Interim Statement on Copyright and Licensing of Archetypes

[thomas.beale]

Interim Statement on Copyright and Licensing of Archetypes

Effective processes for the authoring, review, management and quality assurance of archetypes are of fundamental concern to the openEHR Foundation. They are needed to underpin the clinical value of the work of the openEHR community and contribute to its wider success in what is an increasingly complex environment.

Within the wider health and social care communities that openEHR was created to serve, quality and sustainability of the clinical content of the EHR is what matters. To that end, we have placed clinical requirement, rigorous specification, practical implementation experience, and engagement with health care professionals at the centre of the policy and strategy of the Foundation. We also have to keep in mind potential future business models for the wider governance of openEHR, such as in its partnership with IHTSDO.

The Board of the openEHR Foundation has considered representations made to it, seeking early clarification of its policy on copyright and licensing of openEHR archetypes; i.e. archetypes that use the formalisms and tools that have been pioneered, copyrighted and trademarked by the Foundation. We are pleased to see this issue coming to the fore in the membership of the discussion lists as it undoubtedly reflects the progressive uptake of openEHR in different environments as well as increasing interest in the archetypes that the openEHR community is creating.

Our primary goal and responsibility as a Board is to seek to maximise the benefits of openEHR for health care. To that end, we seek to facilitate the creation of the highest quality of archetypes, reviewed and maintained, in and by the Foundation and its members and partners, and made fairly and equally available to all who wish to use them. The copyright, licence and trade-mark policy we adopt should reflect this objective, as clearly and simply as possible.

openEHR’s formalisms are published openly and some have been adopted as EU/ISO standards, giving them wide access and credibility. Implementations now exist for .NET, JAVA and Python platforms. Much tooling to support the archetype life cycle - from collaborative design, through iterative deployment and evaluation in relevant clinical context, to curation and dissemination, in regional or international repositories - has already been written and made available, much as open source.

There is rapidly increasing professional, government, standards body and commercial interest in openEHR and the Foundation is being encouraged into new international partnerships. As the uptake of openEHR outputs increases, it is becoming ever more important for us to proceed cautiously as a Board, consulting within all these communities to ensure that the IP and trademark owned by the Foundation are appropriately safeguarded. Copyright law is not well attuned to curation of digital artefacts and significant changes are under discussion in some countries. These are likely to be more specific about how digital assets may be managed and safeguarded.

In the openEHR archetype domain we have to consider what specific assets and services we seek, as a Foundation and community: a) to brand, b) to centralise, protect and distribute as custodians (on behalf of our own community, or national programmes, or professional bodies), c) to control/regulate version manage and maintain, and d) to market. We are still learning our way towards answering these complex questions, informed by practical experience.

In line with the policies we have adopted on open publication and contribution to standards, we cannot assume that the openEHR Foundation will be the exclusive developers of archetypes, and we cannot assume that every other archetype producer will elect to “deposit” its archetypes with openEHR, for kite-marking, for custody, or for us to take over distribution and maintenance. We also cannot assume that the openEHR Foundation’s CKM will prove the only way to develop and maintain good quality archetypes.

We need to decide what “stamp” of openEHR we should require on artefacts created and managed separately from the Foundation. For example, is it enough for the formalism to be acknowledged in the archetype through standard wording, or should we require that all archetypes produced using openEHR formalism follow its license and copyright rules and practices? And, should we request that such archetypes be assigned to openEHR in some way?

Should we propose a set of instruments that we will use to protect archetypes that we are adopting within openEHR, but which other content developers might also adopt for themselves? For example, could national archetypes be acknowledged as openEHR compliant archetypes, but developed, validated and copyrighted by a suitable national or regional authority, licensed, and then made available from a repository that it regulates? Or, should we distance ourselves from such separate approaches and place openEHR four-square behind an integrated international framework of core and shared archetypes?

All such choices have relevance for the revenue models we might envisage to support the ongoing operations and development of the Foundation, if it continues to grow along its current pathway. We have made it a clear from the outset that we believe that openEHR or something like it is needed and have been keen for - and indeed have worked very hard to promote - inclusiveness, on all fronts open to us.

We have discussed the issues set out above, at length, and they cannot be quickly decided upon, safely. We view it as our role, at this stage, to publish here an interim statement of the policy issues we have identified and the direction of travel we are following, for the Foundation, which is as follows:

1. To meet immediate needs, we are minded to publish archetypes managed at http://www.openEHR.org/knowledge from the Foundation under the Creative Commons license – specifically the Attribute and ShareAlike (CC-BY-SA). This is the same license that Wikipedia is using.
2. We also propose, at a minimum, that the copyright of all archetypes managed at http://www.openEHR.org/knowledge should be assigned to the Foundation. This is needed to ensure that the Foundation can give permission to others to adapt the work (see the CC license for details).

We will continue to listen and consult on the wider issues discussed in this interim statement. We must align the Foundation’s approach with the requirements and plans of our partners in IHTSDO and EuroRec and with the development of the new governance framework and business plan now needed for the Foundation.

We will keep the plan under close review over the period ahead, as we work with EuroRec, IHTSDO and others to fund a major experimental and clinically driven project for clinical content quality assurance, embracing archetypes and terminology.

