[Data-modeling] Processes

[Jeff Prucher]

> 
> Firstly, there is a Process 
> <http://process.freebase.com/type/schema/base/process/process>
>   type, e.g. Nitrogen fixation 
> <http://process.freebase.com/view/en/nitrogen_fixation>  , 
> which has some Input and Output properties (these would be 
> typically expect a type of material, but to allow for 
> flexibility I'm only going to expect a type of Topic) A 
> process can be composed of other component processes (for the 
> Nitrogen Cycle 
> <http://process.freebase.com/view/en/nitrogen_cycle>  these 
> would be Nitrogen fixation 
> <http://process.freebase.com/view/en/nitrogen_fixation>  , 
> decomposition 
> <http://process.freebase.com/view/en/decomposition>  , 
> Nitrification 
> <http://process.freebase.com/view/en/nitrification>  etc.).  
> Parent and component processes are linked with a Process 
> stage CVT 
> <http://process.freebase.com/type/schema/base/process/process_
> stage>  .
> 
> I was trying to model the order in which stages of a process 
> occur (e.g. decomposition, followed by Nitrification, 
> followed by assimilation) by using "previous stage" and 
> "subsequent stage" properties - but that led to a CVT within 
> a CVT which the client doesn't agree with.  I've worked 
> around this to make Process stage have an Input and Output 
> property, but this means that input and outputs both need to 
> be entered twice - once into the Process topic and another 
> time into the Process stage CVT - not ideal but it works.

Why use a CVT between the Part Of and Component properties?  I _think_ that
if you just made the "input" and "output" properties of "Process" into
disambiguators, they would show up as a tabular display without a CVT, and
therefore without requiring redundant data entry.   

-Jeff