Dear Ian,
I wrote I will consider it.
I can accept your proposition.
It is factually the truth.
Gerard
Gerard Freriks
+31 620347088
gfrer@luna.nl
Many thanks Gerard,
Much appreciated,
Ian
Hi Ian,
I am a bit puzzled by the communications with Gerard you had yesterday.
You seem very pleased with the statements that Gerard did yesterday about the
ownership of OpenEHR foundation, an ownership would not anymore be called
"proprietary". For me is owning pretty much the same as proprietary when it is
about companies. But there seems to to be for you a pleasing difference.
Don't get me wrong, I am happy for you that you are pleased, and I think I
will not disturb that feeling with unpleasant thoughts.
Because I think that you have a good logical reasoning about how Gerard's
statement is an answer to your concern you expressed last Friday, which was
very much OpenEHR-specification-related.
It is only that I don't get it.
I still don't know how Gerard at this moment thinks about the openness of the
OpenEHR specifications.
Do you know, Ian? You cared a lot about this last Friday, so how are your
feelings about this now? Are you also pleased?
Since this is a public mailinglist, in which it is natural that the public is
able to understand the reasonings, it maybe the case that a significant part of
the public reading here does not have enough information to understand why you
are so pleased.
You expressed also concern about what potential customers would think about
the openness of OpenEHR specifications, and Erik even mentioned the word FUD.
Do you think that now potential customers are satisfied by the proceedings
evolving yesterday and their worrying is over?
Thank you very much for further enlightenments.
Best regards
Bert
Hi Bert,
My primary concern was to establish that the use of ‘proprietary’ in the context of openEHR, particularly in the context of specifications / archetypes but also in terms of governance, gives the impression (perhaps unintended), that
a) the specifications and archetypes are licensed using a closed-source license.
b) that ownership of the IP rests with a commercial organisation.
both of which are misleading ,and I am very grateful to Gerard for acknowledging this difficulty.
Gerard, Gunnar and others express a perfectly legitimate concern that the current ‘ownership’ of openEHR rests with a single organisation i.e UCL, and would prefer that a broader member-ownership model would be a better alternative. I am sure the Board of Governors would be very happy to see some proposals to be put forward on this basis, but I also think it was correct for the previous Board to continue with the current arrangement while the new elected MB gets up to speed and hopefully gets things moving at a practical level.
There are other more complex issues around the exact flavour of licensing which some of us feel can act as a deterrent to some potential customers. That is also understood and being discussed by the MB but it does need careful thinking through, as there may be some unintended consequences if we get it wrong. We also have to recognise that against those perceived deterrents, a large number of commercial and governmental organisations are quite happy to use and endorse openEHR so these arguments are quite nuanced.
So am I happy that the worrying of potential customers is over? Of course not!
We still have work to do to understand if changing the licensing arrangements and high-level governance will further reduce these concerns.
We still have to pursue alignment with SDO’s to see if that too will reduce anxieties, but we have to ensure that any such alignment does not conflict with our chosen methodology / philosophy.
I think we also have to recognise that some of these fears are groundless and based on risk-aversion to open-source in general or bodies that do not have ‘official’ SDO status, when such fears are generally diminishing throughout industry.
I am sure there will be ongoing robust discussions about the best approach but I am very encouraged that there seems to me to be 100% consensus on the end goal - open specifications, open governance and agile methodology.
Ian
Hi!
ND on the specification documents is not a big or urgent problem if there are Apace 2 licenced computable artifacts like UML-files/descriptions of all classes, ADL/AQL grammars, openEHR term lists/vocabularies and other things needed for building actual systems. I believe that is already the case (or at least the intention).
In W3C and similar organizations the “ownership” and ways to influence the organization is probably a bit more clear than openEHR.
As a Swede it is easy to compare the governance of openEHR to the constitutional monarchy I live in 
A tradition-based non-elected royal family that most people kind of like and don’t mind and then the real action/power lies in the elected parliament. It is just that the division of power and responsibilities is a bit more obvious in the nation than in openEHR. Is there a clear document describing the division of power between the “royal family” and the elected management board in openEHR?
Having non-elected initial founders in the top of open source projects is not uncommon and often works well. The forking-possibly and many other things contribute to this working quite well, see the list at https://en.m.wikipedia.org/wiki/Benevolent_dictator_for_life for examples. However Standards-organization-thinking-people are likely less used to this kind of thinking, and this creates confusion or concern.
So I’d say either clarify/modify governance or make licences so open that concerns about governance become irrelevant.
The potential confusing effect of what SA (share alike) on the archetypes legally means for some kinds of use inside non-open-source systems is another, probably more burning issue that has already been discussed, I’ll try to avoid that thread right now.
//Erik Sundvall
P.s. David Ingram is very nice and fitting in the role of a founding king, don’t you think? 
Hi Ian, thanks for your answer.
I do not completely agree
My primary concern was to establish that the use of 'proprietary' in the
context of openEHR, particularly in the context of specifications /
archetypes but also in terms of governance, gives the impression (perhaps
unintended), thata) the specifications and archetypes are licensed using a closed-source
license.
