Re: [Crm-sig] Modelling an Event's General Outcome Ideas? Properties?

2021-12-20 Thread George Bruseker via Crm-sig 
Hi Thanasi,

The proposal creates a consistent way of doing the 'type of' version of a
property that relates one particular to another particular.

So  each individual property:
https://cidoc-crm.org/Property/P20-had-specific-purpose/version-7.1.1
has its typed version like:
https://cidoc-crm.org/Property/P21-had-general-purpose/version-7.1.1

Right?

But I contend there IS NO particular property in regular CRM that expresses
the semantics I indicate above (therefore the proposal cannot generate its
typed version). P21 DOES NOT express the semantics I need (hence also not
P23).

O13 triggers more or less does. in particular. But I need the
generalization. Triggered an outcome of type.

Anyhow, not sure if anyone else needs this, but very common in my data.

Cheers,
G

On Mon, Dec 20, 2021 at 4:35 PM Athanasios Velios 
wrote:

> Following Athina's response and in relation to the question about the
> extant properties, I guess the "type of type" can be replicated with
> thesaurus related properties (e.g. P127 has broader term). I would
> consider the instances of E55 Type slightly differently to normal
> instances and not extent the idea to them.
>
> T.
>
> On 14/12/2021 19:42, George Bruseker wrote:
> > Hi Thanasi,
> >
> > Yes that's true. Good reminder. That might be a solution but then we
> > would need the particular property for expressing that two events are
> > causally connected. I avoided to put it in the last email so as not to
> > stir up to many semantic teapots. But obviously to have the general
> > property we should have the particular property. So we have for example
> > we have the particular properties:
> >
> > https://cidoc-crm.org/Property/P20-had-specific-purpose/version-7.1.1
> > 
> > and
> > https://cidoc-crm.org/Property/P21-had-general-purpose/version-7.1.1
> > 
> >
> > so the analogy to this in my situation is probably
> >
> > O13 triggers (is triggered by)
> > https://cidoc-crm.org/crmsci/sites/default/files/CRMsci%20v.1.4.pdf
> > 
> > and we need the analogy of p21 to make the model complete
> >
> > On another note out of curiosity, in the extension where every property
> > has a 'type of' property what happens with the extant 'type of'
> > properties? I assume there isn't any has general purpose of type
> > property... or is there?
> >
> > Cheers
> >
> > G
> >
> > On Tue, Dec 14, 2021 at 9:20 PM Athanasios Velios via Crm-sig
> > mailto:crm-sig@ics.forth.gr>> wrote:
> >
> > Hi George, all,
> >
> > As part of Linked Conservation Data (and with the help of Carlo,
> Martin
> > and Steve) we proposed the idea of Typed Properties which derive from
> > current CRM properties and always have E55 Type as range.
> >
> > E.g. "bears feature" → "bears feature of type" so that one can
> describe
> > the type of something without specifying the individual. It is very
> > economical in conservation where we want to avoid describing
> > hundreds of
> > individuals of similar types.
> >
> > We are still baking the exact impact of such a reduction from
> > individuals to Types. One issue in RDFS is the multitude of new
> > properties. There seems to be a simple implementation in OWL with
> > property paths. Not an immediate solution but a flag for more to
> come.
> >
> > All the best,
> >
> > Thanasis
> >
> > On 14/12/2021 15:49, George Bruseker via Crm-sig wrote:
> >  > Hi all,
> >  >
> >  > I have situations in which I have events where the data curators
> >  > describe events for which they have generic knowledge of the
> > outcome:
> >  > sold, completed, incomplete, this sort of thing. So there is
> > knowledge
> >  > but it is not knowledge of the particular next event but of a
> > general
> >  > kind of outcome.
> >  >
> >  > We have properties like: P21 had general purpose (was purpose of)
> > which
> >  > is very useful for when the data curator only has generic
> knowledge
> >  > knowledge and not particular knowledge regarding purpose. This
> > seems a
> >  > parallel to this case.
> >  >
> >  > Anybody else have this case and have an interest in a property
> > like 'had
> >  > general outcome' or 'had outcome of type' that goes from Event to
> a
> >  > Type? Or, better yet if possible, a solution that doesn't involve
> > a new
> >  > property but that does meet this semantic need without too many
> > contortions?
> >  >
> >  > Best,
> >  >
> >  > George
> >  >
> >  > ___
> >  > Crm-sig mailing list
> >  > Crm-sig@ics.forth.gr 
> >  > http://lists.ics.forth.gr/mailman/listinfo/crm-sig
> >

Re: [Crm-sig] Modelling an Event's General Outcome Ideas? Properties?

