[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

NOTATION FPIs (was: <XML:SCRIPT>)

• From: Toby Speight <tms@a...>
• To: "XML developers' list" <xml-dev@i...>
• Date: 09 Jul 1998 10:59:22 +0100

Peter> Peter Murray-Rust <URL:mailto:peter@u...>

=> In article <3.0.1.16.19980708215906.bbc7f7a4@p...>,
=> Peter wrote:

>> <?XML version="1.0"?>
>> <!DOCTYPE MyDoc [
>> <!NOTATION JavaScript PUBLIC "+//IDN netscape.com//NOTATION Java
>> Script//EN" >
>> ...
>> ]>
>> ...
>>
>> This serves to connect the local name "JavaScript" to the formal
>> specification that can presumably be found at the other end of the
>> public identifier for the notation. By definition, the external ID
>> for a notation is supposed to get you the human-readable definition
>> of the notation. You

Peter> What automatic mechanism is available for finding a document at
Peter> the end of:
Peter> 	"+//IDN netscape.com//NOTATION JavaScript//EN"
Peter> Is there a set of maintained FPI servers like DNS?  Because if
Peter> not, an FPI isn't very useful to me.

It *is* useful: it's an opaque string, which can be used to look up
components (programs, beans, shared libraries...), just as MIME types
can.  You never asked for a set of maintained MIME-type servers like
DNS.  In fact, it's not difficult to write an FPI for a MIME type:

 -//Internet Assigned Numbers Authority//NOTATION MIME image/gif//

perhaps (I'm not sure of the correct owner identifier, but you get
the picture).  You could even use your existing .mailcaps file (or
equivalent) to set up some initial mappings for these FPIs.


Peter> The advantage of MIME types is that there is a well-defined
Peter> mechanism in current software for associating MIME types with
Peter> software.

That's true, but it's not without its limits (I assume you're referring to
mailcap files).  They don't have established conventions for components
other than entire programs; it's hard to define different programs for
different situations (hi-res screen, terminal screen, print, etc.)  Any
new convention for associating FPIs with handlers ought to be more general
than mailcap files.

[It's possible that the current IETF work on URN resolution services
may help with retrieving resources from arbitrary, unknown FPIs]

-- 


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• Follow-Ups:
  • Re: NOTATION FPIs (was: <XML:SCRIPT>)
    • From: Chris Maden <crism@o...>

• References:
  • Re: <XML:SCRIPT>
    • From: Peter Murray-Rust <peter@u...>

• Prev by Date: XSchema: Naming vote count
• Next by Date: XML keyword definitions
• Previous by thread: Re: <XML:SCRIPT>
• Next by thread: Re: NOTATION FPIs (was: <XML:SCRIPT>)
• Index(es):
  • Date
  • Thread