Re: Proposal: Looking inside tables

On Wed, Aug 14, 2013 at 2:07 PM, Thad Guidry <thadguidry@gmail.com> wrote:

> In my mind, I was sort of flipping things around...and telling a parser
> that the set of attributes are my own, or borrowed from Schema.org
>
> I was thinking along the lines of an "attribute set" being dictated by the
> vocab=
>
> So, in my example, the "category", "class", and "rowstype" attributes are
> assumed from Schema.org ... where each would be assumed to mean
> schema.org/category , schema.org/class , schema.org/rowstype  without
> having to fully express each one.  I was thinking having the vocab= would
> mean that all the attributes expressed inside the <table> tag would be
> coming from whatever was on the right side of the = equals sign.
>

That would not work for a few reasons. First the class attribute is already
defined in HTML, so how would the parser know whether the class attribute
should be ignored (if used in the CSS sense), or taken into account (
schema.org table extension).

Secondly, this would not validate.

Finally, that's just not how RDFa and microdata work: they each have their
own attributes inside which you add tokens corresponding to
schema.orgterms... so this would be a major departure from the already
standardized
formats for expressing structured data in HTML... The way I understand this
proposal is that it's a DSL on top of RDFa/microdata, but it remains fully
compatible with HTML. It would require parsers to have a specific
extensions to support this markup (one could imagine a preprocessor
distributing the right attributes to the row HTML element so that a regular
parser can parse the data).

Steph.



>
> But as Dan pointed out....that is probably not a winning design.
>
>
>
> On Wed, Aug 14, 2013 at 12:39 PM, Stأ©phane Corlosquet <
> scorlosquet@gmail.com> wrote:
>
>>
>>
>> On Wed, Aug 14, 2013 at 11:17 AM, Thad Guidry <thadguidry@gmail.com>wrote:
>>
>>> Quick feedback Omar and others,
>>>
>>> <table typeof="Painting" vocab="http://schema.org/">
>>>
>>> I would rather see "typeof" be renamed to "rowstypeof" or simply
>>> "rowstype" to truly indicate that all the rows themselves have the implied
>>> type and not the Table Set type or class.  This way we can reserve the
>>> Table "typeof" for higher kinded types, classes, and categories if need be ?
>>>
>>> My thinking and expression would look something like this :
>>>
>>> <table typeof="Artwork" rowstype="Painting" vocab="http://schema.org/">
>>>
>>> cooler ideas also embedding the use of  http://schema.org/Class :
>>>
>>> <table category="OnSale" class="Artwork" rowstype="Painting" vocab="
>>> http://schema.org/">
>>>
>>
>> In addition to what Dan said, these new attributes would also not be
>> valid HTML. Sticking to existing attributes will avoid validation issues.
>>
>> Steph.
>>
>>
>>>
>>> Thoughts ?
>>>
>>>
>>>
>>> On Tue, Aug 13, 2013 at 4:04 PM, Omar Benjelloun (ط¹ظ…ط± ط¨ظ†ط¬ظ„ظˆظ†) <
>>> benjello@google.com> wrote:
>>>
>>>> Hi,
>>>>
>>>> Many useful datasets on the Web take the form of tables. The goal of this
>>>> proposal is to provide a simple, schema.org-friendly way to "look inside"
>>>> these tables, and map their contents into triples.
>>>>
>>>> This is an early draft proposal developed at Google. We're seeking
>>>> feedback from the community.
>>>>
>>>> The proposal is attached to this e-mail, and will be uploaded to the
>>>> WebSchemas/SchemaDotOrgProposals page shortly.
>>>>
>>>> Thanks,
>>>> -Omar
>>>>
>>>
>>>
>>>
>>> --
>>> -Thad
>>> Thad on Freebase.com <http://www.freebase.com/view/en/thad_guidry>
>>> Thad on LinkedIn <http://www.linkedin.com/in/thadguidry/>
>>>
>>
>>
>>
>> --
>> Steph.
>
>
>
>
> --
> -Thad
> Thad on Freebase.com <http://www.freebase.com/view/en/thad_guidry>
> Thad on LinkedIn <http://www.linkedin.com/in/thadguidry/>
>



-- 
Steph.