- From: Karen Coyle <kcoyle@kcoyle.net>
- Date: Thu, 22 Jan 2015 10:41:38 -0800
- To: Jose Emilio Labra Gayo <jelabra@gmail.com>
- CC: public-data-shapes-wg@w3.org
On 1/22/15 9:23 AM, Jose Emilio Labra Gayo wrote:
> I have edited the requirements document with some comments and
> objections. In the case of static constraints, I said that I didn't like
> the name "static" and maybe, a more intuitive name, like "global"
> constraints, could be more descriptive.
+1
>
> But apart from that, I think a more technology neutral way to describe
> what the requirement means is the selection of which nodes are affected
> by a constraint description. Which in the case of "static constraints"
> means all the nodes.
Which seems to imply that one can address graph+all subgraphs (or
node+all subnodes), something that could be useful in any case, even if
the top node isn't your very top node.
>
> One possibility is that the nodes affected are those nodes that are
> instances of some specific class (which I think is the default
> possibility in SPIN), another possibility is all the nodes ("static
> constraints") and another one is to select one specific node (which was
> also added as another requirement [1].
I can think of situations where a sub-node is of a different class,
however, which is why the "global" part creates a different case. If I
have a book graph that links to chapter graphs that have author graphs,
and my rule is "one label per subject" then the definition by class is
too narrow. Make sense?
kc
>
> I grouped those possibilities in a section titled "Selection of nodes"
> [2] where I included the
> the three possibilities as three different requirements.
>
> I hope this clarifies the requirements in a more technology neutral way...
>
> [1]
> https://www.w3.org/2014/data-shapes/wiki/Requirements#Evaluating_Constraints_for_a_Single_Node_Only
> [2] https://www.w3.org/2014/data-shapes/wiki/Requirements#Selection_of_nodes
>
> Best regards, Jose Labra
>
>
> On Thu, Jan 22, 2015 at 3:50 PM, Karen Coyle <kcoyle@kcoyle.net
> <mailto:kcoyle@kcoyle.net>> wrote:
>
> Sorry, what I actually meant is: do we have a technology-neutral
> statement of this that can be considered a requirement? We can't
> word the requirements in terms of specific solutions.
>
> However, at this point I've lost the original train of thought. it
> seems that we started with: Wherever property X is used, it can be
> used only once per graph.
>
> That seems to me to be a convenient short-hand that can be defined
> in other ways, as Peter and Holger have addressed. If so, then this
> is a candidate for a macro in whatever interface is used.
>
> kc
>
>
> On 1/21/15 6:24 PM, Holger Knublauch wrote:
>
> On 1/22/2015 11:16, Karen Coyle wrote:
>
> The question, though, is how one
> would express this without using SPIN
>
>
> I assume you mean "without using SPARQL"?
>
> - in other words, do we have a
> generic way to express this requirement? I think it gets
> back to how one
> defines to target of the validation. Because these two cases
> have two
> different solutions, should they be different requirements?
>
>
> Any complex constraint expressed using SPARQL can be turned into a
> (SPIN/LDOM) template. This means that some experts can prepare
> high-level lego bricks for people who don't know SPARQL. The items
> mentioned under "Property declarations" are basically the most
> common
> patterns, and those should of course be built-in. We may
> identify other
> recurring patterns that should also be covered by built-ins (such as
> templates).
>
> Holger
>
>
>
>
> --
> Karen Coyle
> kcoyle@kcoyle.net <mailto:kcoyle@kcoyle.net> http://kcoyle.net
> m: 1-510-435-8234 <tel:1-510-435-8234>
> skype: kcoylenet/+1-510-984-3600 <tel:%2B1-510-984-3600>
>
>
>
>
> --
> Saludos, Labra
--
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet/+1-510-984-3600
Received on Thursday, 22 January 2015 18:41:51 UTC