That they’re hidden in a lot of text form which one has to follow through hyperlinks and read even more text.
You stated somewhere - correctly - that companies want to avoid risk, similarly decision makers want to avoid reading through lengthy discussion (from which they have to draw there own conclusions:-) )
If I understand you correctly your main argument is that:
the share alike (SA) requirement will create a risk for lengthy juridical procedures - in every country they operate - for companies who include openEHR archetypes or derivatives thereof in their systems. Since companies avoid risk, they will choose other solutions without an SA requirement.
The reason for this is that it’s not clear what SA exactly means. For instance in the context of building archetype-based GUI- stubs, forms etc. in proprietary systems. As a consequence it could be possible that companies are forced - unwillingly - to open up the source of their proprietary systems. It will take years and many court cases, in different countries, to sort this out. Until then (the large) companies will stay away from openEHR.
This problem can be avoided completely by dropping the SA requirement.
So I guess the first question is: who has a solid argument against Erik’s argument.
The second question is: what are the exact benefits of the SA requirement and are they worth the risk of companies not using openEHR at all (presuming that’s a real risk).
Thank you Shinji, this is an excellent idea - to really put support for language and other localization at the heart. I would propose that one Organisation become an associate and manage local activites - which Organisation should be by a vote of local associates.
This prevents the need for a lot of administrative activity.
The previous Board did not want to make an error and use too loose a licence given that there is no going back.
Our concern is that someone could specialize an archetype and claim copyright, or create a template and do the same. It is our intention at this stage to have a specific clause in the licence that limits it to derived archetypes and templates. At all discussions with industrial parties this has been acceptable, many see it as positive as the corollary of Eric’s approach (which may be the best) is that there are heaps of archetypes out there that have openEHR attribution but are copyright to other parties.
Is it clear what I am saying. It is a conundrum - and needs careful appraisal before going to BY alone.
Hi Tim,
It is tangled up with the CC-BY-SA question. Some one needs to have the copyright or there is a license agreement that is evoked as you enter the the archetype in the repository.
Our advice was that having copyright simplifies the world. Having said that the same archetypes now exist in other repositories with copyright assigned to the national provider, so it is already murky.
The real point is that interoperability depends on sharing archetypes, which need to be available for use without regard to others.
By that, I also mean I can freely use ANY archetype or template out there if I need to.
There are two possibilities for archetypes that are donated to a place like CKM in openEHR.org.
copyright remains with the originator
copyright goes to the publisher, which could be openEHR.org, or some national e-health programme, or some other body
When there is a license attached (e.g. CC-BY), it doesn’t matter that much, since the rights of the user are defined mostly by the license rather than direct reversion to copyright law in some country. However, as with most open source code, it probably makes sense if the copyright is given to openEHR or other publishing entity (e.g. some national e-health programme), since this avoids fights about which of the 20 contributors to an archetype should have copyright. In the end, copyright is a nearly useless concept for changing artefacts, it only really works with unchanging ones like books and paintings. As long as attirbutions to original author(s) and contributors are made in the artefact text, I would say that making the artefact copyright to the relevant publishing organisation makes more sense. Many archetypes won’t really have any ‘original author’ - just contributors, due to being truly team-built artefacts.
I don't think actually changing the text of any accepted, well known
license is a good idea at all - then it becomes something for which no
legal analysis is available, and won't be trusted by anyone. Instead,
openEHR should simply state which kinds of artefacts require which kinds
of license (if any).
It is normal practice with CC to include clarifications and the whole
structure of the license is designed to do this.
Let's stay with the issue of how we stop someone copyrighting and charging
for a specialised archetype? Or a template that is fundamental to health
care (like an antenatal visit)?
Let's stay with the issue of how we stop someone copyrighting and charging
for a specialised archetype? Or a template that is fundamental to health
care (like an antenatal visit)?
So, Sam have you finally dropped the thought that CC-BY-SA would
protect against patent system abuse? We have not gotten a clear answer
to this yet.
Is your only remaining concern now that you believe copyright law
might be so strong that it would stop possibilities to do
specialisations with the same main functionality as possibly
copyrighted useful specialisations? Is that your only remaining "SA"
motive?
Sam, since you are and have been so extremely powerful in the formal
openEHR decision process, we really need to understand your thoughts,
thus I and others have been probing to figure out your reasoning for
years.
Maybe that isn't such a bad thing. They are only roping themselves
into their own corner.
We as a community could always provide help in highlighting those "ropes" via:
- Technical means: license detection upon import of archetypes and EHR
data (as described on the wiki)
- Social & political means: questioning the fairness and wisdom of
parties trying to block the common good of semantic interoperability.
Their income often somehow originates from public funding and they
should be concerned about potential badwill.
We can probably also get around the hypothetical/potential problem by
making a similar (hopefully even better) specialisation ourselves (not
an exact copy) that covers the same needs. It will be hard for the
"copyrighting and charging"-bad guys to claim that another
implementation of the idea is a verbatim copy (prohibited by
copyright) and that their work has enough innovation height that is
not obvious to skilled persons (and thus patentable). The same
argument goes the other way too - even ideas from a CC-BY-SA licensed
archetype from openEHR may be fairly closely re-implemented as
completely closed/copyrighted and it would be both hard and
time-consuming for the openEHR foundation to try and stop this
(anti-interoperable) approach, so let's reduce the temptation by
simply using CC-BY.
Copyright does not stop ideas the same broad way patents do. To
software people I believe it's obvious that improvement-ideas in
commercial closed source forks of e.g. apache-licensed software
projects does not prevent the open source original project to
reimplement similar ideas (as long as the closed source code is not
copied). Copyright is not nearly as harmful as patents in these cases.
Our advice was that having copyright simplifies the world. Having said that
the same archetypes now exist in other repositories with copyright assigned
to the national provider, so it is already murky.
Well, if the national providers had been encouraged to use CC-BY
instead of CC-BY-SA then it wouldn't matter at all who had the
copyright (as Tom has pointed out several times over the years)...
Let’s stay with the issue of how we stop someone copyrighting and charging
for a specialised archetype? Or a template that is fundamental to health
care (like an antenatal visit)?
Cheers, Sam
Why we need to define a license to stop someone to copyright or charge for a specialized archetype?
I think this could be done by defining “user terms” to the archetypes downloaded from the CKM (and every CKM around the world must have the same “use terms”.
You can include something like this on the user terms: all artefacts (archetypes, templates, term sets, etc) downloaded from here are public and free to use and to specialize. All artefacts derived from those, alse must be free. This is a copyleft scheme.
If I want to use artefacts from a “public CKM”, I must follow those rules. Of course, anyone can create its own CKM and create artefacts from scratch and do whatever they want with those artefacts.
I think you can charge for the time you invest in specialize artefacts from public CKMs, but not sell the artifact itself. If you create the artefact from scratch its the creators desition to charge or not.
