RE: Processing instructions for validating that a document may access data

> From: James A. Larson [jim@larson-tech.com]
> Sent: Wednesday, 16 March, 2005 11:54
> To: Paul Grosso; Norman Walsh
> Cc: Brad Porter; connolly@w3.org; www-voice@w3.org
> Subject: Processing instructions for validating
>          that a document may access data
> 
> Paul Grosso and Norman Walsh
> Co-chairs, XML Core
> 
> I'm writing to you as co-chairs of the XML Core 
> Working Group about an issue that Dan Connolly 
> believes will be of interest to you.... 
> Dan Connolly has raised an 
> issue[1] with the VoiceXML 2.1 last call working 
> draft, specifically about the Processing Instruction 
> section [2] which exposes an XML document referenced 
> by the <data> element via the DOM to a voice 
> application in which the interpreter should 
> validate that the host requesting the document 
> is allowed to access the data.
> 
> [1] http://lists.w3.org/Archives/Public/www-voice/2005JanMar/0065.html
> 
> [2]
http://www.w3.org/TR/2004/WD-voicexml21-20040728/#sec-data-security 

Jim, et al.,

At my suggestion, the XML CG discussed this at our 
meeting this week.  I've also done some more research
on my own.

I read at [2]:

 Before exposing an XML document referenced by the
 <data> element via the DOM to a voice application,
 the interpreter must validate that the host requesting
 the document is allowed to access the data. This
 validation is performed by comparing the hostname
 and IP Address of the document server from which
 the document containing the <data> element was
 fetched to the list of hostnames, hostname suffixes,
 and IP addresses listed in the <?access control?>
 processing instruction included in the XML document
 referenced by the <data> element. 

I understand that the information whose access is being 
discussed is the XML information being referenced by 
the VoiceXML <data> element, and the access-control PI 
is embedded within that referenced XML information, not 
within the VoiceXML document.

At http://www.w3.org/TR/2004/WD-voicexml21-20040728/#sec-data
it says:

 The <data> element allows a VoiceXML application to
 fetch arbitrary XML data from a document server
 without transitioning to a new VoiceXML document. 

This implies to me that the referenced document--in
which the access-control PI is to be embedded--is
arbitrary XML, not necessarily something written in
the VoiceXML vocabulary.  Is my understanding correct?

We had at least one member of the XML CG think that
using elements/attributes made more sense in the case
that the access-control information was being embedded
in a document using the VoiceXML vocabulary, but we 
had no one (on the call this week) that expressed any
serious reservations about using a processing instruction
to embed access-control information in an arbitrary
XML document.

While it is true that VoiceXML could invent, say, a 
vxi:access-control element (where vxi is a prefix 
bound to a special VoiceXML "instance" namespace) that 
could be placed within any document instead of some
access-control PI, personally, I would probably lean 
toward the PI in this case.

Regardless, neither I nor anyone on the XML CG expressing 
an opinion in this matter can see any strong technical
argument against using a PI in this case.

paul

Received on Thursday, 17 March 2005 16:40:33 UTC