Hi Richard,
Thank you very much for your comments!
On 22/1/2014 1:43 ??, Richard Light wrote:
On 21/01/2014 18:37, martin wrote:
Hi Richard,
CRM-SIG distinguishes the authoritative textual definition of the
CIDOC CRM, which is at regular intervals (5 years...) submitted to
ISO as draft for updating ISO21127, from implementation
recommendations. Global compatibility with the CRM is not defined at
any encoding level, as explained in the standard itself.
Implementation recommendations are much narrower in scope than the
CRM definition. The
guideline in [1] _is a valid CRM_SIG recommendation_. It serves
compatibility in a narrower RDFS world.
THe RDFS version of the CRM itself is a recommendation, and not part
of the DEfinition of the CRM.
The recommendation in [1] should be integrated into the next RDFS
version recommended by CRM-SIG.
Martin,
I'm perfectly happy with the principle that the authoritative CRM
floats above implementation details, and is the document which goes
through the ISO review process. I'm also happy with the idea that
implementation recommendations are more specific in scope.
However, I don't think that this should lead to a sense that
implementation recommendations are in any sense second-class citizens
in the CRM world, or that they are "optional extras".
That's not meant! Simply the scope of use cases to which they apply are
much narrower than the CRM itself. For instance, when the rdf:literal is
a better choice to transport time than R82a,b etc., has not been defined
more rigorously. That should be done. Smaller scope is not second class :)
Current shortcomings:
a) The CRM-SIG Website needs a better way of making its
recommendations visible. This is envisaged by mid 2014.
Certainly, the current approach is confusing. For example, the
various RDF[S] versions of the CRM, plus the guideline [1] which I
mentioned, appear on the page called "Official release of the CRM".
As you say, they are /not/ the official CRM, so the entries for each
then have to say "this is NOT a definition of the CRM ...". In fact,
looking down the "Official release" page, most of the entries are in
fact different RDF or OWL encodings. :-) Given that there is a page
"Previous versions", and a page "Working drafts", I can't see why the
"Official release" page should ever have more than one thing on it.
(This is perhaps being unkind - I can see how these pages have
developed over time, and for example "Working drafts" appears to have
been abandoned some years ago.)
If there was an actual page called "Resources" (and not just a menu
heading), it could explain the CRM ecosystem clearly and succinctly,
with links to each type of resource. If there was a page called
"Implementation recommendations", the various RDF encodings of the CRM
could go there. This would remove the need for the apologetic "this
is NOT the CRM" notices.
In general, I think this part of the site is crying out for a succinct
explanation of what is there, and why it is there. At present it is
just a list of resources for those "in the know".
Excellent, we presented a new lay-out for the Website on the last
CRM-SIG and asked for comments. We'll check that your points are or will
be met.
b) The exact (functional!!) scope of implementation recommendations
has not been defined. Here someone could identify quite distinct use
cases.
I think we need to start with a firm, /positive/, definition of what
an "implementation recommendation" (henceforth "IR") is. (Something
better than "this is NOT a definition of the CRM".)
This statement is absolutely necessary. Unfortunately, people read only
a small fraction of what the SIG provides.
Otherwise, you're right!
We then need to state which IRs we have done work on, and present the
current version and history of each. I suggest that we use the same
page layout and structure for the official CRM and for each IR, e.g.:
* short description of scope and purpose
* list of those responsible; contact details
* current official version
* working drafts; work in progress
* earlier versions
right, this is what we plan. With the man-power of voluntary work we
have, keeping versions i sync is not always possible. There is no
software on the market that could do such thing.
Am I right in thinking that the RDF[S] work is the only IR that we
actually have?
No, not right. We have guidelines for RDBMS.
I would define the use case for the RDF version of CRM as being "to
enable CRM concepts to be used in Linked Data applications". Discuss ...
Right! Would you write up such thing?
As well as adopting a positive approach to the documentation of IRs, I
think we also need to ensure that their development is subject to the
same rigour as the core CRM. For example, the recommendation [1] was
agreed in November 2011, and so should have been taken into account in
the approved RDF encoding dated November 2012. This suggests that we
maybe need a separate issue tracking system for each IR, as well as
one for the core CRM.
Correct. Actaully the approved encoding needs a use case definition as
well...
Help welcome!
I hope the above is helpful.
Yes indeed!
As regards the web site, didn't you say that only you could update it?
No, I said we couldn't find people wanting to invest real effort into
this in collaboration with us.
We should find a way to distribute editorial work. A business model for
that is needed, which is attractive.
If so, all we can do is make suggestions as to what changes should be
made. In general, I would suggest that "less is more" - imagine the
site from the perspective of someone who lands there for the first
time, wanting to find out about the CRM, and give them a clearly
explained path through to relevant information with as few clicks as
possible.
No chance to make it smaller. We checked everything.
Would you like to go through our redesign document and really provide
feed-back?
Anyone else willing to do so?
Best,
Martin
Best wishes,
Richard
Best,
Martin
On 21/1/2014 7:34 ??, Richard Light wrote:
Hi,
Does that mean that the conclusions of [1] are not to be supported
by the CRM itself? If you recall, we have discussed the
subproperties P81a etc. and pointed out that they are not in the
official CRM. If this is to remain the case, where does that leave
the advice in [1]?
Thanks,
Richard
[1]
http://www.cidoc-crm.org/docs/How_to%20implement%20CRM_Time_in%20RDF.pdf
On 20/01/2014 09:21, Chryssoula Bekiari wrote:
Dear All
You may find the new draft version 5.1.2 of CIDOC CRM in
http://www.cidoc-crm.org/official_release_cidoc.html
FYI
Chryssoula
--
*Richard Light*
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Research Director | Fax:+30(2810)391638 |
| Email:[email protected] |
|
Center for Cultural Informatics |
Information Systems Laboratory |
Institute of Computer Science |
Foundation for Research and Technology - Hellas (FORTH) |
|
N.Plastira 100, Vassilika Vouton, |
GR70013 Heraklion,Crete,Greece |
|
Web-site:http://www.ics.forth.gr/isl |
--------------------------------------------------------------
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
--
*Richard Light*
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
--
--------------------------------------------------------------
Dr. Martin Doerr | Vox:+30(2810)391625 |
Research Director | Fax:+30(2810)391638 |
| Email: [email protected] |
|
Center for Cultural Informatics |
Information Systems Laboratory |
Institute of Computer Science |
Foundation for Research and Technology - Hellas (FORTH) |
|
N.Plastira 100, Vassilika Vouton, |
GR70013 Heraklion,Crete,Greece |
|
Web-site: http://www.ics.forth.gr/isl |
--------------------------------------------------------------