This interim statement is now on the wiki at http://www.openehr.org/wiki/display/oecom/Archetypes±+Copyright+and+Licensing. Subject to any necessary rethinking as a Board, arising from responses we receive before December 1^st 2009, we plan that it will become official openEHR Foundation policy from January 1^st 2010, when a set of rules covering its implementation will also be published. We will also consider whether and in what form we might usefully propose guidelines for how copyright in archetypes might best be managed in other contexts, such as a) when managed by governments on national or regional servers, b) when managed privately by healthcare organisations, professional bodies or companies, and c) when managed experimentally, eg in research programmes.

Please respond to the statement with your views on the wiki, or on the clinical or technical lists. Please also remember our status, still, as a voluntary movement, operating with very limited resources. So, when you propose new work or costs, please treat your proposal as if you yourself were being asked to contribute or pay for the work involved!

Above all, we are concerned to get this next and crucial step right. But we recognise that an interim statement along these lines has become necessary, if not overdue.

David Ingram, Dipak Kalra, Sam Heard, for the Board of the openEHR Foundation

October 1^st, 2009

[erik.sundvall]

Hi!

Thanks for starting to movie towards the Creative Commons family of licenses!

On Fri, Oct 2, 2009 at 02:10, Thomas Beale
<thomas.beale@oceaninformatics.com> (on behalf of the openEHR board)
posted:

In line with the policies we have adopted on open publication and
contribution to standards, we cannot assume that the openEHR Foundation will
be the exclusive developers of archetypes, and we cannot assume that every
other archetype producer will elect to "deposit" its archetypes with
openEHR, for kite-marking, for custody, or for us to take over distribution
and maintenance. We also cannot assume that the openEHR Foundation's CKM
will prove the only way to develop and maintain good quality archetypes.

I agree, and this the way it should be. Imagine if you always needed
to go through W3C in order to develop or distribute artifacts created
using e.g. the OWL or HTML specifications! In that case we would not
appreciate their usage of the word "open" in open
specification/standard.

We need to decide what "stamp" of openEHR we should require on artefacts
created and managed separately from the Foundation. For example, is it
enough for the formalism to be acknowledged in the archetype through
standard wording, or should we require that all archetypes produced using
openEHR formalism follow its license and copyright rules and practices? And,
should we request that such archetypes be assigned to openEHR in some way?

No, if you declare that openEHR is an OPEN specification project, then
I believe it would be inappropriate to at the same time claim that you
have the authority to say what licenses, wording or publication
procedures people claiming to use the specification to create e.g.
archetypes are permitted to use.

(Think of the OWL or HTML file example again, it is easy to
technically, using validation tools, check if the specification is
followed or not, but W3C does not try to do that for authored OWL or
HTML files through policy and licenses.)

To meet immediate needs, we are minded to publish archetypes managed at
http://www.openEHR.org/knowledge from the Foundation under the Creative
Commons license – specifically the Attribute and ShareAlike (CC-BY-SA). This
is the same license that Wikipedia is using.

I still am waiting to hear anybody clearly explain how the SA (Share
Alike) part would possibly benefit the openEHR community and the broad
uptake of openEHR in health care, more than just a CC-BY license
would.

One downside of adding a SA requirement is that both the board and
implementing organisations will have to deal with a number of
questions regarding what is considered to be a derivative work and
probably need to arrange for special permits/exceptions. This might
limit or encumber the re-use of archetypes in settings you have not
thought of yet.

Is e.g. a graphical user interface originally automatically derived
from a set of archetypes and then manually modified considered to be a
derivative work? In that case, can such a user interface be used in a
commercial product that is not licensed under CC-BY-SA? In a recent
thread we have already discussed this, see e.g.:
http://www.openehr.org/mailarchives/openehr-clinical/msg01486.html

As I mentioned above it would be inappropriate to try to impose
certain licenses on the artifacts/files created using the openEHR
specifications, on the other hand the openEHR foundation could of
course impose certain licenses in order to use the foundations
authoring, validation and distribution services. So yes a CC-BY-SA
would be a possible requirement for using the services, but what value
would it add in addition to CC-BY? Does it not only increase the risk
of pushing organisations that dislike (or that suspect that their
contractors/system vendors would dislike) the SA-part to start their
own repositories using e.g. CC-BY instead?

We also propose, at a minimum, that the copyright of all archetypes managed
at http://www.openEHR.org/knowledge should be assigned to the Foundation.
This is needed to ensure that the Foundation can give permission to others
to adapt the work (see the CC license for details).

Seems OK as a hosting requirement as long as the license gives enough
freedoms (in my view CC-BY, not CC-BY-SA) so that the name on the
copyright only is needed to be mentioned for attribution and is not
required in order to authorize possible non-SA derivative works .

We will continue to listen and consult on the wider issues discussed in this
interim statement.

...

Subject to any necessary rethinking as a Board,

...

Above all, we are concerned to get this next and crucial step right.

Great

Best regards,
Erik Sundvall
erik.sundvall@liu.se http://www.imt.liu.se/~erisu/ Tel: +46-13-286733
(Mail & tel. recently changed, please update your contact lists.)

[Segun_Odujebe]

Dear All,

I have with great interest observed the trend of the discussions on IP. I believe Tom's Statement and Erik's response set a direction which I will like to comment on as well.

I kind of liked Erik's comparison to the W3C standards.

In attributing IP to the OpenEHR Foundation, we need to think long-term and pragmatically.

What part of the body of work should we attribute to the Foundation. In basic context, I think the specifications are the primary works. Now if they are open, it means we have to decide and define what the 'open' in openEHR stands for.

