RE: PIDF - same schema with nicer formatting (blank lines)

> * Need to make sure that the <basic> status element isn't 
> mandatory. It should be optional.

I'll change the schema and send it out in separate mail.  I don't think it
says it's optional in the proposed text that was in email.  Is there a
version of the draft with the latest text from mail incorporated?

> * I'm wondering about the use of <sequence> in some places (status?) -
> will we always wish to impose this ordering?

I think it's better in XML for formats that don't have to support an
arbitrary order to to specify a specific order of child elements, both to
make the formats simpler and also to make the format description simpler.
That's the way it is commonly done in formats like SOAP, UDDI, WSDL -- they
use a fixed sequence, not arbitrary orders of elements.  It also makes the
XML Schema far easier to read.  The other alternative is to build a more
complex model that specifically lists all the allowed orders.  It's
especially complex to allow arbitrary orders where multiple extensions from
other namespaces are allowed - would they allow all possible orders with
extensions allowed to appear between any namespace defined elements?  It's
simpler to specify an order where arbitrary orders aren't required.  I'd
suggest we stay with a fixed sequence and whereever we allow extensions, we
require that they be put last - after the impp namespace defined sibling
elements.

> -----Original Message-----
> From: Adrian Bateman [mailto:[email protected]]
> Sent: Wednesday, April 24, 2002 8:38 AM
> To: 'Carr, Wayne'; 'Graham Klyne'
> Cc: [email protected]
> Subject: RE: PIDF - same schema with nicer formatting (blank lines)
> 
> 
> Hi,
> 
> There are a some things that I want to examine in more detail with the
> schema but I haven't found the time yet. I thought it might 
> be best if I
> outlined a couple of my thoughts in a broad sense at this 
> point anyway.
> 
> * Need to make sure that the <basic> status element isn't 
> mandatory. It
> should be optional.
> 
> * I'm wondering about the use of <sequence> in some places (status?) -
> will we always wish to impose this ordering?
> 
> Incidentally, on a slightly different topic, I remember the 
> size of the
> priority value having been mentioned sometime (months) ago. I also
> recall having seen a message on the SIMPLE list recently mentioning
> mapping some SIP(ish?) value onto the PIDF priority. Is there a reason
> why we chose to use a byte for this? I'm worried that people might be
> using broader values in other environments and I'm not sure 
> if there is
> a huge benefit in the current restriction. Is this an issue to anyone?
> 
> Adrian.
> 
> 
> On 24 April 2002 16:03, Carr, Wayne wrote:
> > <?xml version="1.0" encoding="UTF-8"?>
> [snip]
> 
> 


  [reminder: [email protected] for non-technical discussions, please]