On Wed, 14 Jul 2004, Phil Barker wrote:
> After some consideration of Chris's original message, I think I would be
> happier with
> <![CDATA[BEGIN:VCARD
> FN:None
> N:None
> ORG:WH Smith
> NOTE:objectClass\: organization
> VERSION:3.0
> END:VCARD]]>
>
> Since I think Chris's suggestion mis-uses both N's additional names and FN,
> and as Andy has pointed out there's no need to put incorrect information
> there. I'm not sure "None", is the right value, but is the one recommended
> in CanCore, and I don't see much point in choosing a more appropriate term
> (such as "not applicable") just to ignore it.
As an aside, I think this highlights the nonesense that tends to occur
when you mandate an element and then find that it isn't always appropriate
to use it :-(. The logic goes something like this
FN and N are mandatory...
therefore we have to provide them...
but there isn't a real value...
so we'll put 'None' instead.
The trouble is... when I get the string 'None' as a consumer of the vCard,
how do I know whether it means "There isn't an appropriate value' or the
name of the entity represented by the vCard is 'None'?
If the element doesn't have a value, either leave it blank (which I
think metadata consumers should reasonably interpret as "the value of this
element is <null>") or don't provide the element at all (which I think
metadata consumers should reasonably intepret as "no information
provided").
All in all, FWIW, I think I prefer any of the following to the 'None'
suggestion above:
Option 1
<![CDATA[BEGIN:VCARD
FN:
N:
ORG:WH Smith
NOTE:objectClass\: organization
VERSION:3.0
END:VCARD]]>
or
Option 2
<![CDATA[BEGIN:VCARD
ORG:WH Smith
NOTE:objectClass\: organization
VERSION:3.0
END:VCARD]]>
or my original suggestion of
Option 3
<![CDATA[BEGIN:VCARD
FN:WH Smith
N:WH Smith
ORG:WH Smith
NOTE:objectClass\: organization
VERSION:3.0
END:VCARD]]>
I guarantee you that at some point in the future exactly the same kind of
nonesense will occur in UK LOM Core for every element that is currently
set to mandatory - and that we should therefore be relaxing our notions of
mandatoriness by allowing implementors to leave the value blank or not
provide the element at all as appropriate.
Andy
--
Distributed Systems, UKOLN, University of Bath, Bath, BA2 7AY, UK
http://www.ukoln.ac.uk/ukoln/staff/a.powell +44 1225 383933
Resource Discovery Network http://www.rdn.ac.uk/
ECDL 2004, Bath, UK - 12-17 Sept 2004 - http://www.ecdl2004.org/
|