Advantage of ISO

Technical (archive)
41

The one does not exclude the other, I would say.

But on second thought, does ISO prohibit giving a free license, or publishing the specs for free?
I am not sure about that.
I am sure they prohibit publishing their document.

As is with AOM1.4, it is published as ISO's version by ISO (as part of ISO13606) and it is published as OpenEHR's version by OpenEHR , so that can be done.
That both contain the same information.

It is a bit Kafkaesk, but that is normal when bureaucrats get involved.

Bert

42

There are free ISO specifications, like schematron and a handful more.

http://standards.iso.org/ittf/PubliclyAvailableStandards/

You can even ask for an ISO norm to be free. In fact asked for ISO 13606 to be free, but received no answer.

I am happy to debate the relevant merits of the ISO vs. open-source approaches recognising

The one does not exclude the other, I would say.

But on second thought, does ISO prohibit giving a free license, or publishing the specs for free?
I am not sure about that.
I am sure they prohibit publishing their document.

As is with AOM1.4, it is published as ISO’s version by ISO (as part of ISO13606) and it is published as OpenEHR’s version by OpenEHR , so that can be done.
That both contain the same information.

It is a bit Kafkaesk, but that is normal when bureaucrats get involved.

Bert

43

Thank you Ian, that was a very constructive contribution to the discussion. I had started writing a response with some of those thoughs:

44

I forgot to discuss/kill the “fear-of-patent” thought here. At Stackoverflow (http://stackoverflow.com/questions/32010122/are-the-hl7-fhir-hl7-cda-cimi-openehr-and-iso13606-approaches-aiming-to-solve) I wrote:

45

I was very active, 10 years ago, in the anti-software-patent-lobby in Europe. I spent several weeks talking to European members of Parliament. We won. :blush:

I made a presentation for Dutch members of the EP.
http://www.rosa.nl/software_patenten/

It is in Dutch. I made it quick and dirty in OpenOffice version 1.x. So don’t expect a fancy site.

(This was the foreword)

So, I know something about patents. The problem with patents is that they are not easy to find, despite classification systems. It is all about wording and keywords. So the fact that you cannot find a patent disturbing your business does not mean there is no patent.

(This was about finding patents)

It is very important that the OpenEhr foundation, as it actual does, promises in its license that it will not file patent suits against users of its specifications. So we do not need to look for secret patents, the promise is sufficient. This promise is irrevocable.

I too find it unbelievable that people suspect the OpenEhr foundation for having this kind of dirty tricks in their plans. But those people really exist.

Seriously, I even heard people believing the OpenEhr foundation is owning medical data that are stored using OpenEHR specs. Adult serious people, with influence in decisions are saying this. We all know, that medical data in every country are controlled by local law, and the OpenEhr foundation can’t ever be more powerful then the law. People just don’t think well, they don’t add two by two.

In the Netherlands we have books for children, they are called “Jip en Janneke”. The OpenEhr foundation need to explain in “Jip en Janneke” language what really is true and not true.

Maybe a prominent placed link to a FAQ in “Jip en Janneke” language would help.

Maybe I am just accidently meeting the wrong people. That is possible.

46

FUD

Fear Uncertainty Doubt

This is an old technique for making people afraid. Microsoft was always “fudding” Linux. It is a marketing strategy. Thomas mentioned coaching bureaucrats, making them afraid.

It happened before, and it still happens, at least in the Linux world.

47

One last remark: I was reading David Ingrams experience on medinfo, and perhaps it is a good sign that people are fudding openehr like Microsoft started fudding Linux when it grew too big. Maybe it is just that openehr is having success.

In that case, my congratulations and much thanks for designing this which offers so many good opportunities.

48

That is correct.

Some NEN.CEN standards are free to obtain in the Netherlands because of a contract between the government and the SDO.
Recently the ISO policy is to publish all informative parts of the standard but not the normative parts.

Experts nominated by countries have a larger access to full stadard in the context of standards creation/maintenance.

It is my opinion that the SDO’s need an other business model such that standards are made available for free.

Gerard Freriks
+31 620347088
gfrer@luna.nl

49

Hi Erik,

For some reason, that I have not yet fully understood, previous and current leadership of openEHR has not yet dared >taking the step to skip all ND- and SA- clauses. (Like an anxious over-protective parent afraid to give their now fairly >grown teenager enough trust and freedom.)

The MB has been looking at this issue and I think generally minded to take the steps to remove the ND- and SA- clauses but we need to be absolutely clear about the implications.

My understanding is that removing ND (or Public Domain) could only really be safe if we have solid Trademark protection to prevent a fork representing itself as ‘official openEHR’. This was the approach taken by FHIR, I believe that for some technical reason, previous attempts to secure US trademarking was unsuccessful, and course, will cost a few thousand euros to achieve.

Silje, Heather and myself looked at removing -SA in connection with better understanding the copyright requirements for forks / moves of CKM archetypes. There were some concerns that removing -SA might actually make free movement of archetypes between national repos more difficult, particularly if national govts start to fork and apply more restrictive licences. This is not necessarily a blocker but we do need to think through the implications.

