Re: ID namespace for composer and healthcare facility

[Dileep V S]

Thanks Thomas for the additional information

regards
Dileep V S
*Founder*
HealtheLife Ventures LLP
m: +91 9632888113
a: 106, Innovation Centre, IIIT, Electronics City, Bangalore 560100
w: healthelife.in  e: dil...@healthelife.in


On Thu, Dec 13, 2018 at 3:26 PM Thomas Beale 
wrote:

> At the moment I think you can safely default the namespace attribute to a
> value like "EHR", "Demographic", "Terminology" etc, which acts as a service
> name or type.
>
> Now, if you look at the OBJECT_ID sub-types
> ,
> you'll see that they are either UID-based - i.e. based on GUID or maybe OID
> (avoid if humanly possible), or reverse domain names. These are all already
> globally unique.
>
> The subtypes ARCHETYPE_ID, TEMPLATE_ID (not in use) are also globally
> unique. To make archetype ids properly globally unique, they can/should
> have true namespaces prepended, as described here
> .
> Section 7 of that doc gives you some examples. The class that defines those
> identifiers is ARCHETYPE_HRID. All of these id types are for design
> artefacts, so I think not your primary concern right now. The type
> TERMINOLOGY_ID is the same, but less well controlled, but usually reliably
> global, because major terminologies tend to be in global use.
>
> The remaining type, GENERIC_ID, is assumed to be a string with a scheme
> type that makes it unique, but might not. The example is things like
> patient hospital ids. Ideally, national patient ids, social security
> numbers etc are unique.
>
> - thomas
> On 13/12/2018 03:14, Dileep V S wrote:
>
> Hi Thomas,
> Thanks for the info.
>
> Just to clarify my understanding, you feel that the namespace attribute in
> for information only and so can be set as we choose.
>
> However, I am not sure that I understand your statement "all target
> OBJECT_IDs of the various concrete types are already globally unique". Do
> you mean using UUIDs for EHRID, PersonID etc? Can you elaborate some more
> one how this is being managed?
>
>
> --
> Thomas Beale
> Principal, Ars Semantica 
> Consultant, ABD Project, Intermountain Healthcare
> 
> Management Board, Specifications Program Lead, openEHR Foundation
> 
> Chartered IT Professional Fellow, BCS, British Computer Society
> 
> Health IT blog  | Culture blog
>  | The Objective Stance
> 
> ___
> openEHR-technical mailing list
> openEHR-technical@lists.openehr.org
>
> http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org
>
___
openEHR-technical mailing list
openEHR-technical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org

Re: ID namespace for composer and healthcare facility

[Thomas Beale]

At the moment I think you can safely default the namespace attribute to 
a value like "EHR", "Demographic", "Terminology" etc, which acts as a 
service name or type.


Now, if you look at the OBJECT_ID sub-types 
, 
you'll see that they are either UID-based - i.e. based on GUID or maybe 
OID (avoid if humanly possible), or reverse domain names. These are all 
already globally unique.


The subtypes ARCHETYPE_ID, TEMPLATE_ID (not in use) are also globally 
unique. To make archetype ids properly globally unique, they can/should 
have true namespaces prepended, as described here 
. 
Section 7 of that doc gives you some examples. The class that defines 
those identifiers is ARCHETYPE_HRID. All of these id types are for 
design artefacts, so I think not your primary concern right now. The 
type TERMINOLOGY_ID is the same, but less well controlled, but usually 
reliably global, because major terminologies tend to be in global use.


The remaining type, GENERIC_ID, is assumed to be a string with a scheme 
type that makes it unique, but might not. The example is things like 
patient hospital ids. Ideally, national patient ids, social security 
numbers etc are unique.


- thomas

On 13/12/2018 03:14, Dileep V S wrote:

Hi Thomas,
Thanks for the info.

Just to clarify my understanding, you feel that the namespace 
attribute in for information only and so can be set as we choose.


However, I am not sure that I understand your statement "all target 
OBJECT_IDs of the various concrete types are already globally unique". 
Do you mean using UUIDs for EHRID, PersonID etc? Can you elaborate 
some more one how this is being managed?




--
Thomas Beale
Principal, Ars Semantica 
Consultant, ABD Project, Intermountain Healthcare 

Management Board, Specifications Program Lead, openEHR Foundation 

Chartered IT Professional Fellow, BCS, British Computer Society 

Health IT blog  | Culture blog 
 | The Objective Stance 

___
openEHR-technical mailing list
openEHR-technical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org

Re: ID namespace for composer and healthcare facility

[Dileep V S]

Hi Thomas,
Thanks for the info.

Just to clarify my understanding, you feel that the namespace attribute in
for information only and so can be set as we choose.

However, I am not sure that I understand your statement "all target
OBJECT_IDs of the various concrete types are already globally unique". Do
you mean using UUIDs for EHRID, PersonID etc? Can you elaborate some more
one how this is being managed?

regards
Dileep V S
*Founder*
HealtheLife Ventures LLP
m: +91 9632888113
a: 106, Innovation Centre, IIIT, Electronics City, Bangalore 560100
w: healthelife.in  e: dil...@healthelife.in


On Wed, Dec 12, 2018 at 5:51 PM Thomas Beale 
wrote:

> Hi Dileep,
>
> as far as I know, all target OBJECT_IDs of the various concrete types
> are already globally unique. I suspect the namespace attribute will not
> be used in the future, or will have just informational value. It would
> be useful to know what other implementers currently do with this attribute.
>
> - thomas
>
> On 12/12/2018 11:18, Dileep V S wrote:
> > Hi,
> > Compositions require id_namespace for composer and healthcare
> > facility, along with id & id_scheme, to uniquely identify them. How do
> > we ensure universal uniqueness for such name spaces? Is there any
> > central registry where EHR systems are supposed to register their
> > namespaces?
> >
>
>
> ___
> openEHR-technical mailing list
> openEHR-technical@lists.openehr.org
>
> http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org
>
___
openEHR-technical mailing list
openEHR-technical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org

Re: ID namespace for composer and healthcare facility

[Thomas Beale]

Hi Dileep,

as far as I know, all target OBJECT_IDs of the various concrete types 
are already globally unique. I suspect the namespace attribute will not 
be used in the future, or will have just informational value. It would 
be useful to know what other implementers currently do with this attribute.


- thomas

On 12/12/2018 11:18, Dileep V S wrote:

Hi,
Compositions require id_namespace for composer and healthcare 
facility, along with id & id_scheme, to uniquely identify them. How do 
we ensure universal uniqueness for such name spaces? Is there any 
central registry where EHR systems are supposed to register their 
namespaces?





___
openEHR-technical mailing list
openEHR-technical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org

ID namespace for composer and healthcare facility

[Dileep V S]

Hi,
Compositions require id_namespace for composer and healthcare facility,
along with id & id_scheme, to uniquely identify them. How do we ensure
universal uniqueness for such name spaces? Is there any central registry
where EHR systems are supposed to register their namespaces?

regards
Dileep V S
*Founder*
HealtheLife Ventures LLP
m: +91 9632888113
a: 106, Innovation Centre, IIIT, Electronics City, Bangalore 560100
w: healthelife.in  e: dil...@healthelife.in
___
openEHR-technical mailing list
openEHR-technical@lists.openehr.org
http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org