Re: Linked Data Dogfood circa. 2013

On 1/8/13 11:37 AM, Sarven Capadisli wrote:
> On 01/04/2013 02:02 AM, Kingsley Idehen wrote:
>> On 1/3/13 7:50 PM, Sarven Capadisli wrote:
>>> On 01/04/2013 12:34 AM, Bernard Vatant wrote:
>>>> Dog Food People
>>>> http://data.semanticweb.org/person/
>>>
>>> [Off topic]
>>>
>>> Given that the people in that list originally published their papers
>>> using anything but machine friendly Web practices, would anyone care
>>> to enlighten me the dogfood bit in "Dog Food People"?
>>>
>>> I do think that data.semanticweb.org is doing a great thing i.e., they
>>> are the ones dogfooding! It is unfortunately a limited "patch" to a
>>> problem that the Semantic Web / Linked Data community is too careless
>>> to tackle head on!
>>>
>>> -Sarven
>>>
>>>
>>>
>> [On topic]
>>
>> So why don't we all make a concerted effort in 2013 to clean up these
>> kinds of issues. Basically, let's make dogfooding meaningful since its
>> the ultimate demonstrator of technology utility :-)
>
> Enter Sarven's Rant:
>
> I acknowledge that sometimes it is indeed complicated (e.g., too many 
> variables) to eat our own dogfood in every corner, when faced with all 
> the business' needs, for whatever reasons they may be. I don't wish to 
> debate whether the solutions that SW/LD offers are realistic enough or 
> can be fulfilled or not. We use the tools we can to get the work done. 
> Everyone does what they can.
>
> However, what's frustrating to see, not to mention the ongoing 
> facepalms, is the situation that the self-proclaimed SW/LD conferences 
> and academia puts themselves into.
>
> Every year, the conferences primarily requests the research work to be 
> submitted in PDF/Word and then maybe the source in LaTeX for 
> camera-ready versions which is then handed off to the (print) 
> publishers. Similarly, that's what happens in academia too.
>
> That is precisely what data.semanticweb.org is "patching", the bits 
> the SW/LD community shouldn't be intentionally breaking in the first 
> place. It is not good enough to get some metadata when all is said and 
> done. What's that? The leftover from the conferences and publishers? 
> G, thanks, but, no thanks, we can do better. We ought to do better.
>
> All the value in research work is locked up in desktop-friendly 
> formats (yes, Google showing a preview of the PDFs is a hack too) or 
> maybe gets printed for others researchers to consume from. But, wait a 
> minute, wasn't the SW/LD community fighting to get machines to do 
> something? You know it exactly as I do. We need to get a hold of all 
> that awesome information in those papers; from hypotheses, results, 
> claims, conclusions, references, to.. in a way that we can point at it.
>
> Is the current situation seriously something we can't fix or improve 
> on? Is the community at the mercy of cool conferences, organizations, 
> publishers, academia or die hard followers of archaic methods?

Well stated!

I've never groked the prominence of PDF (Paper Description Format) in 
realms that are supposedly religious about 5-Star Linked Data. Ditto 
those conferences that have Linked Data as the headline while being 
bemused by the concept of QRCodes incorporated into the badges of 
conference attendees etc.


>
> How about a possible solution to steer towards in the right direction:
>
> * SW/LD conferences asks research work to be submitted *first* in a 
> machine-friendly format (e.g., HTML+RDFa); if you are responsible for 
> its organization, please make it right and first serve the *Web* 
> community. Otherwise, you are the ones that's holding things back!

Yep!

>
> * If you are an author, researcher, or whatever, put your word where 
> your mouth is and *first* publish it on the Web yourself and make sure 
> it is machine friendly. Give its URI to the conference or your 
> academic institution. Otherwise, you are the ones that's holding 
> things back!

Yep!

>
> * If you are an SW/LD academic supervisor, ask and encourage your 
> students to publish or submit their work in such fashion. Otherwise, 
> you are the ones that's holding things back!

Yep!

And same applies to any EU funded project. You must publish the output 
of your funded work in a Linked Data format.

>
> Do your bit :)

Amen!

>
> -Sarven
>
>
>


-- 

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