Taking the W3C example, it would have been limiting if the consortium had insisted on attributing IP to every xml DTD and instances at the onset. We would never have seen the pervasive implementation of XMLas we now see.

My point is this: archetypes are derived from instantiating the Specifications the same way DTDs and xml documents are derived from following their respective specifications.

There are declarations within xml that uniquely identify which specification or derivative DTDs are being referenced for any XML instance. I believe we should not reinvent the wheel. The XML family of standards provides us examples to tackle the issues we now face.

We can borrow a leaf also from the Open Group- opengroup.org:

Companies like IBM have been known to submit standards for adoption and further development. However in the Open Group attributes the IP but not works derived from them. I think?

It is good that Tom also mentioned that the likelihood of the CKW is not likely to be the only one.

We expect the openEHR specs to move into mainstream adoption.But to fulfill their potential, we must not place hindrances on the path of would be implementers. We must focus on freedom for derivative work while controlling the standards.

How will commercial implementations play out?

Can or should the Foundation really regulate the creation of archetypes.

One implementer may create archetypes that model their focus which may not apply globally.

Does the Foundation have a right over archetypes which are in essence IPs of their creators though they may have used the openEHR specs.

Isn't the saying 'openEHR archetypes shouldn't be in the wild' paradoxical?Specifications can only be proved in the wild. Innovations that will derive from them can not be predicted or totally controlled.

I hope I made sense!

Thanks.

'
Olusegun

Olusegun Odujebe
Enthusia Consulting Ltd.
Lagos Nigeria.
Skype: segun.odujebe
Email: oodujebe@enthusiaconsulting.com

[David_de_Bhal (David de Bhál)]

Dear All

I would really like to question the IP here.
IMHO, there is really very little IP. All you are talking about is a
rearrangement of what is already the clinical method and you are just
detailing a formalization of this.

Over the years, we have been confronted by various IP lawyers and patenting
experts, at great expense I might add, and I have repeatedly asked people at
the various workshops just who is collecting the data items for these
archetypes. Despite all their advice, the information is in the public
domain and if I chose to add one item to, say, a clinical description of
Gait because it is in the openEHR and was not in my original discrete
opportunistic data collection module, I am not breaching anybody's copyright
- it is all prior art. And this is from someone who has as much if not more
personal commitment to EHRs than anybody else I know.

Part of the problem is that for funding, it is often desirable to have some
kind of definable IP or an idea that can be patented. This corrupts the
entire process.

We collect items of data and just re-present it to the relevant
practitioners in the kids of archetypes that they require for their work
flow. You cannot copyright prior art and what we are all doing is just
re-defining some of this prior art.

The sad part of it all is that the technology is taking us away from
Clinical Medicine and into the realm of high-tech, high-cost medicine which
is for the minority of clinical practice, generally. The clinical focus and
the emphasis on the practice of medicine itself is being lost and we may
well bypass what is really so important in the practice of medicine itself
and the art that goes with it. My feeling is that we need to start at the
coalface where we find ourselves and work back.

So, for what it is worth, and while I admire and respect what openEHR is
doing I don’t think that there is copyright, perhaps, and only perhaps, in
the form of its presentation but not in the knowledge because all you are
doing is formalizing it in your own particular way and if I paraphrase it or
pinch a data item here or there then there is no breach of copyright. It is
only if I take openEHR chapter and verse that you might have a case but it
would be really hard to prosecute and would only lead to a lawyers' picnic
and damage our collective attempts to define where we are going
electronically and to communicate, generally.

I would also proffer that the majority of those involved in this EHR
movement generally, will not live to see the benefits and will more than
likely, have retired long before the vision is even clearly defined.
Pathologists, generally, have been in this space for more than 35 years and
there is little effect from this and there is an inability on their part to
even agree to basic definitions of a distinct and defined vocabulary. The
more I think about it the more I think we are building the Tower of Babel.

Medicine itself is supposed to be an open science with no secrets yet most
doctors I now charge, and quite a bit, mostly, for their services. It would
seem to me to be contrary to the basic principles of the Open EHR Foundation
and unproductive to try to lock up the IP and very unproductive to waste
valuable time to try and define and lock up the collective intellectual
pursuits of all those who have contributed. It is also contrary to the basic
Hippocratic principles which some of us, supposedly, profess.

It is notable that since Medicine has been hijacked over the past thirty or
so years, most of the basic ethical principles of western medicine have been
altered and the majority of the unethical behaviours - six of the seven that
I am aware of - have been discarded. It would be sad and a huge loss if we
were to allow these commercial types to hijack the process while we,
individually, can contribute and sell our wares and services as we see fit
in an open market.

Respectfully

David de Bhál

[mailto:openehr-clinical-bounces@openehr.org] On Behalf Of
segunodujebe@yahoo.com
clinical discussions
of Archetypes

Dear All,

I have with great interest observed the trend of the discussions on IP. I
believe Tom's Statement and Erik's response set a direction which I will
like to comment on as well.

I kind of liked Erik's comparison to the W3C standards.

In attributing IP to the OpenEHR Foundation, we need to think long-term and
pragmatically.

What part of the body of work should we attribute to the Foundation. In
basic context, I think the specifications are the primary works. Now if they
are open, it means we have to decide and define what the 'open' in openEHR
stands for.

Taking the W3C example, it would have been limiting if the consortium had
insisted on attributing IP to every xml DTD and instances at the onset. We
would never have seen the pervasive implementation of XMLas we now see.

