Re: scientific publishing process (was Re: Cost and access)

On 2014-10-07 11:39, Norman Gray wrote:
> The original spark to the thread was a lament that SW and LD conferences don't mandate something XMLish for submissions because X(HT)ML is clearly better for... well ... dammit, it's Better.

Straw man argument. Please stop that now!

I will spell out the main proposal and purpose for you because it sounds 
like you are completely oblivious to them. Let me know if anything is 
unclear.

* Conferences on SW/LD research should encourage and allow submissions 
using the Web native technology stack (e.g., starting from HTML and 
friends for instance) alongside the existing requirements. As the 
required submission in PDF can be generated via HTML+CSS, those that 
wish to arrive at the PDF by their own means can still do so, meanwhile 
without asking or forcing the existing authorship or review process to 
change. It is backwards compatible. The underlying idea is to use our 
own technologies, not only for the sake of using them, but also to 
identify the pains as a precursor to raising the quality of the 
(Semantic) Web stack for scientific research publishing, discovery, and 
reuse. This is plain and simple dogfooding and it is important.

* There is an opportunity for granular data discovery, reuse, and 
machines to aid in reproducibility of scientific research. This goes 
completely beyond off the shelf metadata e.g., author, title, subject, 
or what you can stuff into LaTeX+Whatever, not to mention mangling 
around what's primarily intended for desktop and print, to squeeze in 
some "Web" in there. We are talking about making reasonable strides 
towards having scientific knowledge that is "universally" "accessible" 
on the Web. PDF and friends do not fit into that equation that well, 
however, no one is blocked from doing what they already do. Some of us 
would like to do a bit more than that to test things out so that we can 
collectively have more wins.

* There is also an opportunity to attract more funding and interest 
groups, if we can "better" assess the state of Web Science. This is 
simply due to the fact that we would be able to mine more "useful" 
information from existing research. Moreover, we can identify research 
areas of potential value better. It is to elevate the support that we 
can get from machines to excel and to do our work better. This is in 
contrast to what we can currently achieve with the existing workflow 
i.e., the current process is only concerned about making it "easy" for 
the author, reviewer, and publisher, and not about gleaning 
high-fidelity information.

> A more modest goal, which is still valuable and _much_ more achievable, is to get at least some RDF out of submitted articles.  That practically means metadata, plus perhaps some document structure, plus, if you're keen and can get the authors to invest their effort, some argumentation.  That's available for free (and right now) from LaTeX authors, and available from XHTML authors depending on how hard it would be to get them to put @profile attribute in the right places.
> That original lament has overlapped with a parallel lament that PDF is a dead-end format -- it's not 'webby'.  I believe that the demo in my earlier message undermines that claim as far as RDF goes.

Let me get this right: you are advocating that LaTeX + RDF/XML + 
whatever processes one has to go through, is a more sensible approach 
than HTML? If so, we have a different view on what creates a good UX.

It may come as news to you, but the SW/LD community is not in favour of 
authors using RDF/XML unless it is completely within some tool-chain 
left for machines to deal with. There are alternative RDF notations 
which are more preferable. You should look it up. The problem with your 
proposal is that, the author has to boggle their mind with two 
completely different syntaxes (LaTeX and RDF/XML), whereas the original 
proposal was to deal with one i.e., HTML. Styling is no more of an issue 
as the templates in the case of LaTeX is provided, and for HTML, I've 
made a modest PoC with:

https://github.com/csarven/linked-research

However, you are somehow completely oblivious to that even though it was 
mentioned several times now on this mailing list. No, it is not perfect, 
and yes it can be better. There are alternative solutions to achieve 
something along those lines with the same vision in mind, which area all 
okay too.

If this is not about coding, but rather using WYSIWYG editors or 
authoring/publication tools, have a look and try a few here or from a 
service near you:

* http://en.wikipedia.org/wiki/Comparison_of_HTML_editors

* http://en.wikipedia.org/wiki/List_of_content_management_systems

Or you know, take 30 seconds to create a WordPress account and another 
30 seconds to publish. Let me know if you still think that's 
insufficient or completely unreasonable / difficult for Web Science 
people to handle.

So, *do as you like, but do not prevent me* from doing encouraging the 
SW/LD community to dogfood or at least to give a consideration to work 
towards alternative solutions that works better for the *Web* and its 
citizens.

Have a nice day.

-Sarven
http://csarven.ca/#i