I will raise this at the MB meeting this week wth the suggestion that we set up a small working group with reps from Software, Clinical and Specs group Program leads to look at the options and report back.

Ian

50

The ND on the specs, there must be a kind of protection. Brand protection could work, but must be registered in all countries of the world.

You see the same problem at RFC's, they solved it like this, you cannot change them and publish them under the same name.
In the case of RFC a changed version gets a new number.

I don't know what it takes to make an RFC of something and if it would be appropriate for OpenEHR.

http://www.rfc-editor.org/

Bert

51

http://www.rfc-editor.org/rfcfaq.html#howlong

52

ND = No Derivatives ad is the Creative Commons of what W3C has in their licence <http://www.w3.org/Consortium/Legal/2015/doc-license&gt;\. It's just designed to prevent anyone republishing altered versions of the specifications /as the original specifications /- in other words forked publishing, which would create real problems for obvious reasons.

Probably we do want to allow the forking of the specifications into some new specifications, i.e. with new names and identifiers, that clearly cannot confused with the originals, and the ND provision <http://creativecommons.org/licenses/by-nd/4.0/&gt;I believe would prevent this.

I am not sure what the best replacement is though - it's quite important that a specification with the title 'openEHR EHR Information Model' and version xyz really is only one document, and that no modified versions of that can masquerade as that thing.

W3C achieve this with a custom copyright notice (see above). We probably want a different approach. I don't personally have time to research this but ideally we want a licence that does the following for the specifications:

  * requires attribution with all replublishing, sharing
  * prevents republishing in altered form with same document title, id,
    and also publisher i.e. 'openEHR'
  * but allows normal forking into artefacts that are clearly different

- thomas

53

ND = No Derivatives and is the Creative Commons equivalent of what W3C has in their licence <http://www.w3.org/Consortium/Legal/2015/doc-license&gt;\. It's just designed to prevent anyone republishing altered versions of the specifications /as the original specifications /- in other words forked publishing, which would create real problems for obvious reasons.

Probably we do want to allow the forking of the specifications into some new specifications, i.e. with new names and identifiers, that clearly cannot confused with the originals, and the ND provision <http://creativecommons.org/licenses/by-nd/4.0/&gt;I believe would prevent this.

I am not sure what the best replacement is though - it's quite important that a specification with the title 'openEHR EHR Information Model' and version xyz really is only one document, and that no modified versions of that can masquerade as that thing.

W3C achieve this with a custom copyright notice (see above). We probably want a different approach. I don't personally have time to research this but ideally we want a licence that does the following for the specifications:

  * requires attribution with all replublishing, sharing
  * prevents republishing in altered form with same document title, id,
    and also publisher i.e. 'openEHR'
  * but allows normal forking into artefacts that are clearly different

- thomas

54

I understand the purpose of the ND, and I think, as you, that it is important. and all standards that are not registered by an SDO have this problem. So it is a commonly occurring problem, that is why Creative Commons has an answer for that. I think that works allright. I see no problem in the current situation.

But I understand you, and some others feel OpenEHR being attacked, that is why you want another solution.
You suggested a few, I suggested one. You can also consult a specialized lawyer.
Maybe you are solving a non existing problem.

"If it ain't broke, don't fix it"

Bert

55

Dear Ian,

As I wrote you privately I promised to think over my use of words.

Referring to my e-mail with the definition, as I used it, plus the quote from the openEHR website,
it must have been clear that I was pointing at ownership of the openEHR organisation.

I’m aware now, that ‘proprietary’ has an other, different, meaning, when applied to software or specifications.
My original e-mail conveyed an unintended meaning, is my conclusion.
Therefore I will no longer use the word ‘proprietary’ but the phrase ‘ openEHR as a company owned by UCL’.

With regards,

Gerard

Gerard Freriks
+31 620347088
gfrer@luna.nl

56

Thanks Gerard,

That is very positive and helpful. Would you consider adjusting to ‘ openEHR is a not-for-profit company established by UCL’ which I hope captures your reservations about single ownership without giving the impression that this is a 'for-profit 'company?

Ian

57

Can someone help to take me off the list. I have signed into the list and unsubscribe and still unsuccessful.

58

As I recall the main subject of the discussion was the openness of the specifications. Gerard, as I understand you, you are saying that you will use another phrase when addressing the OpenEHR foundation in relation to UCL. You will no longer use the word proprietary in this context. Personally, I don’t care much if you call the OpenEHR foundation “proprietary of UCL”, or “a company owned by UCL”. But how do you value the specifications of OpenEHR, does he still think the word “proprietary” is appropriate for the specifications? Thanks for any clarification Bert

59

correction,

must be:
do you still think the word "proprietary" is appropriate for the specifications

60

I will consider this.

Gerard

Gerard Freriks
+31 620347088
gfrer@luna.nl