[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: Do attributes have scope? Are attributes metadata?

• From: "Len Bullard" <Len.Bullard@ses-i.com>
• To: "John Cowan" <cowan@mercury.ccil.org>,"Norman Gray" <norman@a...>
• Date: Wed, 29 Feb 2012 09:15:52 -0600

It can be muddier than that.   Consider

<wiringdiagram><title>Some Title</title>... </wiringdiagram>

Where the style sheet(s) (say FOSI) contributes "WIRING DIAGRAM".
Should the type definition/schema require the title element given that
it has to be empty or the print copy will have a double entry for that
value possibly in conflict as in "Wiring Diagram"?  The DTD says yes.
The English text only says a title has to be there.  The authority is
the English text but the measuring stick is the DTD.  The status of the
style sheet is it has the power to render the item that will be
inspected visually and redlined.  The DTD errors are factored into the
2% allowable error rate with some comments where a failure to meet an
undocumented but expected is called an error (say using pretext
attributes in xrefs instead of the idref.

Without authoritative clear unambiguous documentation, the tagger is
left testing and trying to work out a less than 2% error rate solution.

I see stuff daily where the organization's practices, habits and
opinions are provably incorrect with respect to the documented
authoritative definitions.   Once again, humans are the power but do not
have the authority so questions of data and metadata will matter.

len

-----Original Message-----
From: John Cowan [mailto:cowan@ccil.org] On Behalf Of John Cowan
Sent: Wednesday, February 29, 2012 8:47 AM
To: Norman Gray
Cc: Costello, Roger L.; xml-dev@lists.xml.org
Subject: Re:  Do attributes have scope? Are attributes
metadata?

Norman Gray scripsit:

> There are probably very few cases where the distinction between data
> and metadata is actually clear or indisputable.  The title of a book
> would seem obvious metadata, compared to the data of the textual
> content, but for the curator of a library catalogue, the book title
> is part of the core catalogue data, and it's the date the catalogue
> entry was created (say) which is the metadata.  

Indeed.  For that matter, the title may appear twice in a marked-up
book:
once, verbatim, as part of the book's content (and therefore data), and
once, possibly normalized, as part of the book's
cataloging-in-publication
metadata.

• References:
  • Do attributes have scope? Are attributes metadata?
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: Do attributes have scope? Are attributes metadata?
    • From: Norman Gray <norman@astro.gla.ac.uk>
  • Re: Do attributes have scope? Are attributes metadata?
    • From: John Cowan <cowan@mercury.ccil.org>