My point is this: archetypes are derived from instantiating the
Specifications the same way DTDs and xml documents are derived from
following their respective specifications.

There are declarations within xml that uniquely identify which specification
or derivative DTDs are being referenced for any XML instance. I believe we
should not reinvent the wheel. The XML family of standards provides us
examples to tackle the issues we now face.

We can borrow a leaf also from the Open Group- opengroup.org:

Companies like IBM have been known to submit standards for adoption and
further development. However in the Open Group attributes the IP but not
works derived from them. I think?

It is good that Tom also mentioned that the likelihood of the CKW is not
likely to be the only one.

We expect the openEHR specs to move into mainstream adoption.But to fulfill
their potential, we must not place hindrances on the path of would be
implementers. We must focus on freedom for derivative work while controlling
the standards.

How will commercial implementations play out?

Can or should the Foundation really regulate the creation of archetypes.

One implementer may create archetypes that model their focus which may not
apply globally.

Does the Foundation have a right over archetypes which are in essence IPs of
their creators though they may have used the openEHR specs.

Isn't the saying 'openEHR archetypes shouldn't be in the wild'
paradoxical?Specifications can only be proved in the wild. Innovations that
will derive from them can not be predicted or totally controlled.

I hope I made sense!

Thanks.

'
Olusegun

Olusegun Odujebe
Enthusia Consulting Ltd.
Lagos Nigeria.
Skype: segun.odujebe
Email: oodujebe@enthusiaconsulting.com

[thomas.beale]

David de Bhál wrote:

[Stef_Verlinden1 (Stef Verlinden)]

David

You’re mixing up patent right and copyright which are both forms of intellectual property.

I would really like to question the IP here.
IMHO, there is really very little IP. All you are talking about is a
rearrangement of what is already the clinical method and you are just
detailing a formalization of this.

On Wiki you can find the following description of copyright: “Copyright is a form of intellectual property that gives the author of an original work exclusive right for a certain time period in relation to that work, including its publication, distribution and adaptation, after which time the work is said to enter the public domain. Copyright applies to any expressible form of an idea or information that is substantive and discrete and fixed in a medium. Some jurisdictions also recognize “moral rights” of the creator of a work, such as the right to be credited for the work. Copyright is described under the umbrella term intellectual property along with patents and trademarks.”

and

“Typically, a work must meet minimal standards of originality in order to qualify for copyright, and the copyright expires after a set period of time (some jurisdictions may allow this to be extended). Different countries impose different tests, although generally the requirements are low; in the United Kingdom there has to be some ‘skill, labour and judgment’ that has gone into it.”

The fact that the underlying knowledge is already in the public domain doesn’t mean that the work can’t be original. With your type of reasoning one can argue that copyright never exist for books since all the words used in the book are already in the public domain. In the case of openEHR the work is original and a lot of skills and effort are put into it. Furthermore some of the ideas developed by openEHR are completely new and weren’t in the public domain before.

Even an archetype falls under the definition of copyright. The authors of an archetypes have rearranged knowledge in the public domain in such a manner that this knowledge can be widely used in a standardized manner and they’ve put a lot of their skills and labour into creating such an archetype.

Cheers,

Stef

[Hugh_Leslie1 (Hugh Leslie)]

Hi David

The issue is coming up because people all round the world want to know how they can use the archetypes that are published on the openEHR clinical knowledge manager. Concerns have been raised that its not clear about the licensing of these artifacts.

The intent of all these discussions is not to prove that the archetype contains some unique content that should be patented or owned but to make sure that the archetypes can be used freely by anyone without constraint.

regards Hugh

[Tim_Cook2 (Tim Cook)]

I do not see why there could be any concerns.

http://www.openehr.org/download/copyright.html#dsy19-OE_Commercial

[tonyshannon]

Dear David,

Many thanks for your email, which has stimulated some
discussion...always a good thing.

Though the focus of your email concerns Intellectual Property, you also
raise some very important clinical points, several of which I thought I
might try to reply to.

(Re: the need to change)

The sad part of it all is that the technology is taking us away from
Clinical Medicine and into the realm of high-tech, high-cost medicine which
is for the minority of clinical practice, generally. The clinical focus and
the emphasis on the practice of medicine itself is being lost and we may
well bypass what is really so important in the practice of medicine itself
and the art that goes with it. My feeling is that we need to start at the
coalface where we find ourselves and work back.

This concerning point is worth directly addressing..
Firstly, I try to see the openEHR effort within the wider context that
most/all healthcare systems around the world are under major pressure to
change, to move towards more sustainable and effective healthcare systems.
In order to support healthcare in the 21st century, it is very clear
that we need major change in all aspects of the people, process,
information and technology elements that healthcare systems currently
struggle with.
Some of us believe that current information management practice in
healthcare systems is very wasteful, with much room for improvement.
Therefore within a wider push for leaner and more effective healthcare
systems, information management must be improved and become a key driver
for improvements in systems, to ensure value for money for patients (and
indeed taxpayers).
However....while healthcare pressures require us all to change (inc.
clinicians) I wouldn't want to make a case for better information
management to be aimed at anything other than improving the working
lives of clinical colleagues (try to balance the art and science of
medicine) and the quality of care of patients at the coalface.

