[x3d-public] Minutes of X3D WG Open Meeting Wednesday 10th May 2017

Tue May 16 10:29:09 PDT 2017

Attendees:  Roy Walmsley, Leonard Daly, Anita Havele, Vince Marchetti, Don
Brutzman, Dick Puk, Michalis Kamburelis, 

Apologies received: 

============================================================================
===========================

Preliminary: Welcome, and introductions, as necessary

Roy welcomed everyone. No introductions were needed.

============================================================================
===========================

Primary discussion topic: SFString and MFString in the XML encoding - which
quotation marks should be accepted?

Ancillary topic: MFxxxx fields - use of commas as value separators

*         Introduction: Review of current and reference standards, led by
the meeting chair

o   ISO/IEC 19775-1:2013 clause 5.3.14 SFString and SFString
(http://www.web3d.org/documents/specifications/19775-1/V3.3/Part01/fieldsDef
.html#SFStringAndMFString)

o   ISO/IEC 19776-2:2015 clause 5.1.2 Description
(http://www.web3d.org/documents/specifications/19776-2/V3.3/Part02/EncodingO
fFields.html#Description)

o   ISO/IEC 19776-2:2015 clause 5.15 SFString and MFString
(http://www.web3d.org/documents/specifications/19776-2/V3.3/Part02/EncodingO
fFields.html#SFString)

o   ISO/IEC 19776-2:2015 clause A.4 Fields
(http://www.web3d.org/documents/specifications/19776-2/V3.3/Part02/grammar.h
tml#Fields)

o   Introducing JSON (http://www.json.org/)

o   ISO/IEC 19776-1:2015 clause 5.1.2 Description
(http://www.web3d.org/documents/specifications/19776-1/V3.3/Part01/EncodingO
fFields.html#5.1.2)

o   ISO/IEC 19776-1:2016 clause 5.15 SFString and MFString
(http://www.web3d.org/documents/specifications/19776-1/V3.3/Part01/EncodingO
fFields.html#SFString)

o   HTML 5.1 W3C Recommendation, 1 November 2016, clause 8.1.2.3 Attributes
(https://www.w3.org/TR/html/syntax.html#elements-attributes)

o   HTML 5.1 W3C Recommendation, 1 November 2016, clause 9.1 Writing XHTML
documents (https://www.w3.org/TR/html/xhtml.html#writing-xhtml-documents)

o   Extensible Markup Language (XML) 1.1 W3C Recommendation 4 February 2004,
clause 2.3 Common Syntactic Constructs [10] AttValue
(https://www.w3.org/TR/2004/REC-xml11-20040204/#NT-AttValue)

Roy opened the discussion with a short review of each of the above
references. The reference from ISO/IEC 19775-1 did not have specific impact
on the discussions. The next three references from ISO/IEC 19776-2, the
Classic VRML encoding, define a string to be have double quotation marks
("), and MFStrings to be an array of strings enclosed in square brackets as
array delimiters. Furthermore, an MFString with a single element does not
require the square brackets. This principle, of not requiring square
brackets for a single value, applies to all MFxxxx fields. The next two
references, from ISO/IEC 19776-1, the XML encoding, define a string to have
double quotation marks(") and MFStrings to be an array of strings in single
quotes ('). The remaining three HTML and XML references define attribute
values to be enclosed in either double quotes(") or single quotes(').

Don highlighted the existing Mantis issue on this topic that collects some
comments. This is issue 1091 -
http://www.web3d.org/member-only/mantis/view.php?id=1091 - entitled
"Confusion between XML and X3D syntax for SFString". The related Mantis
issues 705 - http://www.web3d.org/member-only/mantis/view.php?id=705 -
entitled "Escaping backslashes" and 488 -
http://www.web3d.org/member-only/mantis/view.php?id=488 - entitled "Handling
of quotation marks" were also noted.

Need to be careful with specifying what we mean by "double quotes" and
"single quotes" or "apostrophes". See
https://www.w3.org/TR/REC-xml/#sec-common-syn and
https://dev.w3.org/html5/html-author/charref for XML standard. Michalis
provided references to clarify this - see
http://web3d.org/pipermail/x3d-public_web3d.org/2017-May/006674.html.

The question was also raised about whether the requirement in the XML
encoding for an MFString with a single value to be encoded using two sets of
quotation marks could be relaxed, in effect dropping the outer set of
quotes, just as the Classic VRML encoding drops the array square brackets.

The need for backslash escaping was described. They are needed in the XML
encoding because there is both an XML parser and an X3D parser. The question
of whether there should be a difference between SFString and MFString
escaping requirements was raised. Michalis reported that some
implementations do not recognize escape sequences in SFString values - see
http://web3d.org/pipermail/x3d-public_web3d.org/2017-May/006690.html.

Michalis noted that the X3D XML encoding is specific about what quotation
marks to use, compared to XML. He suggested X3D should permit either sets of
quotes, to match the XML / HTML standards. There was general agreement that
the specific X3D requirements should be relaxed to match the XML
specification for string attributes.

It was agreed that escape sequences are indeed required for MFStrings, but
are not be required for SFStrings. 

It was proposed that alternative text for clause 5.15 in ISO/IEC 19776-1
should be written. During the discussion, there was a large number of e-mail
exchanges on the public list, all stemming from the meeting agenda at
http://web3d.org/pipermail/x3d-public_web3d.org/2017-May/006643.html. At the
end of the meeting discussions Don and Michalis agreed to put forward new
text for consideration.

============================================================================
===========================

Close of Meeting: Thanks for participating and contributing

Roy thanked everyone for their time to participate and the passionate
contributions to the debate.

Roy Walmsley

X3D WG co-chair

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20170516/5d32e907/attachment-0001.html>