2021-12-20 Thread Athanasios Velios via Crm-sig 
Following Athina's response and in relation to the question about the 
extant properties, I guess the "type of type" can be replicated with 
thesaurus related properties (e.g. P127 has broader term). I would 
consider the instances of E55 Type slightly differently to normal 
instances and not extent the idea to them.


T.

On 14/12/2021 19:42, George Bruseker wrote:

Hi Thanasi,

Yes that's true. Good reminder. That might be a solution but then we 
would need the particular property for expressing that two events are 
causally connected. I avoided to put it in the last email so as not to 
stir up to many semantic teapots. But obviously to have the general 
property we should have the particular property. So we have for example 
we have the particular properties:


https://cidoc-crm.org/Property/P20-had-specific-purpose/version-7.1.1 


and
https://cidoc-crm.org/Property/P21-had-general-purpose/version-7.1.1 



so the analogy to this in my situation is probably

O13 triggers (is triggered by)
https://cidoc-crm.org/crmsci/sites/default/files/CRMsci%20v.1.4.pdf 


and we need the analogy of p21 to make the model complete

On another note out of curiosity, in the extension where every property 
has a 'type of' property what happens with the extant 'type of' 
properties? I assume there isn't any has general purpose of type 
property... or is there?


Cheers

G

On Tue, Dec 14, 2021 at 9:20 PM Athanasios Velios via Crm-sig 
mailto:crm-sig@ics.forth.gr>> wrote:


Hi George, all,

As part of Linked Conservation Data (and with the help of Carlo, Martin
and Steve) we proposed the idea of Typed Properties which derive from
current CRM properties and always have E55 Type as range.

E.g. "bears feature" → "bears feature of type" so that one can describe
the type of something without specifying the individual. It is very
economical in conservation where we want to avoid describing
hundreds of
individuals of similar types.

We are still baking the exact impact of such a reduction from
individuals to Types. One issue in RDFS is the multitude of new
properties. There seems to be a simple implementation in OWL with
property paths. Not an immediate solution but a flag for more to come.

All the best,

Thanasis

On 14/12/2021 15:49, George Bruseker via Crm-sig wrote:
 > Hi all,
 >
 > I have situations in which I have events where the data curators
 > describe events for which they have generic knowledge of the
outcome:
 > sold, completed, incomplete, this sort of thing. So there is
knowledge
 > but it is not knowledge of the particular next event but of a
general
 > kind of outcome.
 >
 > We have properties like: P21 had general purpose (was purpose of)
which
 > is very useful for when the data curator only has generic knowledge
 > knowledge and not particular knowledge regarding purpose. This
seems a
 > parallel to this case.
 >
 > Anybody else have this case and have an interest in a property
like 'had
 > general outcome' or 'had outcome of type' that goes from Event to a
 > Type? Or, better yet if possible, a solution that doesn't involve
a new
 > property but that does meet this semantic need without too many
contortions?
 >
 > Best,
 >
 > George
 >
 > ___
 > Crm-sig mailing list
 > Crm-sig@ics.forth.gr 
 > http://lists.ics.forth.gr/mailman/listinfo/crm-sig

 >
___
Crm-sig mailing list
Crm-sig@ics.forth.gr 
http://lists.ics.forth.gr/mailman/listinfo/crm-sig



___
Crm-sig mailing list
Crm-sig@ics.forth.gr
http://lists.ics.forth.gr/mailman/listinfo/crm-sig

Re: [Crm-sig] Official NameSpaces of CRM Extensions?

2021-12-20 Thread George Bruseker via Crm-sig 
Dear all,

Thanks Nicola, that makes sense. I wonder if it is worth talking about what
namespace the extensions have going forward. Taking CRMDig as an example.
It arose from a project within which FORTH was a major partner and is an
outcome of that work. It thus makes sense that it is registered under a
FORTH namespace. But if it is considered an official extension, should it
eventually have a namespace within the cidoc crm world for
generally consistency / understandability / maintenance? May be worth a SIG
conversation?