The changes required will come from a number of directions.
Attempts to impose solutions top-down on healthcare systems, without
clinical support are often liable to fail. (Note the difficulties
experienced by the UK NHS National Programme for IT, which I am all too
familiar with).
On the other hand, as you and others rightly point out, innovation must
equally come from bottom-up solutions, particularly directed at
supporting those clinical needs at the "coalface".
You may be aware of my drive in recent months to focus some of our
efforts within openEHR at the frontline via 2 routes
1)The push for an open source clinical reference framework and
application i.e. Opereffa (though in its very early stages) is aimed at
frontline clinical use, e.g. SOAP noting.
> http://opereffa.chime.ucl.ac.uk
2) The push for top 10 archetypes in an Emergency, is also related to
the clinical requirements of SOAP noting and Emergency Summaries, which
is slowly being progressed internationally via CKM.
I hope the recent push for a Connectathon next year at MedInfo will
allow us to showcase some of the potential value in these efforts to
clinical colleagues, who rightly are looking for clinical value.
I also encourage all my clinical colleagues who want to make a change to
this landscape, to get behind one or other of these efforts, whatever
their clinical coalface is, to help build momentum....
At the same time, as you rightly suggest our focus must be on supporting
other innovations at the frontline. I am very keen to support other
coalface oriented openEHR innovations in this space and would be
interested in any thoughts/plans you have of your own (eg v-practice
work (?)) in that regard...

(Re: bottom-up innovation)

I would also proffer that the majority of those involved in this EHR
movement generally, will not live to see the benefits and will more than
likely, have retired long before the vision is even clearly defined.
Pathologists, generally, have been in this space for more than 35 years and
there is little effect from this and there is an inability on their part to
even agree to basic definitions of a distinct and defined vocabulary. The
more I think about it the more I think we are building the Tower of Babel.

This challenging point relates to an email exchange in the past.
http://www.nabble.com/openEHR-in-2009-and-beyond..-a-view-of-the-way-forward-td21623065.html

If....if I was to wait for an entirely top-down semantically
interoperable solution to my healthcare systems needs then I agree that
  could be like awaiting a Tower of Babel.
On the other hand, if we have agreed that...
- healthcare systems needs to change
- information management systems are key to improvements
- an international health IT platform to openly share clinically useful
components would be a good (if disruptive) thing
- open standards (+/- open source solutions) are needed for that platform
...then *any* effort to evolve healthcare solutions using archetypes
from the bottom up, appears to me to be a move in the right direction.
To reiterate, this is not to suggest we need to await the openEHR
Foundation/IHTSDO/others to design the perfect solution that is adopted
sometime in the future.. but that we, on the ground/at the coalface,
innovate towards this goal here and now.

While I am fortunate to be still in the first half of my career (I
hope!), so I do hope to see some of the benefits before I retire.
(If you are right that I wont see it fulfilled, I guess I'd still rather
try to work towards a better solution now, than simply accept the status
quo for the next x years.....)

(Re: locks and IP)

Medicine itself is supposed to be an open science with no secrets yet most
doctors I now charge, and quite a bit, mostly, for their services. It would
seem to me to be contrary to the basic principles of the Open EHR Foundation
and unproductive to try to lock up the IP and very unproductive to waste
valuable time to try and define and lock up the collective intellectual
pursuits of all those who have contributed. It is also contrary to the basic
Hippocratic principles which some of us, supposedly, profess.

So moving on from the need to change and innovate to the subject of
"locks" and IP...
I would not/will not support any attempt by the openEHR Foundation to
"lock up the IP" in a way that gets in the way of clinical innovation
and improvement in this space.
If anyone is concerned that the interim statement is getting in the way,
please let me know. My understanding of this move towards Creative
Commons licensing is quite the opposite to be honest...(more on that
shortly)

It is notable that since Medicine has been hijacked over the past thirty or
so years, most of the basic ethical principles of western medicine have been
altered and the majority of the unethical behaviours - six of the seven that
I am aware of - have been discarded. It would be sad and a huge loss if we
were to allow these commercial types to hijack the process while we,
individually, can contribute and sell our wares and services as we see fit
in an open market.

One of the key parts of the discussion around Intellectual Property here
is/should be about "sharing the learning".
As I said I wouldnt want openEHR and any related IP concerns to be
getting in the way of any innovation.

My understanding is that this Interim Statement on Copyright and
Licensing of Archetypes is a way to bring these issues out in the open
to encourage discussion and debate in this area, so the more discussion
from the community the better..

As you suggest, there is some tension between the medical worlds
approach to sharing knowledge and that within the informatics community.
I am certainly familiar with several large commercial health IT vendors,
who take a very propretiary stance to their clinical content
(interestingly much of it in their form of clinical "prior art"). Their
approach is I believe a key part of the current challenge in sharing the
innovation in this science and moreover is a block to the integration of
healthcare systems that needs to occur, and one of the key reasons
openEHR gets my support.

As we clinicians cannot overcome any of these hurdles alone, so we need
to work with technical, managerial, legal and commercial colleagues to
effect the paradigm shift that is required.
While I am as keen as anyone to push open standards and openEHR, we need
to be aware of the commercial reality in which we live and operate,
specifically how informatics projects are given funding and ongoing support.

In that context clarity on the copyright & licensing of our openEHR work
is a necessary and good thing.
While I'm not conversant with the legal detail on copyright or
licensing, I welcome this discussion on these issues.
Indeed as far as I understand them, I welcome the move outlined to move
towards the Creative Commons approach to licensing. No doubt more debate
is needed on this, which is again, why I believe why an Interim
statement was released at this point.

