ref_impl_ruby

Implementers (archive)
1

Dear all

We launched ref_impl_ruby project in August 2007 in Japan. Now, we
succeeded in implementation of ADL parser. The bellow URL shows detail
of the project.
http://www.openehr.jp/wiki/Ruby%20reference%20implementation
Ruby is a smart and powerful object oriented script language, which
takes less effort to programming. I believe that Ruby implementation
will enhance the availability of openEHR well.
We have many module to implement to openEHR environment, but members are
not enough in Japan. Please join us if you are interested in Ruby.
And please set up ref_impl_ruby mailing list and repository in openEHR
official site.

2

KOBAYASHI, Shinji schreef:

Dear all

We launched ref_impl_ruby project in August 2007 in Japan. Now, we
succeeded in implementation of ADL parser. The bellow URL shows detail
of the project.
http://www.openehr.jp/wiki/Ruby%20reference%20implementation
Ruby is a smart and powerful object oriented script language, which
takes less effort to programming. I believe that Ruby implementation
will enhance the availability of openEHR well.
We have many module to implement to openEHR environment, but members are
not enough in Japan. Please join us if you are interested in Ruby.
And please set up ref_impl_ruby mailing list and repository in openEHR
official site.

Very interesting, I find this very encouraging.

I don;t know much about Ruby, in fact, almost nothing. Does it implement
mutiple inheritance?
And how does the language behave on scalability, when it class-instance
are "run" as a service, is it well possible to run multiple threads of
openehr instances simultaneous without using to much resources?

I would surely be glad if a ruby-implementation would exist on the
openehr website, it is not me to decide, but that is my opinion

I hope the road to knowledge on the ruby language and environment is not
to hard to do.

good luck
and I hope we will hear more about you and this project
Bert Verhees

3

I don;t know much about Ruby, in fact, almost nothing. Does it implement
mutiple inheritance?

There are a number of discussions around multiple inheritance.
Ruby dose not implement multiple inheritance but implement 'mix-in' by
module, because multiple inheritance is recognised as a source of
complexity by Ruby developers. In Ruby, 'module' act as abstract class
and interfece in Java. Module can not be instantiated but inherit
methods to child class. Mix-in act almost like multiple inheritance with
less complexity.
http://www.juixe.com/techknow/index.php/2006/06/15/mixins-in-ruby/

And how does the language behave on scalability, when it class-instance
are "run" as a service, is it well possible to run multiple threads of
openehr instances simultaneous without using to much resources?

I don't know the answer if Ruby has scalability to run openEHR, because
I have not implemented all the task yet. However, some popular web sites
running on Ruby perform well. Do you know twitter? It is running on Ruby
on Rails. I think Ruby scales for openEHR enough.
http://www.slideshare.net/Blaine/scaling-twitter/
Ruby on rails is a powerful framework for web application. We want to make
openEHR plugin for rails and use it for my systems.
http://www.rubyonrails.org/

4

All,

I believe that there has been huge growth in the awareness around the
world of openEHR and especially archetypes over the past two years.

The keyword in that sentence is 'awareness'. In many places there is a
pretty serious lack of understanding. Tom Beale, et.al. has done a
wonderful job over the years of creating AND maintaining outstanding
documentation for the Foundation. This is a first for many open
source/open content projects.

How can we the community, support the Foundation better so the
Foundation can support us better?

[Caution random thoughts ahead]

At an appropriate time, conferences are very beneficial in community and
awareness building. [5-6 years out for an openEHR conference?]

Meetings at other related conferences are always good. I understand
that there were a significant number at MedInfo this year [I couldn't be
there :-< ]. This is something that should be in the planning. Are
there papers to be presented? Where will they be submitted? BoF
meetings.

How can we come together under the auspices of the Foundation to build
internal support and support each other? [actually I understand this is
already being worked on]

I'd love to have discussion on these and other points. Remember; "a
rising tide floats all boats" [I have no idea who said that but it
sounds cool so I repeat it often]. :slight_smile:

A story about what (partially) prompted this email:

I was invited to give a presentation to a Standards working group on
Archetypes. Now; I am not an "Archetypes" expert by any means. Heck,
I'm not even a clinician. But I do understand the Reference Model and I
understand two-level modeling. Besides, I've never been to Kenya
before. :slight_smile: So I accepted. I was also asked to present a comparison
between OpenMRS "Concepts" and archetypes. Well, I ass-u-me-d that the
people that had invited me saw some similarities between them and wanted
some clarification and a suggestion as to how the two might be used
together.

