Re: Publication of scientific research

On 04/29/2013 10:56 PM, Kingsley Idehen wrote:
> On 4/29/13 4:21 PM, Sarven Capadisli wrote:
>> On 04/29/2013 10:06 PM, Kingsley Idehen wrote:
>>> On 4/29/13 3:23 PM, Sarven Capadisli wrote:
>>>> On 04/29/2013 09:05 PM, Kingsley Idehen wrote:
>>>>> On 4/29/13 1:29 PM, Andrea Splendiani wrote:
>>>>>> Hi,
>>>>>>
>>>>>> ok. Let's see if we can offer xhtml+RDFa as an additional format, and
>>>>>> see how people react. I'll spread the idea a bit.
>>>>>
>>>>> Why stop at xhtml+RDFa when you also have:
>>>>>
>>>>> 1. html+microdata
>>>>> 2. html+turtle -- where you use <script/> for embedding Turtle.
>>>>>
>>>>> Note,  picking winners (overtly or covertly) is always a shortcut to
>>>>> politically induced inertia. It's best to do the complete opposite
>>>>> which
>>>>> has the net effect of demonstrating the innate dexterity of the RDF.
>>>>
>>>> Sure, why not. We can do all of that. Not the challenge.
>>>>
>>>> Will you get the ISWC organizers to accept *HTML*?
>>>
>>> If I had such influence, of course :-)
>>>
>>>> That's what I would love to hear.
>>>
>>> You heard it now.
>>>
>>>> The rest is really details. We can have 20 different machine readable
>>>> versions of the document if we want. Lets have 1 that's acceptable to
>>>> get things rolling!
>>>
>>> Yes, but why do you think xhtml+rdfa is the one? My point is that we
>>> don't know "the one", because that shouldn't matter in a world of URIs
>>> and RDF based Linked Data :-)
>>
>> You are right!
>>
>> I was proposing (X)HTML(+RDFa) because that's arguably most common and
>> simple enough to carry forward.
>
> I doubt it is. I even doubt its broad use.

What's your comparison?

I was comparing it to other; RDF formats, Microdata, and maybe even 
microformats (which may even be more common). I could be wrong. I'm not 
fixated on any as long as we are moving away from PDF :)

>> By having the accompanying CSS which follows the widely used
>> presentations, it is fairly on an equal footing with the currently
>> dominant format. It keeps reviewers happy since the "papers" are
>> fairly consistent.
>>
>> If we are willing to hack around getting structured data in and out of
>> PDF, I'm sure we can run circles around that via HTML+Whatever.
>
> I would hope so, and that's vital.
>
>> I just didn't want us to get lost in those possibilities and missing
>> the main mark :)
>
> So you have negate the inadvertent introduction of new hurdles :-)

The primary win IMHO is to use HTML. Whatever it is accompanied with is 
a bonus considering the current state of things. Publishing the research 
on a webpage is closely tied to that because we can actually start to 
unleash the work on the Web in a friendly manner, as opposed to putting 
up a format that's intended for the desktop on the Web.

The fact that PDF can carry RDF, or that some search companies are able 
to present a view of the cached PDFs directly in the browser (like a 
webpage or an application) are nice things. But they are merely some of 
the workarounds to the problem that we got ourselves into.

-Sarven