Re: [xml-dev] Are namespaces actually crypto-entities or crypto-links? (

[Arjun Ray <arayq2@gmail.com>]

 On Fri, 23 Jul 2021 12:55:24 +1000, Rick Jelliffe
<rjelliffe@allette.com.au> wrote:


(To answer the quesion in the title: crypto-links, because they are
"names" and referential.) 

| There is already another good standard syntax for namespace 
| declaration if we disallow redeclaration and defaults and require 
| predeclaration: text entity declarations.
| 
| <!ENTITY ns "http://www.whatever.com/"; >

Wouldn't notation declarations work just as well?

<!NOTATION ns PUBLIC "http://www.whatever.com/"; >

(Note: URIs of that form also qualify as minimum literals! :-))

The advantage of entity declarations is only in the substitution to
get the Clark form explicitly, but I'm not seeing the benefit from
that. 

| 3. Getting rid of DTDs would not be, to my thinking, a goal in itself, 
| but merely a side effect of achieving the non-modal/parallelizable goals.

The DTD is a grab-bag, but among the things in it are the metadata
that could be needed for an "informed parse", making otherwise
implicit conventions explicit.  Doing away with metadata in a DTD
means placing the metadata in some form within the instance (otherwise
you're back to implicit understandings and DWIM-ing), which generally
means more syntax to distinguish the metadata from ordinary content.  

So the issue is the extra instance syntax, that perhaps does the least
damage to the canonical SGML-derived forms.

I'd suggest the data attribute syntax. 

	xxx [ yyy = "zzz" ... ]

You could have this within a start tag.  Example:

	<tag xmlns [ foo = "foo-URI" ] foo [ att1="val1" att2="val2" ] >

| 1. Go small. Have namespace but no declarations: only support a hardwired
| set of standard prefixes, such as all W3C, DC, ISO or whatever, plus
| whatever namespace-prefix mappings are known by the parser: the rest must
| use explicit Clark notation, a disincentive but a workaround.

Not very attractive, as i'm sure you're aware.

| 2. Go big. Have these link tags to provide a more generalized mechanism.

Go big, because extra syntax is _needed_. No way around this.