After several weeks of just thinking about a presentation and asking a
question here and there. I realized that the conference in next and
maybe I should really understand OpenMRS "Concepts". [usually a good
idea at this point]. BTW: The panel we are in together was labeled
"Vocabulary Management". I had sent an email a few weeks ago suggesting
it should be called "Knowledge Management" if we were presenting
"Archetypes" and "Concepts". No answer.

On Wednesday (yes only 5 days before traveling to the conference) I
started engaging the OpenMRS community with serious questions (I had
been lurking for weeks). Well, one question led to two and two to four
(you get the idea) and pretty soon I think I shut down OpenMRS
development progress because everyone was busy answering my questions
(sorry but thanks, OpenMRS guys).

After my re-education and a basic "aaaaaHHHHHHaaaaaa" I realized that
"Concepts" and "Archetypes" are not even close in similarity! Basically,
"Concepts" are a local vocabulary used to improve computability to
patient responses in a typical EAV data model.

My point is [may be you haven't done this] that it is too easy sometimes
to get so close to something that you do not understand or appreciate
the view point of others. I began to make assumptions. I almost lost
an opportunity to present the wonderful virtues of openEHR and two-level
modeling because of it.

I am looking forward to next week and I really do appreciate my hosts
asking me to attend.

In preparing the media for the conference they asked for presentations
as well as any background documentation. I included (with my
presentation) the openEHR overview, the archetype principles document
(after checking the license) and a couple of archetypes in ADL and HTML.
I believe (hope) that people that would not normally go tot he website
to look for these will at least browse them on the conference media.

I also think that it is very important for us to point out that while
the ideal is a longitudinal, future proof health record. The openEHR
specifications (applications) can deliver independent clinical
applications that can share information. Lets say that a health
authority wants to review TB registries, Malaria registries and HIV
registries. They can do that with the same application using the
specific archetypes from each area. No other model can accomplish that.

So, my point is ...

we as a community need to promote and EDUCATE at every opportunity, the
concepts of openEHR, archetypes and the Foundation so that we can
encouragin funders to recognize the Foundation needs and so that the
Foundation can grow to support us.

DISCLAIMER: I am not a member of the openEHR Foundation Board of
Directors nor do I have any financial stake in the Foundation. I do
however believe that this is the best opportunity that the world has (so
far) to achieve longitudinal, interoperable, computable health
information systems.

Your comments are encouraged,

Tim

5

Hi Tim

I couldn’t agree more! Hope your conference goes very well - let us know if there is anyway that we can support you.

regards Hugh

Tim Cook wrote:

6

Hi Tim,

I fully agree with you!

The openEHR community need to go out and present the idea of Archetype and two-level modelling to wider audience. So far the openEHR presence in scientific conferences like MIE and Medinfo has been quite good. There were 14 papers, workshops and posters about openEHR at Medinfo this year! All of them can be found at: http://www.openehr.org/shared-resources/conferences/medinfo2007.html. There were also coverage of openEHR at MIE2006 as I recall. Maybe we should try to collect the papers and put them on openEHR site.

Besides scientific conference, we should also participate more general conferences for clinical user groups and software vendors. Next year’s MIE2007 in Gothenberg is a good opportunity since it’s collocated with a major Swedish health IT conference Vitalis, which will draw audience from hospitals, software industry and government. The awareness of openEHR and Archetype in Sweden is quite high. Projects on national and county level are investigating Archetype based EHR standards. As it looks now, there probably will be good coverage of openEHR both at MIE2007 and Vitalis. It would be nice to organize some activities for the openEHR community then like what we did at Medinfo. =)

More specifically about the design comparison between OpenMRS and openEHR, Gunnar and I did such a comparison study between a template-based complementary EHR system and openEHR. The result is published here: http://www.biomedcentral.com/1472-6947/7/10. Hope this is useful to you.

Cheers,
Rong

7

Hi Tim

How can we the community, support the Foundation better so the

Foundation can support us better?

