On 2 Feb 2006, at 02:09, Danny Ayers wrote:
On 2/1/06, A. Pagaltzis <[EMAIL PROTECTED]> wrote:
* David Powell <[EMAIL PROTECTED]> [2006-02-01 17:10]:
On the contrary, I have a problem with preventing multiple
revisions from having the same atom:updated value.
But then you have no idea which revision will be picked by
clients that try to show the latest one, which to me sounds like
it fulfills the criteria for a potential interop problem, which
is exactly what "SHOULD" is about, so I think this was the right
choice.
yes. One could of course have also allowed different language
versions of an entry at the same time for example. But then things
would probably have gotten a lot more complicated in the protocol.
It's not straightforward, this came up pretty quickly in the Atom/OWL
efforts. As Bill said, it's a composite key.
yes. Though this is very much true for the web on the whole. A url
names a thing
but there may be many representations for that thing, even at the
same time. So for a given URL there can be any number of different
language dependent representations, or even different mime type
representations. If you want to refer to the representations instead
of the resource you do usually need a composite key. You are not
usually aware of this composite key nature of the web, as you usually
use defaults: the default representation is returned as the
representation given to you at the time you ask for it, in the
default language, for the default mime type.
An atom Entry is a representation of the resource named by the id.
This id can only have representations that vary along the axis of
time it seems.
In RDF/OWL that has so far demanded a rule* - somewhat less
convenient than having a single
URI as id.
The id relation is functional. That is an entry can only have one id.
[ a :Entry;
:id <tag:eg.com,2/t1>;
:updated "2003-12-13T18:30:02Z"^^xsd:dateTime;
:title "an example"
] .
but the same feed can have another anonymous :Entry with the same id.
[ a :Entry;
:id <tag:eg.com,2/t1>;
:updated "2005-08-24T09:32:04Z"^^xsd:dateTime;
:title "an fixed example"
]
The inverse of :id is :version so that we could rewrite the above
triples as
<tag:eg.com,2/t1> :version [ a :Entry
:updated
"2003-12-13T18:30:02Z"^^xsd:dateTime;
:title "an example" ];
:version [ a :Entry
:updated
"2005-08-24T09:32:04Z"^^xsd:dateTime;
:title "a fixed example" ] .
which shows more clearly the relation between the resource and its 2
representations.
Now the question is when we find two entries with the same id and the
same updated time stamp are they the same entry? This is where the
CIFP's Danny metnioned enter into play.
And this is where I am unsure in atom-owl currently. My feeling is
that deep down that is what all of us believe.
The question is, from what is the composite key formed? id
certainly, updated too - but is same id+updated enough? What if other
parts of the entry differ? Different xml:lang even...same entry or
different?
or is it only the case in one feed document?
*see also : http://esw.w3.org/topic/CIFP
Cheers,
Danny.
--
http://dannyayers.com
