Re: Sections for HTML local implementation?

comments inline;

On 10/10/2012 09:47, Felix Sasaki wrote:
> Hi Shaun, all,
>
> I agree that having the table in an appendix wouldn't make sense. 
> Also, there is a huge overlap between the table in sec. 6.1 and your 
> table. And finally, Jirka is addressing the "how to work with HTML5" 
> in a separate section too.
>
> I have a pragmatic suggestion: next week many usual suspects will be 
> away, at localization world. But a useful topic to discuss might be 
> how to integrate above pieces and what to do about the implementation 
> sections. Do you want to continue this on the list this week and work 
> on it during next week's call?
>
> I am confident that you, Jirka and others who might be able to join 
> would develop a good approach to move forward. That doesn't mean that 
> you have to do everything: a good strategy is always to give action 
> items to people who are not at the meeting ;)
>

Hey, we do read the list you know :-)

Clearly, currently the spec comes across as an XML one with HTML tacked 
on. This is partially a hang-over form using the ITS1.0 format. However, 
this mainly appears in;
1) the data attributes definitions are presented in the same camelcase 
for as the XML binding
2) we don't have sufficient HTML examples in some data categories

2 i think we are looking at already. 1 is really easy to fix by changing 
the names used in the data attribute definitions to a form that is 
agnostic to the XML (or the HTML) form, for example change;

"GLOBAL: ThelocNoteRule 
<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNoteRule>element 
contains the following:

  *

    A requiredselector
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#att.selector.attribute.selector>attribute.
    It contains anabsolute selector
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors>which
    selects the nodes to which this rule applies.

  *

    A requiredlocNoteType
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNoteRule.attributes>attribute
    with the value "description" or "alert".

  *

    Exactly one of the following:

      o

        AlocNote
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNote>element
        that contains the note itself and allows forlocal ITS markup
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selection-local>."

    to:


"GLOBAL: Theloc note rule 
<http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNoteRule>element 
contains the following:

  *

    A requiredselector
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#att.selector.attribute.selector>attribute.
    It contains anabsolute selector
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selectors>which
    selects the nodes to which this rule applies.

  *

    A requiredloc note type
    <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNoteRule.attributes>attribute
    with the value "description" or "alert".

  *

    Exactly one of the following:

      o

        Aloc note
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#locNote>element
        that contains the note itself and allows forlocal ITS markup
        <http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html#selection-local>."

or perhaps capitalise the attribute names, so e.g. locNoteType goes to 
LOC NOTE TYPE.

Then use Shaun table in the main body of the text to make the XML and 
HTML mappings clear with equal partity.

Would that work? We'd still need to look at the section 1-5 to make sure 
that it doesn't favour XML over HTML, e.g. we have section 1.4 calling 
out usage in HTML5 while the usage in XML is more implicit, also the 
examples in this earlier section are still largely XML rather than HTML.

cheers,
Dave

> If you do some planning, you just need to make clear what to do when, 
> see the timeline at
>
> http://lists.w3.org/Archives/Public/public-multilingualweb-lt/2012Oct/0074.html
>
> [
> 1.1) PUBLICATION DATE: I propose week of 22 October. The week after is
> TPAC, and there won’t be publications in that week. See
>
> https://lists.w3.org/Archives/Member/chairs/2012AprJun/0093.html
>
> This means that the last edits need to be done by 19. October.
> ]
>
>
> Best,
>
> Felix
>
> 2012/10/10 Shaun McCance <shaunm@gnome.org <mailto:shaunm@gnome.org>>
>
>     Hi Dave,
>
>     I guess I would expect the semantics to be defined in the introduction
>     for each data category, and for the implementation sections to specify
>     how syntax maps to semantics. And I fear that if the sections for each
>     data category don't mention the HTML attributes, people won't bother
>     to look at some table in an appendix.
>
>     --
>     Shaun
>     On Tue, 2012-10-09 at 22:44 +0100, David Lewis wrote:
>     > Hi Shaun,
>     > I think the intention is that the data category definition
>     specify the
>     > semantics in a language independent manner. It's a hang over
>     from its
>     > 1.0 that the camel case names we give these attributes are the
>     same as
>     > used in the XML version.
>     >
>     > The table you produced showing how the XML and HTML mapping a to the
>     > data categories and their attributes helps make the mapping from
>     > semantics to these more explicitly.
>     >
>     > Cheers,
>     > Dave
>     >
>     > On 9 Oct 2012, at 22:07, Shaun McCance <shaunm@gnome.org
>     <mailto:shaunm@gnome.org>> wrote:
>     >
>     > > Currently, the Implementation sections for the date categories
>     give
>     > > the global and local implementation details, but the local details
>     > > generally only give the names of the XML attributes. Should we
>     have
>     > > extra sections for the local HTML implementation details?
>     > >
>     > > Also, would it be worthwhile to use actual subsections, instead of
>     > > leading paragraphs prefixed with "GLOBAL:" and "LOCAL:"?
>     > >
>     > > --
>     > > Shaun
>     > >
>     > >
>     > >
>     >
>
>
>
>
>
>
> -- 
> Felix Sasaki
> DFKI / W3C Fellow
>