Re: Further work on the Use Cases and Requirements document

Kerry-

responding inline:

On 04/10/2015 02:10 PM, Kerry.Taylor@csiro.au wrote:
>
> Peter,
>
>  
>
> As Alejandro mentioned already, that text comes from the charter.  I am
> assuming “  ISO 19123 abstract model“ substantially means UML, which I think
> it does (please correct me if I am wrong!). 
>

Depends on what the writers had in mind, so I'd hesitate to interpret here.

> Curiously in the “coverage” session at OGC TC Barcelona there was  a rather
>  heated discussion about  what it means to be “conformant” to such a model –
> if I understood it correctly (you were clearly making sense of it at the
> time!) the conclusion was that (in the OGC) it is not even possible to be
> “conformant” to such a model as our charter requires...  Note that afaik this
> is not an issue at all for the W3C.
>

well, conformance means interoperability. It is exactly the point of the
abstract model - it is not meant to be a direct guidance for realizing services,
but establishes abstract concepts (such as "a coverage is a function mapping
sets of direct points to value sets"). This means: implementations will not
understand each other. We had our learning curve here - implementations of WCS
1.x versions normally come with their own client because they are not
interoperable. OGC has done a study on that once (presented at EGU) but I cannot
remember now where to find it. If necessary I can drill for that.

Bottom line, conformance is essential for the Semantic Web and m2m
communication. Think of Haiti: merging of different data was done, but manually.
In future we can do this automatically _if_ the services have a common + clear
udnerstanding of the semantics of coverages, subsetting, etc.


>  
>
> We are most certainly not going to re-invent an XML implementation for ISO 19123.
>

Yes, I understand you will want to go for RDF. But the problem remains, whether
a coverage is expressed in RDF, XML, or JSON or whatever. Note that
"implementation model" does not mean a particular encoding - that is still a
matter of the data format you pick (NetCDF, GML, RDF, etc). It just means "a
definition concise enough for implementations to be interoperable".

>  
>
> My own interpretation of the Charter is that we are developing an OWL (or just
> RDFS) ontology to  represent data that could  also be ISO 19123 discrete
> coverage.  I doubt there is any such implementation already in the OGC or
> anywhere else (although I am aware of some work from Munster that has the
> right flavour, and I believe a few people in the group are working on it).
>

I did not intend to question the Charter, actually. Just stumbled into this
sentence - my apologies for not being aware yet of the exact status of work,
will have to immerse into that.

>  
>
> Does this make the issue go away?
>

unfortunately not - question still is whether W3C intends to join the OGC / ISO
/ INSPIRE caravan or wants to establish an independent definition. Should the
conceptual models align then translating coverages between XML, RDF, etc is
-ahem- almost a non-brainer.

best,
Peter

>  
>
> Kerry
>
>  
>
>  
>
> *From:*Peter Baumann [mailto:p.baumann@jacobs-university.de]
> *Sent:* Friday, 10 April 2015 5:18 PM
> *To:* Frans Knibbe; Alejandro Llaves
> *Cc:* public-sdw-wg@w3.org
> *Subject:* Re: Further work on the Use Cases and Requirements document
>
>  
>
> Hi all,
>
> in the below document there is the stanza "The WG will develop a formal
> Recommendation for expressing discrete coverage data conformant to the ISO
> 19123 abstract model" in 2.5. As already pointed out at the joint W3C/OGC
> meeting this will hinder interoperability: many different implementations of
> the same abstract model are possible, and have been done actually.
>
> But there is more: ISO is currently restructuring 19123 into
> - 19123-1, the current abstract model
> - 19123-2, the coverage implementation model
> the latter will be based verbatim on OGC 09-146r2 which establishes OGC's
> authoritative coverage model. This will be done through a fast track procedure
> (ISO is convinced about the approach, based on both concepts and massive
> implementation evidence), so 19123-2 will be ISO standard on very short notice.
>
> Hence, if W3C will not use OGC's adopted coverage model it will own an
> implementation model which is not interoperable with OGC, ISO, INSPIRE - in
> short: the rest of the world. Therefore, let me suggest to rephrase 2.5
> accordingly. At this stage it does not cost much yet (as usual in the software
> life cycle).
>
> Just wanted to raise awareness of the issue in this group.
>
> cheers,
> Peter
>
>
> On 04/09/2015 04:29 PM, Frans Knibbe wrote:
>
>     Hello Alejandro,
>
>      
>
>     I have just made some changes to the Use Cases and Reuirements document
>     <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html>. I
>     have not put much effort in the textual content yet because I think it is
>     best to get the structure of the document right before putting more meat
>     on its bones. 
>
>      
>
>     The main things I changed today:
>
>       * I have placed the deliverables in a separate chapter.
>       * I have added a chapter that should explain the scoping process.
>       * I have replaced the <details> and <summary> tags, which we can not use
>         because they are not yet standard HTML, with a CSS-based solution for
>         expanding and collapsing text.
>       * I placed a template for use cases in a comment.
>
>     What do you think of the document structure as it is?
>
>      
>
>     Regards,
>
>     Frans
>
>      
>
>     -- 
>
>     Frans Knibbe
>
>     Geodan
>
>     President Kennedylaan 1
>
>     1079 MB Amsterdam (NL)
>
>      
>
>     T +31 (0)20 - 5711 347
>
>     E frans.knibbe@geodan.nl <mailto:frans.knibbe@geodan.nl>
>
>     www.geodan.nl <http://www.geodan.nl>
>
>     disclaimer <http://www.geodan.nl/disclaimer>
>
>      
>
>
>
> -- 
> Dr. Peter Baumann
>  - Professor of Computer Science, Jacobs University Bremen
>    www.faculty.jacobs-university.de/pbaumann <http://www.faculty.jacobs-university.de/pbaumann>
>    mail: p.baumann@jacobs-university.de <mailto:p.baumann@jacobs-university.de>
>    tel: +49-421-200-3178, fax: +49-421-200-493178
>  - Executive Director, rasdaman GmbH Bremen (HRB 26793)
>    www.rasdaman.com <http://www.rasdaman.com>, mail: baumann@rasdaman.com <mailto:baumann@rasdaman.com>
>    tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
> "Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)
>  
>  

-- 
Dr. Peter Baumann
 - Professor of Computer Science, Jacobs University Bremen
   www.faculty.jacobs-university.de/pbaumann
   mail: p.baumann@jacobs-university.de
   tel: +49-421-200-3178, fax: +49-421-200-493178
 - Executive Director, rasdaman GmbH Bremen (HRB 26793)
   www.rasdaman.com, mail: baumann@rasdaman.com
   tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
"Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)