Exellent question. Here are some suggestions:

  • is it possible to have on the website a list of interesting conferences and a list of attendees from the community, so we can know who is going to which meeting.
  • I’ve noticed that many presentations are posted on our website. Is it an idea to create/provide some 'basic/ general tutorial presentations free to use for everybody. I know that people like Thomas, Sam and Dipak (and others) are working extremely hard to create awareness, but unfortunately they can’t be everywhere. Such presentations could be of help for the ‘less experienced’ among us to help to create awareness.
  • I don’t know if this is true and maybe it’s just my imagination, but I have the feeling is that less energy is spend in creating awareness amongst the ‘end-users’, i.c. the health-care providers. If so how could we reach those even more, in the end they will be our best ambassadors?
  • in that respect. You stated ‘[5-6 years out for an openEHR conference?]’. Maybe so, but I really think that in the meantime there should at least be some ‘workshops/bootcamps/whatever you want to call them’ for the clinicians to learn, transfer and improve the art of archetyping. This could also be the, in my opinion much needed platform, to discuss archetype creation strategies. It would be a shame if different strategies would arise to tackle similar problems which then later-on needs to be harmonized.

Really looking forward to a report of the Standards working group meeting. I really like the way you write things down. Maybe we can appoint you as the communities PR manager:-)

Cheers,

Stef

8

Tim

I was very interested to read of your engagement with openMRS as
there has been contact with me about this via Don Detmer, CEO of
AMIA, who is a visiting professor in CHIME. What you are doing seems
an excellent way of broadening out the openEHR mission.

After your conference, it would be extremely helpful to hear, in some
detail, your assessment of the commonalities and differences, and the
potential and useful opportunities to develop synergy with them, from
the Foundation.

On the broader issues you raise, the Board has been working on a
number of options for supporting the consolidation and development of
the openEHR community, through a business case built around the
Foundation's mission, which emphasises clinical and health
engagements, rigour and openness of technical approaches, and
building of trust within user communities. We believe that revenue to
the Foundation must come through its direct connection with and
relevance to these three, and from activities and organisations that
identify with or build on its mission. As of now, having a financial
interest in the openEHR Foundation essentially means paying bills,
which have been considerable, in cost and in kind, over a long period of time.

For my part, I am very fortunate in having had the explicit support
of successive Presidents of UCL in holding this position, but we all
recognise that we have to find the right way, now, to change gear and
put future finances onto a proper and sustainable footing. Sam and I
have started some confidential high level discussions, here, about
this, and we have been very well received. I have worked a lot in the
voluntary sector and know the often quite unproductive load that can
be placed on Foundations that depend, for their work, on sponsorship
and voluntary support, alone. We should thus, we believe, explore
appropriate revenue streams, tied closer into the contractual domain,
around use and support of the openEHR IPR and trademark. Any useful
trademark must be asserted, supported and, if necessary, defended,
and that costs money; even acquiring the trademark has cost us a
great deal. But, that said, clearly Foundation revenue streams must
be framed in such a way that they are explicitly connected with
relevance and usefulness in pursuit of the mission of the Foundation,
as an open and not-for-profit enterprise. This is not an easy
balancing act and there are no exact parallels that I know of, though
many quite close analogies to learn from.

We must and are determined to take our time to get this next stage
right. We're in this for the long term and welcome and value all the
huge amount of thoughtfulness and commitment shown within the
community, that is helping us all move forward.

By the way this should perhaps be a post on a different list.

David

'A rising tide floats all boats' is a new one on me; I will check its
provenance and certainly remember it. In the worlds I move in we
often seem to live in a Mr Micawber (Charles Dickens) zone between
'income = one pound, expenditure equals one pound and sixpence
(misery and disaster abound)' and 'income = one pound, expenditure
equals nineteen shillings and sixpence (joy and munificence all
round)'! Change management in the latter scenario seems an Human
Resources Department free breeze; in the former scenario a completely
draining HR
dominated nightmare! So I'm all for rising tides. In England, we have
some tides that rise and fall 10 metres, every day, though!

9

Hi David,

Thanks for the reply.

I 'think' it was the economist Milton Friedman that said the quote about
tides and boats???

I will certainly report to the Foundation & community on the results of
this conference. OpenMRS has a very high profile now and the team has
been very giving of their time. It is great to have the opportunity to
work with them.

We have been VERY fortunate to have the support of UCL over the years.
But, I am very happy to hear that the Board is looking at long term
solutions in maintaining the Foundation. Part of my point is that it is
incumbent upon us all to participate at some level in that
sustainability.

You are correct that this isn't the correct mailing list for this
discussion. However, there is no mailing list for discussing the
Foundation or for publicity or even for education. I thought about
using the Wiki. But really, unless you make a compelling case in an
email people aren't going to click through to the Wiki page anyway. So,
I plead guilty to list abuse! :wink:

Stef Verlinden had some great comments (on the Implementers list). One
in particular is about providing presentations that others can deliver
when the find themselves in a venue/meeting where openEHR would be an
appropriate topic. In the past, Tom Beale has allowed me to use some of
his graphics etc. that were likely created under the auspices of Ocean
Informatics. I think that we should have at least some basic
presentations (with notes) that everyone can use that have explicit open
license's. I will donate to that effort.

