[rfc-i] XML/HTML format: anchor namespaces

[julian.reschke at gmx.de (Julian Reschke)]

Hi there,

as far as I can tell, the vocabulary spec does not restrict the value of 
"anchor" attributes beyond the limitations inherent in XML.

The HTML draft has in 
<http://greenbytes.de/tech/webdav/draft-iab-html-rfc-02.html#rfc.section.5.1>:

"HTML elements that are generated from XML elements that include an 
"anchor" attribute will use the value of the "anchor" attribute as the 
value of the "id" attribute of the corresponding HTML element. The prep 
tool produces XML with "anchor" attributes in all elements that need 
them. Some HTML constructs (such as <section>) will use multiple 
instances of these identifiers."

But then it does on defining lots of IDs that will be automatically added.

We need to somehow manage collisions. I can think of some:

1) Have a hard-wired list of IDs that can not be used in the XML source. 
This will be tricky to maintain and explain, and will break existing XML.

2) Rewrite author ids that collide with auto-generated ones; that may 
lead to surprising results when people try to link to certain items; 
it'll also be problematic if formatters for different output formats 
(say, HTML and PDF), rewrite them inconsistently.

3) Have all system-generated IDs use a common, reserved, prefix, such as 
"rfc." (maybe with a small number of exceptions, for things like 
"sec-....").

Best regards, Julian