Many thanks again for stimulating this discussion, very happy to discuss
any/all items further..

Kind regards,

Tony

Dr. Tony Shannon
Consultant in Emergency Medicine, Leeds Teaching Hospitals
Clinical Lead, Clinical Content Service, NHS Connecting for Health
Chair, Clinical Review Board, openEHR Foundation
+44.789.988 5068 tony.shannon@nhs.net

David de Bhál wrote:

[David_de_Bhal (David de Bhál)]

We are all for interoperability and we are happy to share information but we
find that other systems are not quite inteoperable.

All the information we collect, for the most part, is tagged and is all HL7
compliant.

We are completely against lock in and undertake to provide our clients with
their information in a useable format.

What I am questioning is that there is any copyright in information that is
in the public domain already.

The clinical method has been around for decades if not centuries and
archetypes seem to be to be a way of formalizing these with a view to
transmitting the information, usably, between systems. Any attempt to lock
this information up as copyright is contrary to the spirit of the idea of
interoperability.

Labeling of this information for its transmission electronically is, in my
opinion, not copyrightable. Lawyers have tried to persuade me to patent our
systems but I have desisted.

The other side of the coin is that there is a duty of care to protect the
information because we are custodians of confidential information.

We are quite ready and willing to communicate but with whom?

The exported documents of the products in the market are not really useable
– they are merely exported as blobs of text and not are really useable
products.

We are, actually, on the same page.

David de Bhál

[David_de_Bhal (David de Bhál)]

Part of all this is the general conflict that arises between business and
medicine.
Doctors, for the most part, are small businessmen who work in isolation. One
of our founding partners has an early MD in Medical Informatics and detests
business and refuses to work in a commercial environment or with business
people. He is, however, a clinician and charges patients for his services.

Let me say that the fruits of our pursuits are available for transmission
between systems if they are able to accept it but we find that exports from
other systems are merely blobs of information, subject to the consent of the
patient. We are able to populate our EHRs from pathology providers for the
most part although they vary quite a bit.

My understanding is that Australia is to legislate that the information in
the medical record is to be the property of the patient and not the medical
practitioner.

We don’t claim any copyright on how we do things although how we do it is in
the class of a trade secret and that is how we propose to recoup our
considerable investment just as medical practitioners practice and provide a
service for consideration.

There has to be, ultimately, a business case for the use of EHRs. If they do
not save time and money, they will not be used. Lawyers, and with them, the
copyrighters and trade markers, have always been the great enemy although
they may be responsible for some of the improvements. We aim to facilitate
the collection by practitioners of the highest quality clinical information
and to 'back fill' so to speak any archetypes that may be required to be
filled and if those items of information are properly labelled then we will
be able to communicate.

What I see happening is the fragmentation of the profession into various
disciplines - EM, Cardiology, Neurology etc along with the records when it
is all part of the same patient system. We put the patient at the center of
the equation and that is the paradigm shift which solves many of the
problems concerning electronic medical records in general. We are trying to
centralise the record while commercial interests are trying to divide us up.

One of the great things about the NHS in Britain was the idea that the
record went with the patient wherever they went, one patient - one record,
and I should like to see that happen with the patient record rather than
disparate systems trying to interpret summaries.

Technologically, we should not need to do this any more and I should like to
see EHRs marketed similarly to how Insurance or Banking services are
marketed. The provision of an EHR is a service which must be provided to the
patient and with patient involvement throughout and along with the sealed
envelope. In the present environment there is only one way to do this.

Ultimately there has to be value for the practitioner and the patient and
there must be some commercial consideration for this.

David de Bhál

[mailto:openehr-clinical-bounces@openehr.org] On Behalf Of Tony Shannon

[Stefan_Sauermann]

This looks similar to the LOINC / SnomedCT question (only in the field
of coding laboratory results):
LOINC comes for free, is done by a voluntary organisation (THANKS A
BUNDLE TO REGENSTRIEF!!)
vs
SnomedCT comes for lots of money.

Both do a good job for coding laboratory results. SnomedCT may win in
other areas, agree.

LOINC does have a better chance to be used by many for lab coding. You
may be lucky and your state pays for Snomed, many (all of us in
Austria!) are not and will therefore not use it.

I therefore vote for the Regenstrief / LOINC business model.
I will contribute to the Regenstrief LOINC work without demanding to be
paid (and I actually am, in a tiny small part).
I will not contribute to SnomedCT (and any other artefact with a similar
business model) without being paid. The pros earn the money, they shall
do the work. They are liable.

Hope this helps, and greetings from Vienna,

Stefan Sauermann

Acting Program Director
Biomedical Engineering Sciences (Master)
University of Applied Sciences Technikum Wien
Hoechstaedtplatz 5, 1200 Vienna, Austria
Tel: +43 (1) 333 40 77 - 988
mobile: +43 (664) 6192555
sauermann@technikum-wien.at

http://www.technikum-wien.at
http://www.healthy-interoperability.at

David de Bhál schrieb:

[erik.sundvall]

Hi!

What I am questioning is that there is any copyright in information that is
in the public domain already.

I think you may have misunderstood some things regarding copyright.

Somebody's written formalisation of public knowledge may very well be
covered by copyright. Writing down popular public stories/knowledge in
a book (or any other published form) will grant you copyright to your
version of them. It's not like a patent where you can prevent others
from using similar ideas though, so others can still formalize/write
down the public knowledge in a similar but slightly different form and
get copyright to their version of it.

