The change to "xm-" sounds like a good solution.
On Tue, Aug 19, 2014 at 7:21 AM, Eliot Kimber <> wrote:
Hmm. I'm surprised I didn't remember this aspect of the XML specification.
I'm also surprised that parsers don't report it, but I see that the XML
spec doesn't actually define it as an error. But it's good somebody
noticed it now. (Can I offer the excuse that it was almost 20 years ago
that we published the XML spec and I haven't seen a need to reread it
closely much since then?)
One solution would be to simply remove the "xml" from the names, leaving:
att
element
nsname
pi
However, "att" and "element" in particular are a bit too generic.
I therefore propose adding a prefix, e.g., "xm-" (XML markup) to all the
names to give:
xm-numcharref
xm-parameterentity
xm-textentity
xm-att
xm-element
xm-nsname
xm-pi
The domain short name "xml-d" also has to change, and "xm-d" would match
the proposed prefix. The XML specification only disallows the full string
"xml", it doesn't disallow "xm(not l)".
Cheers,
Eliot
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com
On 8/19/14, 12:43 AM, "Bob Thomas" <> wrote:
>Wow. What a catch, and what a pain. But, I am grateful that Sandor Kekesi
>took the time to point it out.
>
>Because the W3C XML recommendation is explicit about this, we need to
>rename the following elements:
>
>
>xmlatt
>xmlelement
>xmlnsname
>
>
>xmlpi
>
>
>
>I propose that we simply remove "xml" from each name to yield this:
>
>
>att
>element
>nsname
>pi
>
>
>
>Best Regards,Bob
>
>
>
>On Mon, Aug 18, 2014 at 6:52 PM, Nancy Harrison
><> wrote:
>
>Hi,
>
>
>I received a very interesting piece of mail (see below) as a response to
>the DITA 1.3 webinar, and I think we may need to discuss it tomorrow. We
>seem to have lost track of a small but telling piece of the XML spec,
>that reserves element names beginning with 'xml' or 'XML' for the use of
>the XML standard itself. This conflicts with our use of those letters to
>begin a number of elements in our XML Mention domain.
>
>
>We may need to revisit those element names, unless someone has a good
>reason for why we wouldn't have to follow that part of the spec.
>
>
>I responded to Sandor thanking him for his extremely pertinent comment.
>
>
>Regards,
>
>Nancy
>
>
>
>_____________
>Nancy Harrison
>Infobridge Solutions
>
>
>
>---------- Forwarded message ----------
>From: Kekesi, Sandor (LNG-HBE) <>
>
>
>
>
>Date: Mon, Aug 18, 2014 at 11:12 AM
>Subject: DITA 1.3 "XML mention" question
>To: ""
><>
>
>
>Hi Nancy,
>
>I attended the 8/7 webinar and had a question on the DITA Domains
>component of the presentation, in particular “XML mention”. I wasn’t
>sure how to direct this question to the TC so I hope you don’t mind that
>I’m emailing you directly.
>
>So on to my question: I was wondering if the TC had considered the
>proposed naming of the XML mention markup “xmlelement” might be in
>conflict with the following (highlighted) portion of the XML TR:
>
>
>http://www.w3.org/TR/xml/
>3 Logical Structures[Definition:
> Each XML
> document <http://www.w3.org/TR/xml/#dt-xml-doc> contains one or more
>elements,
> the boundaries of which are either delimited by start-tags
><http://www.w3.org/TR/xml/#dt-stag> and end-tags
><http://www.w3.org/TR/xml/#dt-etag>,
> or, for empty <http://www.w3.org/TR/xml/#dt-empty> elements,
> by an empty-element
> tag <http://www.w3.org/TR/xml/#dt-eetag>. Each element has a type,
>identified by name, sometimes called its "generic identifier" (GI), and
>may have a set of attribute specifications.] Each attribute specification
>has a name <http://www.w3.org/TR/xml/#dt-attrname> and
> a value <http://www.w3.org/TR/xml/#dt-attrval>.
>Element[39]
>element
> ::=
>EmptyElemTag <http://www.w3.org/TR/xml/#NT-EmptyElemTag>
>| STag <http://www.w3.org/TR/xml/#NT-STag> content
><http://www.w3.org/TR/xml/#NT-content> ETag
><http://www.w3.org/TR/xml/#NT-ETag>
>[WFC: Element Type Match] <http://www.w3.org/TR/xml/#GIMatch>
>[VC: Element Valid] <http://www.w3.org/TR/xml/#elementvalid>
>This specification does not constrain the application semantics, use, or
>(beyond syntax) names of the element types and attributes,
>except that names beginning with a match to (('X'|'x')('M'|'m')('L'|'l'))
>are
> reserved for standardization in this or future versions of this
>specification.
>
>
>My team here at LexisNexis has been using DITA for a few years now; we
>had designed a similar domain customization to delimit XML and PATH
>structures, and at first we similarly named these elements beginning with
>“xml…” but then reversed
> course when we realized that the XML TR mentioned that such naming is
>reserved.
>
>Just wanted to bring this to your attention and curious to know if the TC
>had indeed considered this and to learn their rationale for whether there
>is a conflict or if there’s an allowable usage of an element name
>beginning with “xml” here.
>
>Thanks,
>
>Sandor Kekesi
>LexisNexis
>Sr. Content Engineer
>Jersey City, New Jersey
>201.435.2051 <tel:201.435.2051>
> <mailto:>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>--
>Bob Thomas+1 720 201 8260 <tel:%2B1%20720%20201%208260>
>Skype: bob.thomas.colorado
>Instant messaging: Gmail chat () or Skype
>Time zone: Mountain (GMT-7)
>
>
>
>
>
>
>
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail. Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
--
Bob Thomas
+1 720 201 8260
Skype: bob.thomas.colorado
Instant messaging: Gmail chat () or Skype
Time zone: Mountain (GMT-7)
