
> needing the level of functionality CAM brings you will
> not have any impediment.
>
> People who want to do more with their information
> exchanges quite simply cannnot get this for "free".
>
> Whether its RDF or OWL or CAM - you have to
> use software to do something useful with it.
>
> But here's the good news about CAM templates -
> they are designed so you can hand-edit and read them.
>
> Its not like you have to have some GUI tool to
> interpret it for you.
>
> And you can paste your CIQ instances right into
> the CAM template structure and then add the
> CAM rules below it - so there is minimal
> impact on your existing content.
>
> The PPT link I provided contains a link to
> an example that uses multiple address formats,
> so you can see how easy this is to engineer.
>
>
http://drrw.net/presentations/XMLWG-01-04/CAM%20Order%20Item%20example.xml
>
> No tricks required!
>
> Cheers, DW
>
> ----- Original Message ----- 
> From: "Max Voskob" <max.voskob@paradise.net.nz>
> To: <ciq@lists.oasis-open.org>
> Sent: Monday, February 02, 2004 3:52 PM
> Subject: Re: [ciq] OLD [new requirements]
>
>
> > Hi David,
> >
> > The idea of different profiles came from studying many use cases where
> xCIQ
> > were applicable.
> > We came across many situations where people were discouraged by the size
> and
> > complexity of the standards.
> >
> > It is possible to achieve the same results with less effort thru
> > simplification and normalisation of xCIQ. This is the main driver for
the
> > change.
> >
> > Take UDDI for example. We are considering some revamp to the contact
> details
> > area. It would be sensible to use xNAL there, but it is just too big.
The
> > contact details part may become larger than the rest of the UDDI schema.
> :-)
> >
> > I don't know why you are talking about bells and whistles - I didn't
> propose
> > any.
> > Neither I mentioned webservices (did I?) - xCIQ are passive data
> structures.
> > If we decide to add any functionality it will be xCIQ API then.
> >
> > CAM is a great idea. I think that any standard will benefit from
adopting
> > it. On the other hand, it is a burden for implementers. Is it possible
to
> > use CAM as an optional add-on feature?
> >
> > We also need to keep in mind that the era of the semantic web is not far
> > away and gear up to this. Ram told me to make an RDF schema for xNAL as
> far
> > as 2 years ago. I also included it as a requirement.
> >
> > Anyway, it's good to see that we got the ball rolling this time :-)
> >
> > Cheers,
> > Max
> >
> >
> >
> >
> > ----- Original Message ----- 
> > From: "David RR Webber" <david@drrw.info>
> > To: "Max Voskob" <max.voskob@paradise.net.nz>;
<ciq@lists.oasis-open.org>
> > Sent: Monday, 2 February 2004 17:18
> > Subject: Re: [ciq] OLD [new requirements]
> >
> >
> > > Max,
> > >
> > > Looking at your document seems to me these are
> > > all OLD requirements!?!
> > >
> > > I like the idea of creating PROFILES.
> > >
> > > I'm a bit hestitant over going with the requirements
> > > document as you have it.  Certainly from the
> > > address point of view - we can categorize
> > > communities (aka profiles).  But specific
> > > technical features (web service, et al) I think
> > > we should be careful of.   Better to stay
> > > more abstract - and then look at a selection
> > > of methods.  People can always bolt on
> > > technical bells and whistles as they desire
> > > to our foundation methods then.
> > >
> > > In this regard - I believe the UPU has a
> > > comprehensive and clear set of requirements
> > > for postal addressing.
> > >
> > > What I'd like to see is us taking that and
> > > creating one sample PROFILE for CIQ
> > > and UPU using CAM templates as the
> > > bridge.  CAM is providing a method
> > > based around UID references, noun,
> > > and sub-assemblies.
> > >
> > > Now we have the open source CAM
> > > processor available ( http://jcam.org.uk )
> > > there is nothing technically impeding this
> > > work.  I know the UPU has some
> > > excellent test data (and I'm sure there is others
> > > such as Australia, Japan, etc) that we can use.
> > >
> > > By making one such PROFILE - I believe
> > > we will then be in a position to understand
> > > how these other areas may also be addressed.
> > >
> > > What I see here is that the challenge for CIQ
> > > is really one of moving beyond what XSD has
> > > to offer - and using techniques that can
> > > provide the level of eBusiness re-use of
> > > address structures that are required.
> > >
> > > If you crack that nut - then all these
> > > profiles will fit neatly into place.
> > >
> > > Thanks, DW.
> > >
> > >
> > > ----- Original Message ----- 
> > > From: "Max Voskob" <max.voskob@paradise.net.nz>
> > > To: <ciq@lists.oasis-open.org>
> > > Sent: Sunday, February 01, 2004 7:15 PM
> > > Subject: [ciq] new requirements
> > >
> > >
> > > > Hi Ram,
> > > >
> > > > As we decided to get started with a new version of our standards I'd
> > like
> > > to
> > > > upload a requirements documents.
> > > >
> > > > Could you please, create a folder for draft documents on v.next?
> > > >
> > > > Cheers,
> > > > Max
> > > >
> > > >
> > > >
> > > >
> > > > To unsubscribe from this mailing list (and be removed from the
roster
> of
> > > the OASIS TC), go to
> > >
> >
>
http://www.oasis-open.org/apps/org/workgroup/ciq/members/leave_workgroup.php.
> > > >
> > > >
> > >
> > >
> >
> >
> > To unsubscribe from this mailing list (and be removed from the roster of
> the OASIS TC), go to
>
http://www.oasis-open.org/apps/org/workgroup/ciq/members/leave_workgroup.php.
> >
> >
>
>
> To unsubscribe from this mailing list (and be removed from the roster of
the OASIS TC), go to
http://www.oasis-open.org/apps/org/workgroup/ciq/members/leave_workgroup.php.
>
>