Re: Final Turtle rollup EARL report

On 8/29/13 2:17 PM, Gregg Kellogg wrote:
> On Aug 29, 2013, at 8:19 AM, Kingsley Idehen <kidehen@openlinksw.com 
> <mailto:kidehen@openlinksw.com>> wrote:
>
>> On 8/29/13 10:50 AM, Gregg Kellogg wrote:
>>> On Aug 29, 2013, at 6:14 AM, Kingsley Idehen <kidehen@openlinksw.com 
>>> <mailto:kidehen@openlinksw.com>> wrote:
>>>
>>>> On 8/28/13 8:12 PM, Gregg Kellogg wrote:
>>>>> I've integrated the last EARL report for Turtle [1]. Note, that as 
>>>>> it uses ReSpec, it can be saved out and frozen too.
>>>>>
>>>>> The report shows 10 fully conforming implementations.
>>>>>
>>>>> Gregg Kellogg
>>>>> gregg@greggkellogg.net <mailto:gregg@greggkellogg.net>
>>>>>
>>>>> [1] 
>>>>> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/reports/index.html
>>>>>
>>>>>
>>>>>
>>>>>
>>>> Gregg,
>>>>
>>>> Shouldn't the Turtle documents associated with each test suite 
>>>> submission be discoverable from these documents? Basically, their 
>>>> URLs should be referenced from this document etc..
>>>
>>> Each input file used to generate the report is listed in appendix B 
>>> [1]. How would you suggest marking this up to make it more machine 
>>> readable?
>>>
>>> Gregg
>>>
>>> [1] 
>>> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/reports/index.html#individual-test-results
>>>
>>
>> Currently, a user agent would end up with something that looks like 
>> the typical URIBurner page that describes a Web Document [1]. If you 
>> add <link/> based relations (e.g., rel="related" since "related" is 
>> IANA registered and maps to xhv:related) in the <head/> section of 
>> the HTML document, we end up with better description pages due to the 
>> fact that the "xhv:related" relations will be much more visible and 
>> comprehensible to both humans and machines :-)
>
> There are two different kinds of referenced files, alternative 
> versions of the EARL rollup (earl.jsonld and earl.ttl) and the source 
> files used to generate the report. For the former, I think that 
> xhv:alternate is most appropriate. For the latter, I agree with 
> xhv:related.

Yes, of course.

> I've updated the repot (adding Jose Labra's recent submission) adding 
> appropriate relations to both the <a> references, and <link> headers; 
> this is redundant for RDFa, but might be useful for FeedBurner.

FeedBurner or URIBurner ? :-)

>
> Also, note that "related" is not a term defined in an XHTML:RDFa 
> initial context (although "alternate" is), in any case, the report is 
> HTML5, which doesn't define any link relations; best is to add both 
> "alternate" and 'xhv:alternate" along with "related" and 
> 'xhv:related", the un-prefixed versions in the head section using 
> <link>, the prefixed versions in the body.

We map "related" to xhv:related. Anyway, you can also use URIs for 
relations that aren't in IANA e.g., 
<http://www.w3.org/2000/01/rdf-schema#seeAlso> .
>
> Re-looking at your uriburner, I do see the xhv:alternate showing up, 
> but not the xhv:related. Let me know if there are other changes you 
> thing would be useful.
I'll take a look.

Kingsley

>
> Gregg
>
>> [1] 
>> http://linkeddata.uriburner.com/about/html/https/dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/reports/index.html#individual-test-results 
>> .
>>
>> -- 
>>
>> Regards,
>>
>> Kingsley Idehen	
>> Founder & CEO
>> OpenLink Software
>> Company Web:http://www.openlinksw.com
>> Personal Weblog:http://www.openlinksw.com/blog/~kidehen
>> Twitter/Identi.ca  <http://Identi.ca>  handle: @kidehen
>> Google+ Profile:https://plus.google.com/112399767740508618350/about
>> LinkedIn Profile:http://www.linkedin.com/in/kidehen
>>
>>
>>
>>
>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen