xml-dev - Re: SDD again

[alex@veosystems.com]

> 
> At 22:55 08/05/98 -0700, [several people] wrote:
> 
> [... important and occasionally emotive discussion of SDD snipped...]
> 
> This discussion exemplifies one of the roles we hoped XML-DEV could address
> - that different people could make different semantic assumptions from the
> spec. I am not surprised that this particular problem has surfaced because
> the distinction between validity and well-formedness is new and because it
> has been clear that it is a difficult one to formalise. In the current
> situation it would appear that different parser-creators can create code
> which can sometimes produce different parse trees from the same document.
> [Please correct me if I'm wrong.] These writers are not incompetent - they
> take different interpretations of what the spec requires to be done and
> under what circumstances.

I think the biggest area of grayness is when standalone=no.  There is an
operational issue of how you get to a proper standalone=yes document without
a validating parser... but that is a different issue.

...but I'm not the right person to summarize this thread.

> If - as appears - we have a spec which is fuzzy in places then we it would
> be extremely useful to know - dispassionately - where the problems lie. If
> they can be identified then we might be able to:
> 	- agree on a common course of action and/or
> 	- agree that the document creator could encode her wishes in some way and/or
> 	- agree that the parserwriter could make various options available to the
> reader/client (e.g. through commandline switches).

In my reasoning I think we should develop an industry standard test suite.
James Clark has done a wonderful job aon making a start.  We took this
test suite and built a testing system around an idea I had about 
conformance.  It works quite well but there is more work to be done.

Such a test suite and conformance to it needs to be governed by some
standards and/or consortium.  OASIS springs to mind...

> Some validating parsers (e.g. DXP) already have commandline switches (e.g.
> -v for validate) and I suspect that there will be increasing pressure to
> add more ('ignore standalone declaration' for example). If we could
> systematise these it could be one way to reduce unexpected effects of parsing.

Although we need to identify "features" that should be common... lets not
get too far down the road of standardizing command line switches, etc.
It is just not necessary at this stage.

=============================================================================
R. Alexander Milowski      alex@veosystems.com     (612) 825-4132       v|e|o
                          MOS | sed s/SG/X/g > DYX

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