8871 2010-02-03 10:53:29 +0000 Empty attribute syntax definition seems incorrect. 2010-03-10 07:13:57 +0000 1 1 1 Unclassified HTML WG HTML5: The Markup Language (editor: Michael(tm) Smith) unspecified PC Linux RESOLVED WONTFIX http://dev.w3.org/html5/markup/syntax.html#syntax-attr-empty editorial P2 normal --- 1 mail mike public-html-admin public-html-bugzilla oldest_to_newest 31415 0 mail 2010-02-03 10:53:29 +0000 H:TML states that "certain attributes may be specified by providing just the attribute name" which suggests that there are other attributes which may not be specified this way. The other HTML5 spec doesn't place any limits on which attributes may use the empty attribute syntax. 32412 1 mike 2010-02-18 18:00:55 +0000 (In reply to comment #0) > H:TML states that "certain attributes may be specified by providing just the > attribute name" which suggests that there are other attributes which may not be > specified this way. > > The other HTML5 spec doesn't place any limits on which attributes may use the > empty attribute syntax. Is there a specific change you would like to see made to the wording? Changing it to "All attributes may be specified by providing just the attribute name" would not be correct, because using the empty-attribute syntax with many (or even most) attributes is a conformance error -- because the values of the attributes must conform to a specific datatype/microsyntax. So the HTML5 spec does in fact place limits on which attributes can use the empty-attribute syntax. And the per-element subsections of "HTML elements" section of the H:TML draft indicate explicitly which attributes can be empty. 32465 2 mail 2010-02-20 00:36:26 +0000 (In reply to comment #1) > Changing it to "All attributes may be specified by providing just the attribute > name" would not be correct, because using the empty-attribute syntax with many > (or even most) attributes is a conformance error -- because the values of the > attributes must conform to a specific datatype/microsyntax. So the HTML5 spec > does in fact place limits on which attributes can use the empty-attribute > syntax. The some/all quantifier is not needed: "Attributes may be specified by providing just the attribute name." If further clarification is needed: "This sets the attribute value to the empty string, so is a conformance error if the empty string is not a valid value for the attribute." > And the per-element subsections of "HTML elements" section of the H:TML draft > indicate explicitly which attributes can be empty. It does for some allowably empty attributes, but not all. e.g. @href, @rel and @ping may all be empty on the <a> element, so the following is conformant: <a href rel ping>Foo</a> But the language in the H:TML draft appears to forbid this. 32467 3 mike 2010-02-20 03:39:42 +0000 (In reply to comment #2) > The some/all quantifier is not needed: "Attributes may be specified by > providing just the attribute name." If further clarification is needed: "This > sets the attribute value to the empty string, so is a conformance error if the > empty string is not a valid value for the attribute." That's the same as saying, "Certain attributes may be specified by providing just the attribute name, with no value.", it's just more words. > > And the per-element subsections of "HTML elements" section of the H:TML draft > > indicate explicitly which attributes can be empty. > > It does for some allowably empty attributes, but not all. > > e.g. @href, @rel and @ping may all be empty on the <a> element, so the > following is conformant: > > <a href rel ping>Foo</a> > > But the language in the H:TML draft appears to forbid this. In fact it doesn't. It says that, e.g., the allowed value of the rel attribute is "tokens", and explicitly defines "tokens" as "A space-separated list of zero or more token instances." Anyway, I did make a refinement to clarify that the value of the @href attribute is allowed to be empty. See http://dev.w3.org/cvsweb/html5/markup/spec.html?r1=1.171&r2=1.172&f=h 32873 4 mike 2010-03-10 07:13:57 +0000 EDITOR'S RESPONSE: This is an Editor's Response to your comment. If you are satisfied with this response, please change the state of this bug to CLOSED. If you have additional information and would like the editor to reconsider, please reopen this bug. If you would like to escalate the issue to the full HTML Working Group, please add the TrackerRequest keyword to this bug, and suggest title and text for the tracker issue; or you may create a tracker issue yourself, if you are able to do so. For more details, see this document: http://dev.w3.org/html5/decision-policy/decision-policy.html Status: Rejected Change Description: no change Rationale: Most attributes cannot be specified in a conformant manner by providing just the attribute name; to phrase that in a slightly different way: Using empty-attribute syntax to specify those attributes is a conformance error. Therefore, it's accurate to state that "certain attributes may be specified by providing just the attribute name", with the implication being that some cannot by specified that way.