James M Snell wrote:
Robert Sayre wrote: > I'm not sure they are that antagonistic, but I agree with Paul when he > says they could be added later. If profiles aren't supposed to matter to > "Core Atom" processors, why not put our money where our mouth is and > make profiles declare themselves in their own namespace?
Suppose I want to require that a feed head MUST specify at least one category element for a specific type of use case. (just pulled the example from the air). I want folks to be able to know that I'm using alternative containment requirements. I want everything else to remain the same so that processors that aren't familiar with my modified profile can simply ignore it and process according to the core. Just so I understand exactly what you're thinking on this, how would I handle this using the extension approach? Could you provide some strawman angle brackets?
So, you're looking for a way to include a "schema" association in the feed, and you want a standard way to do it. The only processors that will do anything useful with this information are those that know about the "profile".
<feed ...> <head> <ex:MyPetSchema>GottaHaveCategories</ex:MyPetSchema> </head> ... </feed>
You're using Atom, and processors who know about your additional requirements will know about your schema. What you're asking is for the spec to lend weight to your extensions, and I'm not very positive about that.
Robert Sayre
