RE: Error recovery

I am strongly in favor of seperateing out the relaxed and strict specs.
My preference for ease of simplicity and clarity is for the strict spec to be the basis
And this "SMLDM" spec be derived from it rather than the other way around.
That way the MicroXML as we have been discussing for months still stays the standalone smaller spec.   Also one could conceive of many variants of relaxed specs based on MicroXML ...
If we had more than one ... James's way would be multiple inheritance with restrictions,  where as if MicroXML was the root then we have simple derivation with extensions.
Also adding more derived specs as people gain experience is easier than going back and changing the basis of everything.




-----------------------------------------------------------------------------
David Lee
Lead Engineer
MarkLogic Corporation
dlee@marklogic.com
Phone: +1 812-482-5223
Cell:  +1 812-630-7622
www.marklogic.com<http://www.marklogic.com/>


From: James Clark [mailto:jjc@jclark.com]
Sent: Wednesday, December 05, 2012 5:51 PM
To: David Carlisle
Cc: public-microxml@w3.org
Subject: Re: Error recovery

On Wed, Dec 5, 2012 at 5:22 PM, David Carlisle <davidc@nag.co.uk<mailto:davidc@nag.co.uk>> wrote:
On 05/12/2012 08:37, James Clark wrote:
or (more likely?) by redefining valid name tokens in the data model
to be unconstrained arbitrary strings.


I still prefer this option.

yes I think I a coming to that view too, but would prefer that the data
model part of the micro-xml spec acknowledges that rather than this
parser generating something that is in no-mans land specification wise.

On the other hand, there is something weird about a data model that calls itself the "MicroXML" data model not always being able to be represented in MicroXML syntax.

Here's another possible approach to slicing things up:

1. Define the relaxed data model in a separate specification, say, Simple Markup Language Data Model (SMLDM). This would

- define the data model (same as in the MicroXML spec but without the constraints on names and data characters)
- define the JSON syntax
- define the mapping to the XML infoset
- explain that it can be used for MicroXML, XML 1.0 (various editions), XML 1.0 + XML Namespaces, HTML, Markdown etc

2. The MicroXML spec references SMLDM and says that the MicroXML data model is SMLDM plus constraints on names and data characters.

3. The MicroXML error recovery spec references SMLDM and says that in general it produces SMLDM but there's an optional fixup stage that can make it comply with the constraints of the MicroXML data model.

James