RE: [xri] Links as atomic packages of optional attributes

["Scott Cantor" <cantor.2@osu.edu>]

Eran Hammer-Lahav wrote on 2009-09-26:
> I am not sure how much we gain from that and if this is any less confusing
> than just keeping it all as element (with the new cardinality).

It's more heat than light, but if you're looking for ammo, the argument I
usually use for attributes vs. elements (aside from verbosity) is that you
get some implementers screwing up the fact that whitespace nodes inside of
an element are significant and have to be dealt with when you have content
like:

<Link>
	<URI>
	Value
	</URI>
</Link>

The actual string inside the URI element there of course has two line feeds
and a tab.

Attributes can be somewhat simpler for bad code to handle. Not much of an
argument I admit.

-- Scott