Best,

George

On Mon, Dec 20, 2021 at 9:51 AM Nicola Carboni 
wrote:

> Dear George,
>
> The namespace to be used should be the xml:base value in the RDF
> document. Example:
>
> http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; xml:lang="en" 
> xml:base="http://www.cidoc-crm.org/cidoc-crm/CRMsci/;>
>
> http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
> xml:base="http://www.ics.forth.gr/isl/CRMgeo/; xml:lang="en">
>
> The confusion started because the namespace has changed over time
>
> CRMdig 3.2.2 had
>
> http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
> xml:base="http://www.ics.forth.gr/isl/CRMext/CRMdig.rdfs/; xml:lang="en">
>
> The latest version is
>
> rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
> xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
> xml:base="http://www.ics.forth.gr/isl/CRMdig/; xml:lang="en">
>
> Generally they are both documented in prefix.cc, hence someone is still
> using the old ones.
>
> For clarifying the confusion, It is possible to write explicitly in the
> RDF itself the preferred namespace and prefix, using the properties
> vann:preferredNamespaceUri and vann:preferredNamespacePrefix . Example (in
> ttl) from VIR  :
>
> vann:preferredNamespacePrefix "vir" ;vann:preferredNamespaceUri 
> "http://w3id.org/vir#; ;
>
> Best,
>
> Nicola
>
> --
> Nicola Carboni
> Visual Contagions
> Digital Humanities - dh.unige.ch
> Faculté des Lettres
> Université de Genève
> 5, rue de Candolle
> 1211 Genève 4
>
> On 15 Dec 2021, at 11:58, George Bruseker via Crm-sig wrote:
>
> Dear all,
>
> I am wondering if anybody else struggles with what official namespace ot
> use for the CRM extensions. I'm not really sure how the situation stands.
> Should the minisites for each extension have a prominent place where they
> display the namespaces just so we all follow the same procedure? Do I miss
> what is already there?
>
> Best,
>
> George
> ___
> Crm-sig mailing list
> Crm-sig@ics.forth.gr
>
> http://lists.ics.forth.gr/mailman/listinfo/crm-sig
>
>
___
Crm-sig mailing list
Crm-sig@ics.forth.gr
http://lists.ics.forth.gr/mailman/listinfo/crm-sig

Re: [Crm-sig] Official NameSpaces of CRM Extensions?

2021-12-20 Thread Nicola Carboni via Crm-sig 

Dear George,

The namespace to be used should be the `xml:base` value in the RDF 
document. Example:

```xml
http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; xml:lang="en" 
xml:base="http://www.cidoc-crm.org/cidoc-crm/CRMsci/;>

```
```xml
http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
xml:base="http://www.ics.forth.gr/isl/CRMgeo/; xml:lang="en">

```

The confusion started because the namespace has changed over time

CRMdig 3.2.2 had

```xml
http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
xml:base="http://www.ics.forth.gr/isl/CRMext/CRMdig.rdfs/; 
xml:lang="en">

```
The latest version is
```xml
rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#; 
xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#; 
xml:base="http://www.ics.forth.gr/isl/CRMdig/; xml:lang="en">

```

Generally they are both documented in [prefix.cc](http://prefix.cc), 
hence someone is still using the old ones.



For clarifying the confusion, It is possible to write explicitly in the 
RDF itself the preferred namespace and prefix, using the properties 
vann:preferredNamespaceUri and vann:preferredNamespacePrefix . Example 
(in ttl) from [VIR](http://prefix.cc) :


```ttl
vann:preferredNamespacePrefix "vir" ;
vann:preferredNamespaceUri "http://w3id.org/vir#; ;
```

Best,

Nicola



--
Nicola Carboni
Visual Contagions
Digital Humanities - dh.unige.ch
Faculté des Lettres
Université de Genève
5, rue de Candolle
1211 Genève 4

On 15 Dec 2021, at 11:58, George Bruseker via Crm-sig wrote:


Dear all,

I am wondering if anybody else struggles with what official namespace 
ot
use for the CRM extensions. I'm not really sure how the situation 
stands.
Should the minisites for each extension have a prominent place where 
they
display the namespaces just so we all follow the same procedure? Do I 
miss

what is already there?

Best,

George
___
Crm-sig mailing list
Crm-sig@ics.forth.gr
http://lists.ics.forth.gr/mailman/listinfo/crm-sig___
Crm-sig mailing list
Crm-sig@ics.forth.gr
http://lists.ics.forth.gr/mailman/listinfo/crm-sig