Re: [Fwd: RE: The syntax issue]

Hi Orri,

From: "Orri Erling" <erling@xs4all.nl>
> For human authorship, it should be neither XML nor RDF since both  
> are as
> good as unwritable by humans.

This is a bit of an overstatement. I have seen humans write large  
amounts of both Turtle and XML. People write books in XML.

> RDF with blank nodes for representing trees is an order of magnitude  
> less human readable than a corresponding XML syntax.

<!-- Some tree in XML -->
<aaaaa>
     <bbbbb>
         <ccccc>value</ccccc>
         <ddddd>12.5</ddddd>
     </bbbbb>
     <eeeee>foo</eeeee>
</aaaaa>


# Some tree in Turtle
<> :aaaaa [
     :bbbbb [
         :ccccc "value";
         :ddddd 12.5;
     ];
     :eeeee "foo";
].


Honestly, I don't think there is an order-of-magnitude difference.

> This is also why we never considered XML or RDF as our own outside  
> syntax,
> even though the internal mapping schema happens to be RDF we would  
> not dream
> of anyone  interacting with this except via a SPARQL/SQL hybrid  
> syntax.

I'm prepared to accept that a custom syntax can be more user-friendly  
than either XML or Turtle. I certainly would be interested in seeing  
an example to get a better idea. Are you or someone else from the  
OpenLink crew up for translating Souri's example to an RDF Views-like  
syntax?

Raw files of Souri's XML example and of my Turtle conversion are here:
http://github.com/cygri/r2rml/raw/master/examples/emp-dept.xml
http://github.com/cygri/r2rml/raw/master/examples/emp-dept.ttl

The DB schema is here:
http://www.w3.org/2001/sw/rdb2rdf/wiki/R2RML_in_Turtle

Best,
Richard



>
>
> Regards
>
> Orri
>
>
>
> PS:  Harry, this will probably bounce from the list, can you forward  
> this
> there since this sender address has never worked with the list?
>
>
>
>
> -----Original Message-----
> From: public-rdb2rdf-wg-request@w3.org
> [mailto:public-rdb2rdf-wg-request@w3.org] On Behalf Of Harry Halpin
> Sent: Wednesday, August 25, 2010 6:20 PM
> To: Souri Das
> Cc: RDB2RDF WG
> Subject: Re: The syntax issue
>
>> I do see the point for RDF serialization and we should go for it.  
>> If the
>> mapping specification is written by people familiar with RDF (like  
>> us),
>> that will most likely be the preferred syntax.
>> However, what if we have a DB person writing the mapping. The  
>> widespread
>> familiarity of and availability of tools for XML could make the
>> XML-based syntax more suitable for their use than the RDF syntax.
>
> I would second Souri here. I think what we should have is a XML  
> syntax for
> DB designers who may "just" be getting into RDF and a RDF/Turtle  
> syntax
> for those who are more familiar with RDF. However, RDF/XML does not  
> count
> as a human or even machine-usable XML syntax for people who are not
> familiar with RDF, and while I'm tempted by EricP's suggestion have  
> having
> the best possible XML syntax that is RDF compatible, I'd have to see a
> good example to be convinced that the result will not be needlessly
> awkward. Therefore, I'd suggest that we use the simplest possible XML
> syntax and have a GRDDL (XSLT) transformation to the Turtle/RDF  
> syntax.
>
>          cheers,
>              harry
>
>
>
>>
>> Thanks,
>> - Souri.
>>
>> Juan Sequeda wrote:
>>> +1 for RDF serialization
>>>
>>> RDF people will be happy with this. But what about the DB people?   
>>> I'm
>>> guessing Richard can confirm this with the D2R experience. But  
>>> Souri,
>>> what do you think? A DB person with bare little experience in RDF,
>>> would they be comfortable?
>>>
>>> Is there another serialization that we should think about/ plan for
>>> the future?
>>>
>>>
>>> Juan Sequeda
>>> +1-575-SEQ-UEDA
>>> www.juansequeda.com <http://www.juansequeda.com>
>>>
>>>
>>> On Wed, Aug 25, 2010 at 10:11 AM, Michael Hausenblas
>>> <michael.hausenblas@deri.org <mailto:michael.hausenblas@deri.org>>
>>> wrote:
>>>
>>>
>>>> But, we should also consider having an XML syntax just b/c of its
>>>> popularity and widespread familiarity.
>>>
>>>    Hmm, I think in terms of manual editing Richard has made very  
>>> good
>>>    points
>>>    and I've so far not really seen good arguments for XML beside the
>>>    above
>>>    (which is, I think, not the strongest one ;)
>>>
>>>> However, as Ashok said, if we have more than one syntax we
>>>    should also
>>>> have tools to translate from one syntax to another.
>>>
>>>    That's easy. If we have Turtle as syntax (which I do prefer due  
>>> to
>>>    many
>>>    reasons, most of them already covered by Richard), then I'd claim
>>>    that any
>>>    RDF processor out there can immediately turn it into RDF/XML :)
>>>
>>>    Cheers,
>>>         Michael
>>>
>>>    --
>>>    Dr. Michael Hausenblas
>>>    LiDRC - Linked Data Research Centre
>>>    DERI - Digital Enterprise Research Institute
>>>    NUIG - National University of Ireland, Galway
>>>    Ireland, Europe
>>>    Tel. +353 91 495730
>>>    http://linkeddata.deri.ie/
>>>    http://sw-app.org/about.html
>>>
>>>
>>>
>>>> From: Souri Das <Souripriya.Das@oracle.com
>>>    <mailto:Souripriya.Das@oracle.com>>
>>>> Date: Wed, 25 Aug 2010 11:01:46 -0400
>>>> To: RDB2RDF WG <public-rdb2rdf-wg@w3.org
>>>    <mailto:public-rdb2rdf-wg@w3.org>>
>>>> Subject: Re: The syntax issue
>>>> Resent-From: RDB2RDF WG <public-rdb2rdf-wg@w3.org
>>>    <mailto:public-rdb2rdf-wg@w3.org>>
>>>> Resent-Date: Wed, 25 Aug 2010 15:02:36 +0000
>>>>
>>>> I have not had time to carefully go thru Richard's
>>>    justifications for
>>>> RDF serialization yet, but I think RDF serialization may be
>>> needed.
>>>> But, we should also consider having an XML syntax just b/c of its
>>>> popularity and widespread familiarity.
>>>> However, as Ashok said, if we have more than one syntax we
>>>    should also
>>>> have tools to translate from one syntax to another.
>>>>
>>>> Thanks,
>>>> - Souri.
>>>>
>>>> ashok malhotra wrote:
>>>>> If we are arguing syntax then we are done :-)
>>>>>
>>>>> If we end up with more than one syntax it would be good if it was
>>>>> possible
>>>>> to automatically translate from one syntax to the other.
>>>>> All the best, Ashok
>>>>>
>>>>>
>>>>> Michael Hausenblas wrote:
>>>>>>> I propose to proceed based on the concepts of Souri's
>>>    approach, but
>>>>>>> with an RDF serialization instead of XML as the surface syntax.
>>>>>>>
>>>>>>
>>>>>> +1
>>>>>>
>>>>>> Cheers,
>>>>>>      Michael
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>
>
>
>

Received on Thursday, 26 August 2010 22:16:49 UTC