Re: Do we really need a notion of profile "conformance"?

Simon,

Actually, what I noticed most wasn't the "document" part but the
"recommendations, requirements and conformance tests for those
requirements" which is quite a bit more specific than "how something
should be done". That latter has no mention of requirements or
conformance or tests, and "how" is pretty broad. I think the key here is
that without a qualifier, "specification" covers a lot of ground, some
of it having nothing to do with IT activites.  After all, examples of
"how things should be done" include:

- a recipe for chocolate pudding
- the "parts fly together" diagrams that come with Ikea purchases
- a table-setting description from Godey's Ladies book
- etc.

If we do not define anything beyond the dictionary sense of "how
something *should* be done" (not *must*) then a statement that a profile
is constraints on a specification could not reliably provide any measure
of conformance.

I also notice that the OGC definition uses "and" which could mean that
anything without all three of those elements would NOT be considered a
specification, rather than those being examples of the characteristics
of specifications. I cannot tell if that is what is intended.

I'd prefer to limit our scope to information technology, and even within
that to a subset of that environment.

kc

On 6/19/19 2:18 PM, Cox, Simon (L&W, Clayton) wrote:
> Yeah - that probably needs fixing. 
> That document is over 10 years old now, and these days we would definitely distinguish the 'specification' from the 'specification document'. 
> ISO (which strongly influenced OGS processes in the first 20 years) is very very document-centric - the PDF is the point of truth. 
> OGC has grown up a bit now. 
> Simon 
> 
> -----Original Message-----
> From: Karen Coyle [mailto:kcoyle@kcoyle.net] 
> Sent: Wednesday, 19 June, 2019 11:03
> To: public-dxwg-wg@w3.org
> Subject: Re: Do we really need a notion of profile "conformance"?
> 
> The OGC definition is:
> 
> "4.21 specification
> document containing recommendations, requirements and conformance tests for those requirements" [1]
> 
> Which is narrower than (I almost said "more specific than") the common language usage, which is something on the order of:
> 
> "a detailed description of how something should be done, made, etc.:
> All products are made exactly to the customer's specifications.
> A specification has been drawn up for the new military aircraft.
> a job specification
> The cars have been built to a high specification (= a high standard)." [2]
> 
> Given the variability between those two definitions it seems that we would need to be clear exactly how it is being used in DXWG documents.
> 
> kc
> [1] https://portal.opengeospatial.org/files/?artifact_id=34762
> [2] https://dictionary.cambridge.org/dictionary/english/specification
> 
> 
> On 6/18/19 5:16 PM, Cox, Simon (L&W, Clayton) wrote:
>> 'Specification' is used by OGC as a generalized abstraction of technical standards/profiles/whatever. 
>>
>> OGC has a policy directive 'The Specification Model - A Standard for Modular specifications' which lays out an overall structure for building specifications based on reusable modules from existing specs. The key goal in this case was to formalize the way that groups of requirements are chunked up so that conformance certificates are at a useful level of granularity. But the overall principle is that each new specification has dependencies, and that these should be explicit and managed. 
>>
>> https://www.opengeospatial.org/ogc/policies/directives
>> https://portal.opengeospatial.org/files/34762
>>
>> -----Original Message-----
>> From: Thomas Baker [mailto:tom@tombaker.org]
>> Sent: Wednesday, 19 June, 2019 07:09
>> To: Antoine Isaac <aisaac@few.vu.nl>
>> Cc: public-dxwg-wg@w3.org
>> Subject: Re: Do we really need a notion of profile "conformance"?
>>
>> On Tue, Jun 18, 2019 at 09:35:18PM +0200, Antoine Isaac wrote:
>>> In particular I like very much the notion of 'specification' rather 
>>> than 'document'.  'document' points to a rather concrete instance of 
>>> a profile, while 'specification' is more conceptual, and allows to to 
>>> group things together. For example, 'specification' allows me to 
>>> consider the Europeana Data Model as a specification that can be 
>>> detailed in an XML Schema, an RDFS/OWL document, or a SHACL 
>>> represention. If we jump straight to the definition, then all the 
>>> 'instanciations' of the Europeana Data Model would live in splendid 
>>> isolation. There are many more downsides to focusing on 'documents'
>>> but this one seems a major showstopper to me.
>>
>> +1 to "specification" with this nice explanation of the distinction.
>>
>> Tom
>>
>> --
>> Tom Baker <tom@tombaker.org>
>>
>>
>>
> 
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net
> skype: kcoylenet
> 

-- 
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net
skype: kcoylenet

Received on Thursday, 20 June 2019 23:00:44 UTC