DR-08-0012 Namespace Mapping Table v2

[Rick Jelliffe]

Shawn Villaron wrote:
> Sorry, my mapping table doesn't tell the full story.  I believe we're scoping the namespace change to only STRICT documents; does that change your view on this?
>   
Probably not.  But certainly better.

Developers will still need to duplicate their software. Examples will 
still all be wrong. WG4 still needs to warn that  IS29500:2800 Strict is 
obsolete pending the changes. (I do not believe that a change in 
namespace can be called an editorial change, since it directly alters 
bits in files.)

Is there any commitment from Microsoft that you will implement loading 
and export of strict OOXML using the new  namespace?  In the absence of 
commitment, it would be a waste of time to consider it, and it would be 
better to treat Strict IS29500 as the desirable profile of Transitional, 
a meaningful target for procurement and development, rather than moving 
Strict to being an incompatible dialect.

Personally, I think there is high value in Transitional systems being 
able to make a good stab at reading Strict documents, even if it is with 
some "graceful degradation" or  drop-out of content (with user agent 
notification.)

I also suggest that changing the namespace to make a new language may 
indeed go much further than what the BRM wanted to establish. While WG4 
is not bound by the BRM for ever more, of course, it needs to be a 
significant indicator of the expectations of the NBs for OOXML in the 
short to medium term.

I suppose this is a crucial issue to me: I think it is possible to for 
WG4 to remove BRM implementation SNAFUs in order to make sure that 
Transitional does indeed describe real current Office 2007 document. I 
think it is reasonable to infer this as the intention of the BRM: 
individual changes get re-assessed in the light of the big picture that 
emerged.  But I am not sure that BRM in making the strict/transitional 
distinction wanted them to be utterly incompatible dialects. I thought 
the BRM's instructions were along the lines of labelling some 
functionality as being appropriate for transitional only, not labelling 
some other functionality as being appropriate for strict only, if you 
see the difference it makes.

Furthermore, the change of namespace looks like the kind of thing that 
will, since it does not relate to functionality and breaks things, smell 
like a rat to people. Now it may take a vivid imagination to come up 
with a cogent reason why it is a rat (I am lacking in that imagination), 
but I think that many NBs will be suspicious of changes that are not 
obviously rooted in functional requirements of the kinds that came out 
of the BRM.   I am not trying to invoke FUD; at the least I think WG4 
will need a good story on why the change in namespace would not be 
license for Microsoft to ignore or marginalize OOXML Strict (or some 
other similar claim.)

Cheers
Rick Jelliffe