Shall I copy/reformat this for the wiki?

Cheers,
Tim

10

Isn't this?
http://en.wikipedia.org/wiki/A_rising_tide_lifts_all_boats

I also agree that many conferences make better the community, but it is
difficult for worldwide project, such as openEHR. Therefore, we have
local activity in Japan to have many conference and parties.

I think about open source community norm is in terms of another JFK
speech,
"And so, my fellow developers: ask not what your community can do for
you - ask what you can do for your commnunity. My fellow citizens of the
world: ask not what Foundation will do for you, but what together we can
do for the freedom of man."

We have a freedom to develop openEHR systems within the licenses. The
freedom in research and development is great worthy in sustainability.
Both OpenMRS and openEHR are deffer in modeling but have the same
concept, 'Freedom'

I am looking forward your educational tutorials and send cheers for your
act as an 'evangelist' of open source in medicine. I will also write
codes for me, medical providers and open source community in the world.

11

Actually, I think that the really exciting thing is that we now have a growing community of openEHR people from around the world who want to contribute and make it grow. Shinji, we really appreciate you and your groups growing participation in the community.

I look forward to coming to Japan one day soon and sharing some of that really good Sushi that you promised me!

regards Hugh

KOBAYASHI, Shinji wrote:

12

Hi Rong,

I had not previously seen this paper.

It should be "required reading". :wink:

Thanks for the link.

Cheers,
Tim

13

Tim Cook wrote:

Stef Verlinden had some great comments (on the Implementers list). One
in particular is about providing presentations that others can deliver
when the find themselves in a venue/meeting where openEHR would be an
appropriate topic. In the past, Tom Beale has allowed me to use some of
his graphics etc. that were likely created under the auspices of Ocean
Informatics. I think that we should have at least some basic
presentations (with notes) that everyone can use that have explicit open
license's. I will donate to that effort.

*well, most of the PPTs I created were paid for by Ocean, UCL or me
(earlier days). If they are on the website, they are fair game for
re-use but the original authors should always be acknowledge of course.
Do we need to go as far as having a license to cover presentations?
Maybe we do.... sugestions are welcome.

- thomas

14

If they are on the website, they are fair game for
re-use

actually, legally, no

if they are posted with no details concerning copyright, then
default copyright applies, which is called "fair use". But like
everything legal, "fair use" has a legal definition which differs
from sensible interpretation. In some countries, this difference
has been underlined by applicable case law

So it's important to be explicit about allowed usage. I use
either EPL (http://www.eclipse.org/org/documents/epl-v10.php) or
Creative Commons (http://creativecommons.org/licenses/by/2.5/au/)

> Maybe we do.... sugestions are welcome.

yes, you should, and I suggest either the applicable openEHR
licenses or creative commons.

Grahame

15

HI!

I was also just about to suggest Creative Commons (CC) when I saw the
reply from Grahame.
There are several CC versions for different purposes, see:
http://creativecommons.org/about/
http://creativecommons.org/license/
http://creativecommons.org/about/license/

I'd suggest using the most liberal ones (e.g. "by" or "public domain")
for presentational material intended for reuse:
http://creativecommons.org/licenses/by/3.0/
The "by" licence mixes well with the common scientific publication process.

Using software licenses like GPL or EPL for documents can be
unnecessarily confusing.

The current licence for the openEHR specifications...
http://www.openehr.org/releases/1.0.1/html/architecture/overview/Output/front.html
... seems to be close to...
http://creativecommons.org/licenses/by-nc-nd/3.0/
...so for next specification release one could consider switching to
that if a more well known license is wanted.

By the way, the web-remake seems to have killed the link referred to
in the openEHR specification licence undetected:
http://www.openehr.org/free_commercial_use.htm

Best regards,
Erik Sundvall
erisu@imt.liu.se http://www.imt.liu.se/~erisu/ Tel: +46-13-227579

16

if they are posted with no details concerning copyright, then
default copyright applies, which is called "fair use". But like

In the Netherlands, default copyright means "no right to copy" all rigths
are with the writer, not even his employer has any right, which was a
problem on an application on which the at that time employed programmer
claimed copyright, sometime after he was fired.

Bert