oship-dev team mailing list archive

[Diego Manhães Pinheiro <dmpinheiro@xxxxxxxxx>]

2010/7/9 Tim Cook <timothywayne.cook@xxxxxxxxx>

> Hi Diego and All,
>
> We are redefining what OSHIP is in the following way.  It is the Open
> Source Healthcare Information Platform coded in multiple programming
> languages.
>
> So, my idea is that we have a development branch for each of the
> following:
>
> OSHIPpy - Python
>
> OSHIPrb - Ruby
>
> OSHIPjava -Java
>
> OSHIPlua - Lua
>
> OSHIPcpp - C++
>
>
> We have code for the first three in various states of completion.  So
> what are the pros and cons of using one OSHIP repository divided into
> separate branches as compared to creating new projects for each?  I lean
> towards using the same project with multiple repositories.  It certainly
> makes management easier and will easily allow for Blueprints etc. to be
> applied to all when it makes sense to do so.
>
I don't think it is a good idea. Take two projects with the same goal,
doesn't mean that they will follow the same steps to achieve the goal.
Reuse blueprints isn't a good idea. I think the focus of each these projects
is implement the openEHR specifications. Each language has your own ways to
implement it. Each language has different strategies to solve a problem (
some complicated, others less complicated). OSHIP Blueprints are related to
the how you will implement a behavior described by the openEHR
specifications. So, If you want implement the same behavior in different
languages,with the same object roles, with the same message trades defined
by the Launchpad blueprints, I reckon this is a waste of time. An better
approach is choose only one language, implement your thoughts, learn with
the mistakes and try to evolve them with a code that make them true.
If you want create code components reusable by different languages, use a
.NET framework or a VM framework seems to be a better solution to this.
SOA[1] also can be an alternative to this.

In brief: Use different projects in Launchpad with different
repositories,with distinct blueprints, bug reports, etc; seems to be more
reliable to achieve openEHR implementations to each language.

[1] - http://pt.wikipedia.org/wiki/Service-oriented_architecture

PS: Please, include IMHO into my all points.

>
> Thoughts?
>
> Cheers,
> Tim
>
>
>
> --
> ***************************************************************
> Timothy Cook, MSc
> Project Lead - Multi-Level Healthcare Information Modeling
> http://www.mlhim.org
>
> LinkedIn Profile:http://www.linkedin.com/in/timothywaynecook
> Skype ID == timothy.cook
> Academic.Edu Profile: http://uff.academia.edu/TimothyCook
>
> You may get my Public GPG key from  popular keyservers or
> from this link http://timothywayne.cook.googlepages.com/home
>
>
> _______________________________________________
> Mailing list: https://launchpad.net/~oship-dev
> Post to     : oship-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~oship-dev
> More help   : https://help.launchpad.net/ListHelp
>
>


-- 
http://dmpinheiro.net
-----------------------------------
Diego Manhães Pinheiro