Re: FHIR on schema.org

On 05/26/2016 08:16 PM, Renato Iannella wrote:
>
>> On 26 May 2016, at 17:42, Marc Twagirumukiza
>> <marc.twagirumukiza@agfa.com <mailto:marc.twagirumukiza@agfa.com>> wrote:
>>
>> However it's the responsibility of every one to make its data public
>> or not.
>> This doesn't prevent us to provide a*way of expressing data*(for those
>> who want to do so) with FHIR standard usingschema.org <http://schema.org/>
>
> I think the onus is on the spec development side to show how privacy
> issues are addressed (mitigated).
> Hence, using “Privacy-By-Design” principles [1] (for example).
>
> The current FHIR core spec uses a secure protocol for exchange of data
> (ie good design) for XML/JSON.
> But if we then say - here is how you encode FHIR data in public web
> pages and publish schema.org <http://schema.org> URIs - then we must be
> able to specifically address these privacy concerns. (I imagine a lot of
> Privacy Advocacy groups would be interested if they saw that.)

I think it is important to distinguish two separate and orthogonal concerns:

  1. What data should be shared or exchanged?

  2. What does the data mean?

Security and privacy are all about #1 -- not #2.   Schema.org and 
healthcare vocabularies address #2 -- not #1.  They are all about 
creating a shared understanding of what the data means, in order to 
achieve interoperability between parties that have already decided to 
exchange data.  This is completely orthogonal to the question of whether 
that data should be shared or exchanged.

Privacy-By-Design principles are very relevant to protocol 
specifications, but they are not very relevant to vocabularies whose 
purpose is to enable shared understanding of data meaning.

David Booth


>
>> Just for your example, we are already expressing our internal
>> healthcare data (and EHR data) usingschema.org
>> <http://schema.org/>(although they are not public).
>> This has a benefit when we need to share such data with another APIs
>> and there HL7 FHIR comes in as a standard.
>
> Does that mean you are encoding FHIR Data using RDFa/Microdata? (and
> using schema.org <http://schema.org> URIs for all the FHIR concepts?)
>
> Cheers - Renato
>
> [1] https://www.w3.org/TR/2012/NOTE-app-privacy-bp-20120703/

Received on Friday, 27 May 2016 01:10:09 UTC