If I could interject here, one of the main reasons for my interest in
dataportability.org is as someone in the life science and healthcare
sector. Google is not the first company to manage healthcare data,
which thus far is anything but portable, but the issues around the
portability of healthcare data as it gets more complex and standards
(like HL7) are developed are still not fully understood. While some
of the dataportability stack does apply to healthcare related
services, one should be careful not to get too dogmatic here. It's
more important to make systems interoperable and resolve the data
ownership issues first.
I am a little concerned about the data portability group considering
data portability as a brand. It shouldn't be considered such.
In the life sciences we have been able to move data from one resource
to another for years (and the data sets are huge and quality is
critical), albeit using rather poorly designed and clunky systems. We
just never called it data portability.
Sorry for the rant, but reading the tech sectors coverage of consumer
genomics and phr's in recent months does that :)
Deepak
On Feb 29, 12:17 pm, nitinborwankar <
nborwan...@gmail.com> wrote:
> Google has recently announced Google Healthhttp://
googleblog.blogspot.com/2008/02/google-health-first-look.html.
> In this post Marissa Meyer says :-
>
> "We are trying to make user data accessible and portable," Marissa
> Mayer, Google's vice-president of search product and user experience,
> says in an interview.
>
> However major issues remain - what touchstone do we use to confirm
> that the data is indeed portable - can I move it freely using common
> formats - can I have a guarantee that data will be deleted when I
> decide to leave the service.
>
> The casual use of the word portable muddles the Data Portability brand
> and is also somewhat cynical 'spin'.
> If Google cared to make health data secure and portable - they would
> enable a platform where all records in the cloud were strong encrypted
> and could only be decrypted using a key not in the cloud but on a
> keychain drive.
> The definitive originals of my medical data would be on my keychain
> drive and all conflicting data would be resolved by reference to my
> original data. Now that would enable accessibility and portability.
>
> Aside from the muddling of the brand Google Health raises a number of
> issues that are may be out of scope for DP as an effort but are so
> close in context that we need to pay very close attention to the
> implications if we are to create DP best practices that people will
> use in real life scenarios like these.
>
> Other posts about this topic are athttp://
gigaom.com/2008/02/28/google-creates-giant-ssn-database/which
> is a rather superficial and confusing description of the issues
> and athttp://
www.scmagazineus.com/Google-Cleveland-hospital-begin-medical-r...
> which is a solid analysis of the problems
> Chris Saad has also just blogged about this athttp://
chrissaad.wordpress.com/2008/02/29/mean-what-you-say-say-what-...