[archetypes.com.au] Functionality to compare two archetypes

Technical (archive)
1

Dear all,

We have added prototype functionality to the Archetypefinder to compare
two versions/revisions of an archetype at
http://www.archetypes.com.au/archetypefinder/compare.html. This
functionality will later be integrated into the appropriate tools, but
can be useful on its own too.

You can upload two archetypes and then see what changes have been made
between the two versions of an archetype. Also, the type of the changes
with regard to compatibility of the two archetypes (e.g. 'compatible
with revision', 'requires new version' or 'descriptive change') is
shown.

Have a look - USAGE:

You can use URLs to archetypes as well as upload archetypes directly
from your local system .

For example, you can use a modified blood pressure archetype and compare
it with the current blood pressure archetype from the openEHR site. For
this just

1. Go to http://www.archetypes.com.au/archetypefinder/compare.html,

2. Copy the URL
http://svn.openehr.org/knowledge/archetypes/dev/adl/openehr/ehr/entry/ob
servation/openEHR-EHR-OBSERVATION.blood_pressure.v1.adl and paste it
into the webform as URL for the original archetype (1st row, last
column)

3. Copy the URL
http://healthinformatics.cqu.edu.au/archetypefinder/openEHR-EHR-OBSERVAT
ION.blood_pressure.v1.adl and paste it into the webform as URL for the
modified archetype (2nd row, last column)

4. Click the 'Compare Archetypes'-button

Please note that this is a prototype. We would be glad about any
feedback, bug reports, ideas, discussion.

Especially, if you disagree with some of the conclusions about the
compatibility of two archetypes you submitted, please drop me an email
or discuss via this list!

It is important to make the compatibility of archetypes explicit, so
that archetype-enabled software can decide this - we want to settle this
issue once and for all!

Many thanks to Stefan Fuchs from the UMIT in Austria for heaps of hard
work on this during an internship with us!

Best regards,

Sebastian

Dr Sebastian Garde

Dr. sc. hum., Dipl.-Inform. Med, FACHI

Research Fellow

Faculty of Business and Informatics, Central Queensland University

Austin Centre for Applied Clinical Informatics, Austin Health

Heidelberg Vic 3084, Australia

s.garde@cqu.edu.au

Ph: +61 (0)3 9496 4040

Fax: +61 (0)3 9496 4224

http://healthinformatics.cqu.edu.au
<http://healthinformatics.cqu.edu.au/&gt;

http://www.acaci.org.au/&gt;

Visit the new open access electronic Journal of Health Informatics
(eJHI): http://ejhi.net/&gt;

2

Hi Sebastian!

It is a very nice piece of work you and Stefan have done! I fully
agree that we need to make archetypes compatibility checking explicit
and this is a very good start.

It will be useful to publish the algorithm that you used for
comparison so others could comment. And the same algorithm could be
implemented in different tools to ensure interoperability of
archetypes.

Cheers,
Rong

3

Rong Chen wrote:

Hi Sebastian!

It is a very nice piece of work you and Stefan have done! I fully
agree that we need to make archetypes compatibility checking explicit
and this is a very good start.

It will be useful to publish the algorithm that you used for
comparison so others could comment. And the same algorithm could be
implemented in different tools to ensure interoperability of
archetypes.
  

I have just been thinking about the specification for this area myself,
and how we should approach it. In our software, we have just
successfully done the 'flat-form' -> 'differential form' transformation
of specialised archetypes, and we are now working on the differential ->
flat-form transform (i.e. compression through inheritance). These two
transforms allow hierarchies of specialisations to be properly
processed. The excellent work Sebastian mentions above gives us some
generic differencing, even between non-parent/child pairs. There are
probably some more variants we need as well.

My proposal would be that once we have perfected the
transformations/comparisons in software, that we document the algorithms
in the Archetype Semantics document (currently an early draft lives here

4

Hi Sebastian,

Nice work!

I, who quite often work with medical information in more than one language, think it would be nice to add a column with the language code for each line when it is suitable and maybe the possibility of selecting to only show the differences for one of the languages in the archetype. Otherwise it can be quite annoying to compare archetypes which contain many languages.

Greetings,

Mikael

5

Hi everyone

We need to be careful here as this will show the changes and if there have (inadvertently) been changes in other languages then we do want the author to know. There should not be changes in more than one language for most of the archetype…

I am not sure a dynamic filter will be easy at this stage. Ajax?

Cheers, Sam

Mikael Nyström wrote:

(attachments)

OceanCsmall.png

6

Hi Sam,

Of course it improves the risk for the user to unconsciously hide the changes the user is looking for, but in general I believe that it is a good thing to implement useful functionality even if it improves the risk for a beginner to do to something wrong.

(I believe that we non native English speakers think about managing information in multiple languages as a more complex task than native English speakers.)

Greetings,

Mikael

(attachments)

OceanCsmall.png

7

Hi Mikael, hi Sam,

Thank you for your comments.

There is room for improvement in displaying this - I think often Javascript (without server interaction like with AJAX) is used for this.

I created a feature request for this.

Also I'd like a better display of the hierarchy within the archetype, e.g. if a cluster is added with heaps of elements in it, they should be displayed 'below' the cluster.

Cheers

Sebastian

(attachments)

image002.gif