[Home] [By Thread] [By Date] [Recent Entries]
Ian Graham wrote: > On Wed, 28 Nov 2001, Tim Bray wrote: >> It's also coherent since the NS rec explicitly >>rules out the empty string ever being used as a namespace name. > But the NS rec (Section 5.2 [1]) does say that the default namespace can > be set to an empty string ... although this is equivalent to the 'no > namespsace' partition. From the definition of "namespace prefix" in Section 2: "In such declarations, the namespace name may not be empty." The passage from 5.2 "The default namespace can be set to the empty string. This has the same effect, within the scope of the declaration, of there being no default namespace." i.e. making it empty is an assertion there's no namespace. Thus I stand behind my claim: there is no way to associate an element or attribute with a namespace whose name is the empty string. Thus this is a reasonable way of expressing the absence of a namespace in a programming language. -Tim
|

Cart



