W3C home > Mailing lists > Public > semantic-web@w3.org > July 2014

Re: Call for Linked Research

From: Sarven Capadisli <info@csarven.ca>
Date: Wed, 30 Jul 2014 10:22:01 +0200
Message-ID: <53D8AB29.3020702@csarven.ca>
To: Linking Open Data <public-lod@w3.org>, SW-forum <semantic-web@w3.org>
On 2014-07-30 09:25, Giovanni Tummarello wrote:
>     Like I said, it is mind-boggling to think that the SW/LD research
>     community is stuck on 1-star Linked Data. Is that sinking in yet?
> So Sarvem let us be rational and pick Occam's razor style simplest
> explanation
> Could we all be stupid or could it be instead an indicator of how little
> relevant LD is perceived to be (at this task, as for many others)?

I think the most obvious and simple reason is:

The publishing industry dictates how things should be written down. 
Conferences pass on the orders. Authors follow-up.

It just turned out to be that PDF is the first-class format for the 
publishers. If it was something else, authors would do that. So, no, it 
is not due to the "shortcoming" of LD as you put it - that's a more 
complicated possible explanation.

> If there was a pain that LD could really solve people would have used it
> without anyone "calling" for it. The mith of "put the data and it wil be
> useful" was nice to believe in for a while (early LOD efforts, 2007 or
> so ) but has - quite unfortunately -  disolved long since then.

You are misinterpreting my central argument.

The primary argument is not even about "LD". It actually centers around 
authors taking control of their own work at all steps. The 
recommendation to use LD is just one way to foster intelligent knowledge 
preservation and discovery.

I would even go one step back say that whatever is Web-native e.g., an 
ordinary HTML page, is still an orders of magnitude improvement over PDF 
(which is a desktop-native format). Do not tell me that PDF is a better 
way to disseminate knowledge on the Web than HTML. With HTML, there is 
at least one path towards LD. That option is open if people want to take 
it. If PDF was the better way, we'd see using LaTeX/PDF to create their 
Web pages.

The Occam's razor: authors are not adopting LD because they are told to 
use something else.

> All this while things like schema.org <http://schema.org> have exploded
> on the web and form what is a real semantic web of marked up pages. (but
> quite sadly, not a mention of this in this mailing list).

A lot of things happen outside of SW/LD mailing list. Computer Science, 
for instance. What's your point?

Off topic: When schema.org was put together, it had a history to look 
at. Smart folks behind it came up with a solution that would work for 
many. The industry was ready for it. Lets not forget all the efforts 
that tried to capture and discovery information on ordinary web pages 
e.g., Yahoo! SearchMonkey, microformats.

I am not trying to sell a vocabulary here or some exact how-to list on 
how you should construct a web page or LD paper. Start somewhere that's 
"Web-friendly" and make sure that you have total control over your 
words. No one is stopping anyone from passing a PDF copy to the 
conference along the way. We are able to do this now.

> my2c If you want to have some impact then your best bet by order of
> magnitudes is to liase with Dan Brickley (Google) to have some more
> specific schema.org <http://schema.org> markup for scientific experiments.

I do not think that effort is the best way to use our energy right now. 
The problem is not "I can't put my research on a Web page because there 
is no nice clean vocabularies for me and everyone else to use". The 
problem is "Can we get authors to start publishing their research on 
their own?" As I've said numerous times, this is not a technology 
problem. I think we have a long list of tools and services e.g., 
Wordpress?, and plenty of great vocabularies (coming out of this 
community), and technical means to do it ourself.

So, no, I completely reject your argument on why things are the way they 
are is due to LD's shortcomings.

The L(O)D efforts are successful - from GYM's search/knowledge graphs, 
to government/public making reasonable efforts to get their stuff up in 
a Web-friendly way i.e., unchaining themselves from PDF/Word etc. They 
are on board (some taking longer than the others, but still). There is 
good and noticeable progress. It is the LD research community that's 
stuck, and that's embarrassing! So, what is your real argument?


Received on Wednesday, 30 July 2014 08:22:35 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:52 UTC