Now regarding archetypes we want to share _exactly_ the same
formalisation (not just any somewhat similar ideas based on public
knowledge) and thus there is always a risk of somebody starting to
make a fuzz about copyright. That's why there is a need to officially
waive some of the "rights" e.g. by using a CC-BY licence so that all
future users can feel safe that nobody can come later and give them
copyright problems.

Best regards,
Erik Sundvall
erik.sundvall@liu.se http://www.imt.liu.se/~erisu/ Tel: +46-13-286733
(Mail & tel. recently changed, so please update your contact lists.)

P.s. I am still waiting for somebody to motivate how CC-BY-SA could
possibly bring benefits over CC-BY in the case of openEHR-hosted
archetypes. There must either have been a mistake or some unexplained
motivation behind the choice since the "SA" issue was brought up
before the Iiterim statement was made.

[thomas.beale]

David de Bhál wrote:

We are all for interoperability and we are happy to share information but we find that other systems are not quite inteoperable.

All the information we collect, for the most part, is tagged and is all HL7 compliant.

We are completely against lock in and undertake to provide our clients with their information in a useable format.

What I am questioning is that there is any copyright in information that is in the public domain already.

The clinical method has been around for decades if not centuries and archetypes seem to be to be a way of formalizing

‘formalising’ is the key point here - prior to the archetypes, these typical ideas were not formalised openly in this way.

these with a view to transmitting the information, usably, between systems. Any attempt to lock this information up as copyright is contrary to the spirit of the idea of interoperability.

I think you are missing the point completely here David. What is needed to guarantee openness for an artefact is a situation where:

• if it is published openly it can’t be co-opted by particular interests and made private
• if it is published as a private artefact, it is protected for the author in the normal way
  This does involve a combination of copyright + license. For an explanation of this, which for Gnu software is called ‘copyleft’, see http://www.gnu.org/copyleft/ From that page:

To copyleft a program, we first state that it is copyrighted; then we add distribution terms, which are a legal instrument that gives everyone the rights to use, modify, and redistribute the program’s code or any program derived from it but only if the distribution terms are unchanged. Thus, the code and the freedoms become legally inseparable.

Proprietary software developers use copyright to take away the users’ freedom; we use copyright to guarantee their freedom. That’s why we reverse the name, changing “copyright” into “copyleft.”

Labeling of this information for its transmission electronically is, in my opinion, not copyrightable. Lawyers have tried to persuade me to patent our systems but I have desisted.

I am not sure what you are referring to here; it is true that labels for transmission can be derived from archetypes, but that happens to be a useful side-effect, but what is being copyrighted is not a specific way of sending data with tags (although systems to do that are always copyrighted, e.g. Snomed, LOINC, etc).

The other side of the coin is that there is a duty of care to protect the information because we are custodians of confidential information.

We are quite ready and willing to communicate but with whom?

The exported documents of the products in the market are not really useable – they are merely exported as blobs of text and not are really useable products.

there is no doubt that this is true…

[erik.sundvall]

Hi Tim and others!

First, thanks for pointing out the new location of the "openEHR Free
Commercial Use License"
(http://www.openehr.org/download/copyright.html#dsy19-OE_Commercial)
since the link http://www.openehr.org/free_commercial_use.htm stated
in the licence of the specifications no longer works.

The issue is coming up because people all round the world want to know
how they can use the archetypes that are published on the openEHR
clinical knowledge manager. Concerns have been raised that its not
clear about the licensing of these artifacts.

The intent of all these discussions is not to prove that the archetype
contains some unique content that should be patented or owned but to
make sure that the archetypes can be used freely by anyone without
constraint.

I do not see why there could be any concerns.

http://www.openehr.org/download/copyright.html#dsy19-OE_Commercial

I can still see the concerns since it was far from obvious that
openEHR-hosted archetypes would be using the same license as the
openEHR specifications. I believe it is a good thing that we discuss
the CC-family of licenses for archetypes.

Second, when now finally finding and getting to read the "Free
Commercial Use License" it seems (to me a non-lawyer) very open, even
more open than the "openEHR Public License" as long as you remember to
display the somewhat longer and more detailed "copyright notice and
disclaimer of the openEHR Warranty" in any works you create based on
the openEHR specifications.

Since the "Free Commercial Use License", without asking for additional
permissions, allows use defined as...

   1. copying
   2. distribution
   3. licensing (including sub-licensing) whether or not for
commercial and/or financial gain and/benefit
   4. modification
   5. adaptation
   6. incorporating the Materials in whole or part into other software
or documentation.

...and since CC-BY includes a disclaimer of warranty (see #5 on
http://creativecommons.org/licenses/by/3.0/legalcode) I don't see a
big practical difference between them. The Creative Commons licenses
are a lot better known, understood and internationally adapted though
so I'd suggest moving also the openEHR specifications (not only
openEHR hosted archetypes) to a CC-BY license unless somebody can come
up with a good motivation not to.

Third, on the OSHIP-list
(https://lists.launchpad.net/oship-dev/msg00559.html) I noticed what
might be described as a "fork" (
http://en.wikipedia.org/wiki/Fork_(software_development) )
including the openEHR specifications named "MLHIM Specifications", see
https://launchpad.net/mlhim. I'd currently guess this move is
unnecessary but I can understand the thinking behind it. Anyway it
proves that the current openEHR specification licensing situation
allows "forks" (wich is a very healthy sign for projects claiming to
be open) and that a possibly increased "fear of forking" if moving the
specifications to CC-BY is irrelevant (since it is possible already
today). If forks in general and this one in particular are
necessary/good/bat etc could preferably be discussed in some separate
thread...

