See also: IRC log
<shadi> http://www.w3.org/WAI/ER/Pointers/WD-Pointers-20080623
<shadi> ACTION: ERT WG discuss issue on rdfs:domain (see http://lists.w3.org/Archives/Public/public-wai-ert/2008Jul/0002.html) [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action01]
saz: comments on pointers draft?
ms: abstract could be more verbose, general reader might not understand it
<shadi> ACTION: ERT WG discuss abstract section (to make it clearer for the average reader) [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action02]
saz: test samples TF yesterday: there are no rules about when/how to use the different pointers
<shadi> ACTION: ERT WG discuss rules or conventions for using the pointers, to ensure consistency [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action03]
ms: HTMLPointer (section 2.3.4)
<shadi> ACTION: ERT WG discuss what we gain by having a distinct HTML Pointer class [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action04]
2.3.4. HTMLPointer Class [...] The following property must be included in resources of type CSSSelectorPointer. <-- change to HTMLPointer
ci: HTMLPointer can be used when markup is not valid/well-formed
ms: to me it looks like just an XPath expression
saz: maybe it needs more explanation in the
document
... previous discussion XPath/XPointer
<shadi> ACTION: ERT WG discuss difference between XPointer and XPath (see http://lists.w3.org/Archives/Public/public-wai-ert/2008Jul/0000.html) [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action05]
<shadi> actions?
<shadi> http://lists.w3.org/Archives/Public/public-wai-ert/2008Jul/0000.html
jk: likes the XPointerPointer examples in CI's mail
ci: i just looked at the xpointer scheme, we don't need anything from XPointer framework apart from xpointer scheme, so let's reference the xpointer scheme, not the XPointer framework
+1
<MikeS> +1
saz: proposal: make XPointerPointer subclass of XPathPointer
ci: SinglePointer points to single point,
RangePointer points to a range, jk mentioned that you can select a range by
using the XPointer range function(s) within a XPointerPointer which is a
SinglePointer
... it's not the number of locations you get, but the number of expressions
you use
... also CSSSelectorPointer can point to different locations with only one
expression
... return to CompoundPointer vs. SinglePointer approach
cv: use examples with pointers in connection
wirh document snippets
... so that readers can verify how the pointer is to be used
saz: e.g. an EARL document
<shadi> ACTION: ERT WG consider a real EARL report that could be used as examples in the different documents being developed [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action06]
saz: a fifth level of hierarchy for pointer subclassing looks to complicated
<shadi> ACTION: ERT WG think about renaming RangePointer too CompundPointer [recorded in http://www.w3.org/2008/07/02-er-minutes.html#action07]
saz: why is HTMLPointer not an ExpressionPointer
ci: we currently don't know what it is
... we don't know how it works
saz: it's just not documented
... note/guidance on how to use the pointers
+1
ms: e.g. radio button group (input elements) without fieldset element results in 5 errors
saz: you could use RelatedPointer
ci: we need consistency; e.g. LineCharPointer: do you point to the first character of the start tag, or last character of end tag, or ...?
<shadi> proposal: add a new section 2 that describes the overall convention/usage (for example "point to the most specific element/attribute" etc), then add guidance/usage notes within the specific sections of the document where relevant
saz: W3C technical plenary week Mandelieu near
Cannes (France) October 20-24 2008
... we already missed the deadline, but maybe we can get a room
ms: difficult for me during that particular week
cv: I think I can manage to go to TPAC
saz: alternatives?
cv: we can host it any time in October
saz: Fraunhofer FIT St. Augustin near Bonn
(Germany)
... or at W3C office before TPAC?
no telcon next week; next telcon: July 16th