On 2013-10-16 12:24, John S Warren wrote: 

> Just to clarify, the reason I believe it is important to lay out 
> high-level design alternatives and their implications is because it 
> will help in making decisions about how the Message class is to be 
> changed. In other words, the requirements for a class's behavior 
> might be drastically different, depending on whether it is a 
> replacement for an existing type (in which case all the possible 
> use-cases of the existing vs. new type need to be taken into 
> account) or it is going to be used in a new context, where adherence 
> to existing behaviors is not a factor. 
> 
> My hope was that the design alternatives might be discussed, 
> possibly other alternatives proposed, and finally an alternative 
> chosen before proceeding with discussions about implementation 
> details.

Hi John, 

Have you seen the discussion from the last Oslo meeting?
http://eavesdrop.openstack.org/meetings/oslo/2013/oslo.2013-10-11-14.00.log.txt


I think a lot of these issues came up there. 

-Ben 
 
_______________________________________________
OpenStack-dev mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Reply via email to