Personally the most useful part of seeing the fork, where CC-BY is
actually applied to the extracted specifications is that, if the
reasoning is correct, I can licence my teaching materials/slides etc
as CC-BY even if they include material directly from the
specifications (provided I include "copyright notice and disclaimer of
the openEHR Warranty" somewhere). Please correct me if I'm wrong.

Best regards,
Erik Sundvall
erik.sundvall@liu.se http://www.imt.liu.se/~erisu/ Tel: +46-13-286733
(Mail & tel. recently changed, so please update your contact lists.)

P.s.
TODO for somebody with openEHR-web write permissions: Please create a
http-redirect or a page on
http://www.openehr.org/free_commercial_use.htm so that it leads to the
license.

[mikael (Mikael Nyström)]

Hi,

This is a general comment and not a specific comment to Stefan Sauermann’s
e-mail.

I have nothing against comments from people stating that they know what the
licence fee for using SNOMED CT is and that the cost is too high for what
they get. However I am quite bored from comments like
IHTSDO-takes-a-licence-fee-for-using-SNOMED-CT-and-I-guess-that-the-licence-
fee-is-extremely-high-so-IHTSDO-is-a-big-evil-empire-making-a-huge-profit.

My own impression is that the license fee for using SNOMED CT is quite
moderate and also adapted to each member country. For example is it free to
use SNOMED CT in the developing countries. It is also free to get licences
for academic use.

IHTSDO, which is the organisation that owns SNOMED CT, is a Danish
not-for-profit association and my impression is that the licence money is
used in a proper way for enhancements of SNOMED CT and to supporting people
that works with enchantments of SNOMED CT.

(I should probably also add that I belong to the people that work with
improving SNOMED CT without being paid by IHTSDO.)

  Greetings,
  Mikael

[mailto:openehr-clinical-bounces@openehr.org] On Behalf Of Stefan Sauermann

[system]

Hi Olesugen

This is a very helpful statement. A couple of potential misunderstandings.

The archetypes on CKM that the community is working on need to be clearly stamped that they come from here because the international interoperability in health care is going to depend on using the same archetypes or their specialisations as the content specifications. We want people to be able to use them and specialise them, but for these to be freely available as well. This is the SA part of the CC-BY-SA licence.

Eric is concerned about the breadth of what might be called derivative works; that is reasonable and we need to make sure the Foundation is not taking holding any copyright or license on anything that is not a specialisation of an archetype. If someone does their own thing it is very important that it does not have the openEHR copyright unless people want to donate it to the community.

Sharing archetypes - exact copies and specialisations - is interoperability. The Schema is fixed.

It seems we have agreement. That we are close to the right space.

The issue that remains is:

Will CC-BY-SA do more good or harm than CC-BY? I suspect not but I am not sure as the benefits and difficulties are in very different spheres. If we start with the CC-BY-SA and then loosen it if there are problems and we cannot make the necessary clarifications, then we can always move to the CC-BY. Starting with CC-BY and then moving to CC-BY-SA would be very problematic from a legal perspective I believe.

Cheers, Sam

[tonyshannon]

Many thanks for those further interesting points David,

We are indeed on the same page in many respects.

I hope openEHR will help move us away from the current fragmentation of
healthcare teams and related systems towards a more patient centred
system (with related EHR).
As you say their must be a business case to support these efforts, we
must absolutely be adding value to the practitioner and the patient's
journey.

My sense is that we have to embrace some legal issues, around the
formalism of clinical content in the record as Tom explains, rather than
the "clinical art" which could be seen as in the public domain.
This seems a necessary step, in order to engage with the commercial
partners who are needed to make this change happen.
I'm certainly no "legal eagle", but the discussion in this thread about
CC-BY versus CC-BY-SA seems an important one we need to work through.

You mention your openness and attempts to share data and interoperate
are frustrating..
Anyone like yourself who is now building healthcare systems using the
openEHR/ISO archetype standard and keen to share information &
archetypes with others in this field are at the pioneering end of the
informatics spectrum and should be proud of their efforts. No doubt
others would be interested in your important lessons learnt.
Aside from the interoperability angle, you may be able to make an
important case on the value of archetypes in saving clinical time and
effort and improved system maintenance through archetype reuse.
Perhaps you might like to join the Medinfo 2010 Connectathon effort in
Capetown to show you stand out from the crowd?

Kind regards,

Tony

David de Bhál wrote:

[Segun_Odujebe]

Dear Sam,

Thanks for the further clarification.

I think we are all in the same book though sometimes on different pages.

I see archetypes as a technology of sorts.Why?

Because the specification is a system of concepts. According to Peter Drucker, technology is a system of concepts.

The fact that there is a lot of discussion so far on IP and licensing shows that we dealing with a technology that is in its infancy.

How we want this spec to grow, depends on how free we make its use.

I believe whatever we come up with, true freedom for innovation should be at the core of it.

We want the openehr standards to living and not dead like some others.

The Foundation and the promoters are doing a great work though. We just need to move forward more rapidly.

Thank you again for your leadership.

Best Regards,

Olusegun