Re: [ubl] Comment Disposition Clarification: 1.6

["Stephen Green" <stephen_green@seventhproject.co.uk>]

Mike

I'm pretty sure we decided that this just wasn't happening like
that. The issue was more with the fact that the end of one word
didn't reset the (metaphorical) alphabetic order counter to zero
but treated the first letter of the next word as part of the consideration.
i.e.
OrderLine comes after Order and also comes after Ordering
(not a real example).
The TC meeting decided that this was the correct understanding
of alphabetic order.
So you'd get
Order then OrderGoods then Ordering then OrderLine
and not
Order then OrderGoods then OrderLine then Ordering.

No real change is needed to the above situation.

Case only comes into it because the second word happens to start
with a capital and the first word ends with a lower case letter.

I hope that makes sense.

I suppose the rule could make this clearer but the matter that
there needs to be an exception when Type is appended due to
Schema terminology rather than business semantics (i.e. complex Type)
is more of a requirement since this is happening when we'd prefer
that it didn't.

Steve

----- Original Message -----
From: "Grimley Michael J NPRI" <GrimleyMJ@Npt.NUWC.Navy.Mil>
To: <ubl@lists.oasis-open.org>
Sent: Thursday, July 15, 2004 9:32 PM
Subject: RE: [ubl] Comment Disposition Clarification: 1.6


> Steve,
>
> I thought case came into play because someone had pointed out that, if
case *is* considered, then "an object with a small 'a' would appear after an
object with a capital 'Z'. User will consider this as user infriendly and
will complain for sure." (from comment sheet)
>
> Thanks,
> Mike
>
> -----Original Message-----
> From: Stephen Green [mailto:stephen_green@seventhproject.co.uk]
> Sent: Thursday, 15 July 2004 16 14
> To: ubl@lists.oasis-open.org
> Subject: Re: [ubl] Comment Disposition Clarification: 1.6
>
> Mike
>
> At present it is possible to have (and we do have it) in one place
AddressLine following Address (quite rightly) and, in the same Schema,
elsewhere to have AddressLineType preceding AddressType.
> This is because '..Type' is included when deciding the order
alphabetically
> - T comes after L.
> We need a rule to say that Schema-associated terms such as the 'Type'
> suffix should be ignored when calculating the order (wherever the order is
not dictated by the model, e.g. when producing an xsd:sequence).
> (And case isn't really a factor.)
>
> That's my recollection of the result of the Pacific call on 1st July.
>
> All the best
>
> Steve
>
> ----- Original Message -----
> From: "Grimley Michael J NPRI" <GrimleyMJ@Npt.NUWC.Navy.Mil>
> To: <ubl@lists.oasis-open.org>
> Sent: Thursday, July 15, 2004 8:47 PM
> Subject: [ubl] Comment Disposition Clarification: 1.6
>
>
> > Greetings,
> >
> > Disposition states: "Need NDR rule,...".
> >
> > This would be a rule stating that case will *not* be considered when
> alphabetizing schema elements, correct?
> >
> > Thanks,
> > Mike Grimley
> >
> > 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/ubl/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/ubl/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/ubl/members/leave_workgroup.php
.
>
>