Re: Publication of scientific research

On 04/24/2013 08:01 PM, Hugh Glaser wrote:
> Thanks Sarven, Trying to make sure I understand: So you are saying
> that conferences should say that it will accept (only?) HTML. And so
> the process for paper production for me is to save/export from
> Microsoft Word as HTML, instead of doing a "print" to PDF, which is
> what I usually do? Or do you have some other HTML production system
> in mind?

Hi Hugh. Yes, I think HTML(+RDFa) would be ideal as far as data and 
structure is concerned. For presentation, naturally we use CSS, and if 
behaviour is needed we have JavaScript. So, one can style their papers 
in ACM or LNCS or XYZ. I wrote these stylesheets already! For everything 
else in between, we have other SVG, MathML to cover majority of the 
things that goes into papers these days.

As far as how people end up with HTML, I don't know. There are a number 
of ways I suppose. People use Word or WYSIWYG editors, or simply type it 
out. LaTeX, or anything else for that matter can still be transformed to 
HTML.

As PDF is still the dominant format, what I've done in the past is 
"print to PDF" the HTML+RDFa page (given that it used ACM, LNCS, or a 
thesis style). Needless to say, this doesn't get too far at the moment, 
in majority of the cases since the source LaTeX needs to be provided 
sooner or later.

I'm not proposing some fixed rules - this or nothing. If conferences 
want to accept PDF as a secondary format (even though I personally don't 
like this idea), it might still be reasonable for the remaining hardcore 
PDF fans. They can figure that part out. I'm only trying to emphasize on 
the need for us to get our priorities in order.

What I'm also saying is that we i.e., organizers, authors, and funders, 
all have an equal responsibility and need to contribute their own share.

Seeing major Semantic Web / Linked Data conferences announcing the 
events in these lists, meanwhile requesting the works to be submitted in 
*everything but* what they are trying to accomplish is rather contradictory!

> I hate PDF with a passion, by the way, but in the socio thingy of
> being an editor of a proceedings, it can be an enormous pain when
> people submit HTML that has local links to images, etc., even from MS
> Word documents.

Just as conferences recommend template X to be used, it is no different 
for HTML. We can prepare these templates and stylesheets. But you are 
right, the potential issue with things like that is definitely there. 
There is probably no sure way of avoiding that. They need to fix their 
bugs eventually I suppose. I don't know. It'd be great if authors simply 
published their work at their URL and just "submit" or notify the 
reviewers of its location. Reviewers can save the document (at review 
time) to their local disk. With that, they get a snapshot for review. If 
authors want to continue to improve their work for the rest of society 
at that URL, they can go nuts. Reviewer need not be concerned. Oh, and 
how about an open comment system? Did we just accidentally, or 
indirectly improve the broken *closed* peer review system?

Again, I don't see a problem with the technologies nor our ability to 
put them to use. If one can type /paragraph, I'm sure they can manage 
<p>, if they don't know it already. For sure, WYSIWYG type of tools are 
still very important. If the conferences make the call, people might 
actually follow. But, I'd love to see everyone involved step up to the 
plate.

-Sarven