oship-dev team mailing list archive

["Sergio Miranda Freire" <sergio@xxxxxxxxxxxxxxx>]

Hi Guys,

I have said it before, but I may not have emphasized it enough: My
"original" miniterm is a translation to Python from Rong Chen's java code. I
did that only to have a provisional implementation that could support the
coding of several classes of the rm. Certainly, parsing xml files is not the
most efficient way to do that. Observe, however, that the parsing would be
done only once at the start of the server and then everything would be in
memory. 
Also, I do not think that this terminology server performance would be a
bottleneck for any application.

Anyway, the main point here is that I haven´t broken any contract provided
by the specification. You can implement the same interface in many different
ways, you know that. Then you can substitute one implementation for another
as you wish without disrupting your code. Had I implemented some classes of
the rm following the specifications, I would have now to recode them,
because Tim implemented a termserver, but not the terminology interface.
Even worse, I would have to dig into his code to know how to call the
server, because I do not have a contract to look at.

IMHO, the terminology API in the specification is a starting point for
building an API for accessing terminology servers. Probably it will have to
be improved as we gain experience in working with several terminologies
(from a simple one like openEHR codes to very complex terminologies, e.g.,
SNOMED).
If Roger implements the ICD server his way, soon we will have a plethora of
terminology servers, and then we will feel the need for a common API.

Finally, it seems to me that the terminology package will not raise any of
those problems that Tim had with the rm. The methods there are simple to
implement, once you define how you are going to implement the terminology
server or how you are going to access a server outside OSHIP domain. 

Cheers,

Sergio

-----Mensagem original-----
De: oship-dev-bounces+sergio=lampada.uerj.br@xxxxxxxxxxxxxxxxxxx
[mailto:oship-dev-bounces+sergio=lampada.uerj.br@xxxxxxxxxxxxxxxxxxx] Em
nome de Tim Cook
Enviada em: Thursday, March 19, 2009 9:52 AM
Para: 'OSHIP-Dev'
Assunto: Re: [Oship-dev] RES: Recent re-factoring to Grok was: RES: openEHR
Terminology source file

Hi Roger,

I think I read your questions "slightly" differently than Sergio so I will
try to add some more information.

On Thu, 2009-03-19 at 08:51 -0300, Sergio Miranda Freire wrote:
> Hi Roger,
> 
> Answering your questions:
> A) it is in oeterm directory when you update the source code from the
trunk.

Sergio's original minitermserver is in the ~vcs-imports/oship/trunk code
tree.  This is what is getting imported from openehr.org svn.
See: http://bazaar.launchpad.net/~vcs-imports/oship/trunk/files/head%
3A/openehr//minitermserv/


> B) I am not in my computer now, but I think there is just one module in
that directory.
There is one module in that directory that parses the XML (excluding the
country and languages sections which are already available in Grok/Zope3 and
updated by others as things change.

> C and D) The package is org.openehr.rm.support.terminology and the
specification is in the support_im.pdf doc in the openEHR site.
> 
Located here: http://www.openehr.org/releases/1.0.2/roadmap.html though I
can't seem to get to the xml file right now but it is in the oeterm
directory in OSHIP.

> Since you are working with ICD10, this issue directly concerns you. I
agree with most of Tim's answer, except what concerns the termserver.
> I will return to this issue later, but before I will have a closer look at
his code and at the specs again. 

Looking at the code and the specs is a good idea.  However, I also STRONGLY
suggest that you include in that study; the capability and overlapping
functionality of Grok.  

Compare parsing and searching XML files to reading from a cataloged and
indexed BTree container.

Cheers,
Tim



--
Timothy Cook, MSc
Health Informatics Research & Development Services LinkedIn
Profile:http://www.linkedin.com/in/timothywaynecook
Skype ID == timothy.cook
**************************************************************
*You may get my Public GPG key from  popular keyservers or   *
*from this link http://timothywayne.cook.googlepages.com/home*
**************************************************************