W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > January 2013

Re: If no datatype is specified for @datatime

From: Shane McCarron <ahby@aptest.com>
Date: Thu, 17 Jan 2013 14:21:43 -0600
Message-ID: <CAOk_reHK9wmW7+58_3UEr7F99rpSm9nx1RAzcmYk4Y8wWTrYLw@mail.gmail.com>
To: Gregg Kellogg <gregg@greggkellogg.net>
Cc: W3C RDFWA WG <public-rdfa-wg@w3.org>
Markdown - really?  I will need to look into that.  Thanks1

On Thu, Jan 17, 2013 at 2:17 PM, Gregg Kellogg <gregg@greggkellogg.net>wrote:

> On Jan 17, 2013, at 9:06 PM, Shane McCarron <ahby@aptest.com> wrote:
>
> Sorry - forgot to copy the list.  The document should use the <aref>
> markup convention, and <aref> should translate to starting with an '@' just
> as in RDFa Core.
>
>
> Yes, I saw that it wasn't doing this, and didn't understand why not. I'm
> glad that that's been straightened out, as <aref> is much simpler for this.
>
> BTW, just saw on GitHub that ReSpec now accepts Markdown within HTML,
> which could make some of these spec writing chores even easier.
>
> Gregg
>
> ---------- Forwarded message ----------
> From: Shane McCarron <ahby@aptest.com>
> Date: Thu, Jan 17, 2013 at 1:26 PM
> Subject: Re: If no datatype is specified for @datatime
> To: Stéphane Corlosquet <scorlosquet@gmail.com>
>
>
> Although even if it is in orange, it should still be preceded with an '@'.
>  All of our attribute references start with an '@'.  Or they should.
>
>
> On Thu, Jan 17, 2013 at 12:23 PM, Stéphane Corlosquet <
> scorlosquet@gmail.com> wrote:
>
>>
>>
>> On Thu, Jan 17, 2013 at 1:09 PM, Gregg Kellogg <gregg@greggkellogg.net>wrote:
>>
>>> On Jan 17, 2013, at 5:31 PM, Stéphane Corlosquet <scorlosquet@gmail.com>
>>> wrote:
>>>
>>>
>>>
>>> On Thu, Jan 17, 2013 at 11:09 AM, Gregg Kellogg <gregg@greggkellogg.net>wrote:
>>>
>>>> On Jan 17, 2013, at 5:03 PM, Stéphane Corlosquet <scorlosquet@gmail.com>
>>>> wrote:
>>>>
>>>> http://www.w3.org/TR/rdfa-in-html/#additional-rdfa-processing-rules
>>>>
>>>> "If no datatype is specified, and the value does not match any of the
>>>> automatically detected xsd datatypes, a plain literal must be generated
>>>> with the associated language of the node, if available. "
>>>>
>>>> Could we be more precise regarding datatype here, and use "datatype
>>>> attribute" instead of datatype. I'm particularly interested in the case
>>>> where there is an empty datatype attribute with @datetime.
>>>>
>>>> to give more context, I found this part of the spec not very clear
>>> while working on the easyrdf RDFa parser.
>>>
>>> in other words, what does "no datatype" means? does it include the
>>> scenario when there is no datatype attribute and also when there is an
>>> empty datatype attribute? or the former only? then what happens in the case
>>> of the later?
>>>
>>>
>>> In this context, "no datatype" means that there is no @datatype
>>> attribute. If you have @datatype="", you have specified a datatype, which
>>> is the empty datatype; this is consistent with the (former) treatment of
>>> plain literals from non-text nodes.
>>>
>>
>> ok, the plain text version of the spec doesn't it justice. looking at the
>> HTML spec, datatype is in orange and therefore means the attribute, not
>> datatype in general, so nevermind, I retract my proposal. The spec is fine
>> as it is.
>>
>> Steph.
>>
>>
>>>
>>> It seems to me that a plain literal could be generated when there is an
>>> empty @datatype (in line with RDFA-CORE).
>>>
>>>
>>> Yes.
>>>
>>> To make a concrete proposal, add "If an empty datatype is present" to
>>> the second sentence:
>>> [[[
>>> If datatype is specified, it must take precedence. If no datatype is
>>> specified, and the value does not match any of the automatically detected
>>> xsd datatypes, a plain literal must be generated with the associated
>>> language of the node, if available.
>>> ]]]
>>> would become
>>> [[[
>>> If datatype is specified, it must take precedence. If an empty datatype
>>> is present, or if no datatype is specified and the value does not match any
>>> of the automatically detected xsd datatypes, a plain literal must be
>>> generated with the associated language of the node, if available.
>>> ]]]
>>>
>>>
>>> Yes, but I thought you preferred saying "if there is no @datatype
>>> attribute", to if the datatype is not specified.
>>>
>>> Gregg
>>>
>>> Steph.
>>>
>>>
>>>>
>>>> +1
>>>>
>>>> Gregg
>>>>
>>>> --
>>>> Steph.
>>>>
>>>>
>>>>
>>>>
>>>
>>> --
>>> Steph.
>>>
>>>
>>>
>>
>>
>> --
>> Steph.
>
>
>
>
> --
> Shane P. McCarron
> Managing Director, Applied Testing and Technology, Inc.
>
>
>
> --
> Shane P. McCarron
> Managing Director, Applied Testing and Technology, Inc.
>
>
>


-- 
Shane P. McCarron
Managing Director, Applied Testing and Technology, Inc.
Received on Thursday, 17 January 2013 20:22:17 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:19:58 UTC