Re: Cost and access (Was Re: [ESWC 2015] First Call for Paper)

On 2014-10-03 13:36, Eric Prud'hommeaux wrote:
> Let's work through the requirements and a plausible migration plan. We need:

Agreed. In favour of taking action.

Just to separate and emphasize on the issues. The original request was 
merely:

"Will you consider encouraging the use of Semantic Web / Linked Data 
technologies for Extended "Semantic Web" Conference paper submissions?"

or

"Will you compromise on the submission such that the submissions can be 
in PDF and/or in HTML(+RDFa)?"

This, in my view, attempts to retain the existing workflow. There is 
nothing here that tries to "solve everything" (as some misinterpret or 
paint it as such). Incremental actions are preferable than throwing our 
hands into the air and running away frantically from the problem that 
the community brought it onto itself.

This is about creating awareness and embracing Web-native technologies 
for SW research submissions, provided that the final presentation (i.e., 
in PDF) complies with the requested template, which is passed to the 
publisher in the end.

Just to elaborate on that, while the submissions in the end may only be 
in PDF (although, it would be great to work it out without that, but one 
step at a time right?), the fact that the submission line acknowledges 
the importance and flexibility in creating, sharing, and preserving 
research knowledge using the technologies in what the conference is all 
about, should not be underestimated.

As a plus, authors that are on their way to going from, say HTML+CSS to 
PDF, have the opportunity and willingness to make their research 
contributions publicly accessible under a Web space that they control. 
The source method to represent this information sets the tone for the 
rest of the phases. That is, if LaTeX/Word is source, then it is extra 
work to get HTML out of that, and many would not and do not (in fact) 
bother. However, if HTML is source (for instance), then we retain that 
possibility. All meanwhile that the publisher gets their PDF (e.g., via 
HTML+CSS to print file), as well as authors fulfilling their 
academic/research requirements.

Moving on:

> 1 persistent storage: it's hard to beat books for a feeling of
> persistence. Contracts with trusted archival institutions can help but
> we might also want some assurances that the protocols and formats will
> persist as well. It would be possible to have a fallback contract with a
> conventional publisher but it's hard to see what's in it for them if
> they have to paper print everything or migrate to a new format when the
> Web loses way to something else. Maybe it's more pragmatic to forgoe
> these assurances of persistence and just hope that economic interests
> protect the valuable stuff.


This is out of my area, but as I understand it, going from digital 
source to print is just a "view" or "materializing" of said knowledge. 
History has shown that, both, PDF and HTML are sufficient for storage.

Those that wish to archive via PDF can do so. It is just a view after 
all. However, that one particular view to store knowledge need not set 
the tone for everything else. I think the tool-chain around HTML/XML 
tries to lift those restrictions. For instance, with HTML we are free to 
create any suitable presentation for any device with CSS.

> 2 impact factor: i have the impression that conventional publishers have
> a bit of a monopoly and and sudden disruption would be hard to engineer.
> How do to get leading researchers to devote their work in some new
> crackpot e-journal to the exclusion of other articles which will earn
> them more points towards tenure and grants? Perhaps the answer is slowly
> build the impact factor; perhaps it's some sort of revolution in the
> minds of administrators and funders.

I'd like to be optimistic about this and entertain the idea that, either 
the current journals evolve or a new line of journals will seek, embrace 
and truly employ the scientific method with the aid of available 
technologies. At this time, it is difficult to solely rely on 
"human-only" peer reviews, because it is time consuming and error-prone. 
If reviewers have the opportunity to better investigate, by raising the 
support that's available from machines, the truthfulness and 
reproducibility of given research can be better verified.

We are certainly heading in that direction with all the work that goes 
on in SW and other fields. The bottleneck is that, right now, it is not 
seriously given the light of day, or even tested out. When SW/LD 
conferences resist to come to terms with supporting their own 
fundamentals or visions towards research submissions, how is what we 
currently have any desirable?

Just to be clear, the original proposal is not for all of sciences to 
adopt. It is for "international" "semantic web" conferences. That's the 
minimal step we can take.

So, I agree, some revolution, or maybe just evolution on the idea of 
putting our own technologies to test will contribute towards increasing 
the impact factor of journals. Currently, the impact factor is 
artificial in the big picture because we can't and do not really check, 
verify, reproduce, or even double-check existing research at high-fidelity.

On a related note, majority of the "accepted" papers favour positive 
results. By encouraging the publication of negative results (i.e., not 
being bound to the page limits or number of acceptances allowed per 
proceeding), we can mine more from the collective information than what 
we are currently used to.

We can raise the quality of the SW scientific output, if we work with 
machines. PDF puts a bold stop on that goal because the controls that 
employ it have different interests. If the SW stack is worthwhile for 
science, we should test that head on.

> I work towards a network of actionable data just like the rest of you so
> I don't want to discourage this conversation; I just want to focus it.

I would like to propose a third point here:

3 user experience:

The Web stack works, well, pretty good. Some will argue that it barely 
holds, but, who can argue against the fact that it has somehow 
transformed our daily lives. The fact that we can view a simple HTML 
document virtually under any device that's connected to the Web (or just 
a Web browser) by definition fulfills ubiquitous computing. In contrast, 
with PDF, that is not the case. We simply can't navigate from one 
research document to another via PDF. PDF is forced to employ additional 
tooling or protocols to achieve that. Even then, it doesn't work that 
well. It breaks the user experience as the user goes from one interface 
and interaction (e.g., PDF reader) to another (e.g., Web browser). There 
are all sorts of accessibility issues or interaction limitations with 
PDF. What people/orgs behind PDF are counting on is that, print is "the" 
way to consume and share scientific knowledge. In fact, PDF is less 
human-friendly and accessible than a POSH.

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