Loading ...
Sorry, an error occurred while loading the content.

Re: [openhealth] Re: Open Source?

Expand Messages
  • Tim Churches
    ... First of all, I must point out, as I said before, that the openEHR Foundation and Ocean Informatics, the Australian company formed to develop openEHR ideas
    Message 1 of 116 , Nov 30, 2006
    • 0 Attachment
      Fred Trotter wrote:
      > What we, as the community need from Ocean Informatics is a very clear
      > commitment about
      >
      > - What you are releasing open source
      > - When you will be releasing it
      > - What license it will be under
      > - Where it can be downloaded
      > - What you are not releasing open source
      > - What your reasoning is for releasing some things as FOSS and some
      > things not
      > - We need a public commitment.
      >
      > We need this to be on a page on your website written in very clear terms so
      > that when your company is sold, the new owners cannot pursue a "new
      > strategy" for your company without breaking clearly articulated promises to
      > the community. In short answering questions on this thread is not enough.
      >
      > Now, granted, you may already have a public statement like this that is web
      > accessible. If so, great! then all you need to do to repy to Mark is to post
      > a link.

      First of all, I must point out, as I said before, that the openEHR
      Foundation and Ocean Informatics, the Australian company formed to
      develop openEHR ideas and related software, have both exemplary in their
      efforts to make the intellectual property behind openEHR freely
      available - they have released documents which cover most (although not
      all) of what is needed for suitably skilled and resourced organisations
      to create their own openEHR implementations, and they have not, as far
      as I know, pursued any patents on these ideas.

      If you examine http://oceaninformatics.biz you will see that they have
      also released source code under approved open source licenses for two
      bits of software: an Archetype Editor, and a Archetype Technical
      Workbench, both written in Microsoft C# in the .NET environment, I think
      (although earlier versions were written in Eiffel?). So they have put
      their (open sourced) code where their mouth is.

      Also listed on that Web page are several other software components
      needed to build a functional openEHR-based information system. None of
      these are available under open source licenses, or any other kind of
      license, although Thomas suggests that they may be made available as
      open source in the future. I don't think that Ocean Informatics should
      be criticised for not immediately, or possibly not ever, making them
      available as open source - apart from the fact that the components are
      not completed, there are, as Thomas pointed out, business models to be
      considered, and fair enough too.

      However, as others have pointed out, circumspection is probably wise
      regarding the ultimate licensing arrangements for these other openEHR
      software components, Thomas' assurances about their eventual open
      sourcing notwithstanding.

      My own frustration with Thomas' post to this list stems from the fact
      that it the last of quite a few such posts, stretching back many years
      (at least 5 years, possibly 7 or more) announcing that openEHR software
      implementations would be available very soon now. Yet they never seem to
      be quite finished or ready for others to use and test and examine in
      detail. Thomas's offer of arms-length use of an closed source openEHR
      kernel/data storage repository via a closed-source C# DLL over an (as
      yet undocumented?) Web API is not quite the same as being able to verify
      for oneself using an openEHR installation under one's own control, that
      the openEHR ideas do in fact cut the mustard in real-world use, rather
      than as thought experiments and specification documents. Thomas tells us
      that sound, complete openEHR data storage and retrieval implementations
      do exist, but no-one seems to be able to point to where such
      implementations can be had, either as open source or as commercial
      closed source products. Hence my ongoing frustration and, I must admit,
      rising irritation with the whole openEHR initiative. Anyway, the world
      keeps turning, and people who have never even heard of openEHR are
      happily getting on with building "two-level", metadata-driven,
      "soft-coded", flexible and evolvable data storage and retrieval
      mechanisms in health care and other settings.

      Tim C
    • Tim Churches
      ... No, free, open-source licensing is something that only the copyright holder themselves can do. After they have attended to that, *then* others can
      Message 116 of 116 , Jul 1, 2007
      • 0 Attachment
        Adrian Midgley wrote:
        > Tim Churches wrote:
        >> Thomas Beale wrote:
        >>> Tim,
        >>>
        >>> all archetypes published on openEHR are for free use, forever.
        >>> Otherwise they don't go there. Free archetypes is our credo.
        >>
        >> Yeah, but where, in writing, does it say that? Any lawyer will
        >> tell you that credos don't count in court when push comes to
        >> shove. Properly written and executed license agreements do.
        >
        > In the FLOSS model, this may be a matter for other workers to
        > volunteer their labour to get written and executed.

        No, free, open-source licensing is something that only the copyright
        holder themselves can do. After they have attended to that, *then*
        others can contribute to other aspects. But it has to happen in that order.

        Adrian Midgley wrote:
        > Tim Churches wrote:
        >>
        >> ... perhaps it deserves to be written up
        >> as a brief formal paper? There other sociological aspects of openEHR
        >> archetype licensing and control which I have not discussed here, too.
        >
        > It does.
        >
        > The sociological aspects give us more trouble than the technical
        > nowadays, I think, if we include the travails of the NHS and NPfIT and
        > their putative suppliers in "sociological".
        >
        > There is http://www.josmc.org/ JOSMC which could do with material and is
        > a reasonable place for that, and there are always the BMJ and so on if
        > it is short and tight.

        I was a fan of BioMed Central, but recently we have found their
        peer-review model to be broken, with papers sent to reviewers who
        because they are identified are overly critical, and hence perfectly
        good papers seem to get held up for endless rounds of revision and
        re-review.

        PLoS ONE would be a good choice, and I probably find funds for the
        publication fee if someone suitable agrees to be an active co-author.

        Tim C
      Your message has been successfully submitted and would be delivered to recipients shortly.