We know that this is not true, but it is not explicitly acknowledged by
Gerard. His wording is vague.
Now is Gerard not that important (or is he? I start to wonder), so we don't
need to push for that. If he does not want to explicitly and clear state that
the OpenEHR license is not proprietary, we have to deal with that. And for me
that is easy, just ignore it.
It just that I do not agree that his contribution from this respect yesterday
was helpful.
b) that ownership of the IP rests with a commercial organisation.
That is no problem. Not-for-profit organizations can have proprietary licenses
and commercial organizations can have non-proprietary licenses.
It is easy to find examples. See RedHAT, Oracle, IBM, Microsoft, they all have
non-proprietary licenses. HL7 has been for years a specification-organization
which was not free, which was proprietary of its members. Only since five years
or so, it is a free non proprietary standard/specification organization.
So regarding this second point, the statement which you found helpful has
also not of much value.
Hi Ian, thanks for your answer.
I do not completely agree
My primary concern was to establish that the use of 'proprietary' in the
context of openEHR, particularly in the context of specifications /
archetypes but also in terms of governance, gives the impression (perhaps
unintended), thata) the specifications and archetypes are licensed using a closed-source
license.
We know that this is not true, but it is not explicitly acknowledged by
Gerard. His wording is vague.
Now is Gerard not that important (or is he? I start to wonder), so we don't
need to push for that. If he does not want to explicitly and clear state that
the OpenEHR license is not proprietary, we have to deal with that. And for me
that is easy, just ignore it.
It just that I do not agree that his contribution from this respect yesterday
was helpful.
b) that ownership of the IP rests with a commercial organisation.
That is no problem. Not-for-profit organizations can have proprietary licenses
and commercial organizations can have non-proprietary licenses.
It is easy to find examples. See RedHAT, Oracle, IBM, Microsoft, they all have
non-proprietary licenses. HL7 has been for years a specification-organization
which was not free, which was proprietary of its members. Only since five years
or so, it is a free non proprietary standard/specification organization.
So regarding this second point, the statement which you found helpful has
also not of much value.
Sorry for sending it twice, that was an accident
Bert
we probably need to perform an audit on all of these artefacts to check the licences. One thing we need to change is to allow more types of software licence, e.g. AGPL. Large companies and huge health institutions like the NHS simply cannot expect to be able to use everything for free when it costs quite serious investment on the part of typically small companies or research groups in academic settings. They need to consider contributing resources. Viral licences need to be allowed to enable conditional use; if funding is made available, such licences can be converted to other types of licence. - thomas
We must not forget, this discussion (the ND-part) is in the context of the specifications, and specifications is NOT software. If you are going to ask money for the specifications, or membership-construction, OpenEHR is not anymore open. And most important, you cannot close down this version the OpenEHR specs, because you have given it to the world with the right to share it. It will always compete with your paid version on the market. So the AOM-part and other more technical parts will be stable and for free for coming decade. The Reference Model can change version and in a new version close down the free distribution, but many can write a Reference Model. I, personally, consider the Reference Model as least innovative. We have so many Reference Models, Tim Cook created one, Grahame Grieve created one, the guys from 13606 created one. And don’t forget the current version of the Reference Model. Even in the Netherlands there are a few as datamodel which can easily be converted to a two level Reference Model. Bert
Bert,
my comments relate to software only, contributed by companies and other organisations at their own development expense.
It has nothing to do with specifications, nor specification-related computational artefacts (grammars, XSDs, and the like). These are all issued by the foundation, copyrighted to the foundation and will always be free to use under all circumstances, as has always been the case for 15 years. This will never change.
- thomas
Good to be crystal clear about this, with all the FUDders waiting for their chances. Remember, they talk with not so well informed people (govt. bureaucrats, for example) regarding licensing issues.
Bert,
I fail to see the origin of any ambiguity from within openEHR. The specifications have been free and open for 15 years, since 2000 (or soon thereafter, since some were issued around 2002/2003 for the first time, and some later). This has always been clearly stated even in the old copyright notice, as well as in the current CC licence.
All of this is verifiable simply by going to the website and viewing the specifications, and that has always been the case, again, since the beginning.
Therefore, any 'ambiguity' on this subject has been manufactured from (very great) ignorance or else malice, for the agendas of other parties. There's probably not much we can do about this other than rely on the intelligence of potential user organisations to ignore nonsense when it surfaces, and get on with doing real work. If they can't even be bothered to inspect the website and primary materials, they probably are not serious about e-health.
Secondly, Ocean Informatics has nothing specific to do with any of this. It does not have any IP that is special in any way. There are a number of vendor companies that have become Industry Partners, as can be seen on the openEHR.org home page. Most of these companies have contributed some IP, which can be seen either at openEHR@Github <https://github.com/openEHR>or else in other locations that are generally well-signed from the website.
It is not up to Ocean to determine anything to do with openEHR software or other IP status - that is done by the openEHR Management board <http://www.openehr.org/about/management_board>, and/or Board of Governors <http://www.openehr.org/about/board_of_governors>, in cases where legal advice may be required.
I hope this is clearer.
- thomas
It sure is.
I totally agree.
Bert