Gang,

SInce I was a bit of coward and wished to have some private feedbackback 
first, before exposing my thinking to the world.

Here we go,

I am very much "use-case driven" in much approach to programming, always been. 
I also always assumed everything can be done easily, if you just take really 
small steps.

So, I want to get the Eclipse plug-in development going, and I was first 
looking at an AssemblerEditor, but found that a bit over my head as an 
initial attempt. I turned to a RepositoryView.

ATM, my definition of a repository in this context is;

A location where a RepositoryAgent (representative) can retrieve 
metainformation about components stored in such location. The RepositoryAgent 
also knows how to retrieve the component itself.

A component in this context is, any digital media that can be abstracted into 
a single entity, with an Identification (globally unique), Name (smaller 
human-readable form), Description, Compliance, Version.

Compliance is basically; Do you support X? yes/no, where X is any known 
standard.

So much (to me) is fairly clear. I am still struggling a bit with the concept 
of Dependency and Standard (exposure of service, interfaces or other 
standards in the market place.), so at the moment I just assume the Avalon 
model, Service Exposed and Dependence On Service, but expect to change that 
in the near future, when I/we get a better understanding of the abstraction.


The Eclipse Plugin is to support that you can install RepositoryAgent 
implementations, so basically any repository type can be browsed.
For components that have Merlin Compliance, the UI will allow to be 
drag-n-dropped to a created merlin container, and manual or automatic 
dependency resolution can take place, but that is a bit further down the 
line.
Later on, we will see where the ship goes. Small steps!!!


Niclas


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to