[tei-council] documenting xml:space

[Rebecca Welzenbach]

Hi all,

One of my tasks was to reconcile two tickets dealing with the xml:space
attribute.

1) Bug 3223636 (closed before the September F2F)
2) FR 3554294

Currently, Chapter 1 of the Guidelines describes the usual treatment of
whitespace by XML processors and suggests that in TEI there is usually
little reason to change this--but if required, the encoder can use
@xml:space and should refer to the XML spec for guidance. (see:
http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ST.html#index-body.1_div.1_div.3_div.1_div.1_div.4
)

FR 3554294 asks the Guidelines to use clearer and stronger language in
explaining this, and proposes some sample language. It also asks for these
recommendations to be included in the xml:space attribute spec. Most what
it says, though, is already accounted for in the prose of the Guidelines.
There are two exceptions:

1) "If encoders expect applications to process whitespace otherwise, this
should be noted in <encodingStmt>."

Here I think he means <encodingDesc>. This warrants discussion and perhaps
its own ticket: do we want to expand <normalization> to include how the
encoder expects a processor to handle whitespace? In addition, although it
is not mentioned in the ticket, in his wiki page, John suggests that even
in cases where no modification to whitespace normalization is expected,
encoders should state this explicitly, rather than letting it be implied by
omission. Do we agree?

2) "For further background and recommendations, see XML Whitespace in the
TEI Wiki (http://wiki.tei-c.org/index.php/XML_Whitespace) and the XML
specification (http://www.w3.org/TR/REC-xml/#sec-white-space)." Do we have
a policy or practice about linking from the Guidelines to wiki pages?

I propose:

   - Close this ticket. The things that John is asking for have been, or
   are being, addressed elsewhere. John asks for these recommendations to be
   added to the attribute spec, but I propose that we stick with handling them
   in the prose.
   - Pending council feedback, open a new ticket to address the question of
   documenting assumptions about whitespace normalization in the header
   - Open a new ticket to revise/edit for clarity the prose dealing with
   whitespace at 1.3.1.1.4. The prose as it stands is not wrong: it is
   consistent with what John asks for in this ticket, and covers the same
   ground, but it could be clearer and easier to read.

Best,

Becky