dita — archive

[Dana Spradley]

  
  


Glad to hear you also think it's an important
feature, Eliot.


I'm sure we can work out the details during the 1.2 development process.


I didn't want to overengineer it in the proposal.


Nor afterwards, for that matter...enough of that in the high tech world
already.


--Dana



W. Eliot Kimber wrote:

Dana
Spradley wrote:

  
A brief, perhaps placeholder, proposal for
DITA 1.2 to support implementation-specific attributes on arbitrary
elements.

  


Obviously we need to put some design thought into this feature, but I
think that having this apply to *any* unrecognized attribute is
probably a little too strong.


I would suggest that it would make more sense to have a separate
"metadata attributes" declaration attribute. DITA-aware processors
would then use this attribute to determine which attributes are in fact
metadata attributes, e.g.:


<p metadata-atts="foo bar"

  foo="a metadata value"

  bar="another metadata value"

  fred="not a metadata attribute">


Cheers,


Eliot