[
Lists Home |
Date Index |
Thread Index
]
Chiusano Joseph wrote:
>
> ... UDEF attempts to solve (or at least help) this by creating
> "registered universal IDs" that provide a common frame of reference to
> establish semantics. For example, the UDEF ID of "g.9_8" means the
> following in the following contexts:
>
> MIL-STD-2549: Part Product Identifier
> X12 EDI: Product/Service ID
> etc.
>
> An XML representation would be:
>
> <ProductPartIdentifier UID=“9_5.8”>123-456-789</ProductPartIdentifier>
>
> The communicated benefit is that one can change tag names, but use of
> the UID signifies "what one is *really* talking about".
>
why bother encoding the tag name. why not just do
<!DOCTYPE SOME_UDEF SYSTEM "data:,<!ELEMENT UDEF_9_5_8 (#PCDATA) >" [
<!ATTLIST UDEF_9_5_8 ProductPartIdentifier #FIXED "TRUE">
]>
<UDEF_9_5_8>123-456-789</UDEF_9_5_8>
or
<!DOCTYPE SOME_UDEF SYSTEM "data:,<!ELEMENT UDEF_9_5_8 (#PCDATA) >" [
<!ATTLIST UDEF_9_5_8 MIL-STD-2549 #FIXED "Part Product Identifier">
]>
<UDEF_9_5_8>123-456-789</UDEF_9_5_8>
or just
<UDEF_9_5_8>123-456-789</UDEF_9_5_8>
either you have got it right or you have not.
despite the truism, that the generic identifer is just a special attribute,
what is the advantage to making the universal depend on the ideosyncratic?
...
|
