[WF-Protocols] Design Goals

[Eason Choo]

Matze Braun wrote:

>On Thu, 3 Jul 2003, Aglanor wrote:
>
>  
>
>>Ok, I've re-written the goals taking into accounts the comments since
>>yesterday:
>>
>>Design Goals:
>>
>>	* A format for interchange of model data. The main criteria of decision
>>should be interchangeability, above others such as efficiency.
>>
>>	* Must be free and open. Must be independent from platform and SDK.
>>	
>>	* Must be scalable: new features added to the standard should not break
>>backwards compatibility. Applications should ignore features they don't
>>understand.
>>
>>	* Should be easy to implement, read and write.
>>
>>	* Oriented to rendering engines and model tools. Specifically, it is
>>not intended for complex editors.
>>    
>>
These look good to me too

>>    
>>
>These goals are fine IMO.
>
>  
>
>>*****************************
>>
>>	If everyone more or less agree with the former goals, there is only one
>>point left to decide: extensibility. Should we allow third parties to
>>add extensions to the format? 
>>
>>	<IMHO>I think extensibility is a needed thing, not everyone will be
>>happy with what we provide so this could give the means to add the
>>needed extra functionality. Most standards out there are not "locked",
>>but allow extensions, as long as they are marked as such, and that the
>>data is usable without the extensions.</IMHO>
>>    
>>
>I think we should allow extensions, but
>	-don't define any official extensions. Everything we define 
>	 should be normal part of the spec.
>	-if extensions popup then we should imediately discuss what causes 
>	 the need for these extensions, and how we could cleanly add them 
>	 to the a new version of the spec if appropriate.
>  
>
Agreed, so then #openmodel should be a permanent thing, we don't want to 
be dispanded as soon as omf 1.0 specs are released

    Wallaba