Re: Final CFP: In-Use Track ISWC 2013

On Thu, May 2, 2013 at 10:16 AM, Sarven Capadisli <info@csarven.ca> wrote:

> On 05/02/2013 08:12 AM, Paul Groth wrote:
>
>> As you know, the community produces the semantic web dogfood site
>> that contains Linked Data about the papers published by the
>> community's conferences and workshops. This refers to that.
>>
>
> Thanks.
>
>
>  I thought my response and Carsten's  addressed your email with
>> respect to LISC. There was also a very long thread appearing
>> afterwards talking about the various aspects to this around
>> publication formats. I think you have to acknowledge that there are
>> disagreements with your position with respect to technology for
>> publication.
>>
>
> The disagreements that you speak of are completely off topic.
>
> The core of the discussion is to have plain old simple HTML welcomed.
> Maybe I've miscommunicated that at some point. It is really not a debate
> around whether to use HTML5 or XHTML or whichever accompanying semantic
> representations for the data, and being a show-stopper. It is pretty much a
> matter of fact that there is a difference of opinion on that, and it is not
> going to be resolved in the near future. Trying to resolve that is not the
> best use of our time. Those that want to go for it, no one is stopping them.
>

Please read more careful from conference chairs like myself and Daniel
Schwabe why experiments including even "plain old" HTML have failed so far
unfortunately. Namely, broken links to images, lack of convertibility to
PDF needed for book printing, etc.

What I do (see for example, my thesis [1]) is publish HTML versions of
paper on my website when I have the time. However, the main problem even
with self-publishing HTML of academic papers is that tex2html, which is
necessary for the conversion (I don't use Word) seems not to have been
updated since like 1996 :)

The  burden of making a working toolchain I think falls on those who want
it. I'd like it, but not enough to devote the time to work on it as my time
is already overbooked.

   cheers,
      harry

[1]http://www.ibiblio.org/hhalpin/thesis




>
> At the end, it really doesn't make a difference any way which (X)HTML to
> use when our goal is a compromise: 1) use some HTML + CSS (with style guide
> requested) 2) get a resulting view (e.g., in PDF) that's fairly consistent.
> That's something important that we can accomplish. It doesn't matter if it
> is also accompanied with RDFa, Microdata, miroformats - the community can
> sort that out as they wish.
>
> See also example responses [1] [2] [3] [4] that's trying to keep it
> focused.
>
> Here is one proof of concept [5].
>
>
>  I don't know why your pessimistic - we are publishing rdf data about
>> our publications , more authors are providing links to code and data
>> sources, we have open access forums for our papers. Journals like SWJ
>> are experimenting with new forms of peer review.
>>
>
> Those are all important but different issues.
>
> If someone is doing "science", it goes without saying that the data needs
> to be provided as well as a clear path to reproducing the same results. And
> the possibility to communicate due to open access to that knowledge. That's
> not some feature that SW/LD community invented and lets not get over
> ourselves by pretending to have something remarkable in place. It is the
> minimum requirement for sciences. No source code? Cya! Any barrier to
> acquire that knowledge? Cya!
>
> Lets be clear on "we are publishing rdf data about our publications". It
> is exactly that; "about" it. There goes an opportunity to have countless
> invaluable structured statements. Paper after paper. Even if some RDF is
> stuffed into PDF, c'mon, that's a total hack as far as LD is concerned and
> is mostly metadata. We should aim higher in that regard.
>
>
>  The linked data community is full of experiments and great stuff to
>> do better science.
>>
>
> If you want to participate in an experiment, *please*, welcome a flavour
> of HTML and if the committee wants, their favourite structured formats.
> That would mean a lot. Some other track, journal or workshop can do their
> own thing as long as we have a base like HTML.
>
> If you can't do that, I would like to learn about the main reason and not
> off-track on minute discussion points given the big picture. Whatever it
> is, I want a fix and make it happen.
>
> [1] http://lists.w3.org/Archives/**Public/public-lod/2013Apr/**0376.html<http://lists.w3.org/Archives/Public/public-lod/2013Apr/0376.html>
> [2] http://lists.w3.org/Archives/**Public/public-lod/2013Apr/**0318.html<http://lists.w3.org/Archives/Public/public-lod/2013Apr/0318.html>
> [3] http://lists.w3.org/Archives/**Public/public-lod/2013Apr/**0346.html<http://lists.w3.org/Archives/Public/public-lod/2013Apr/0346.html>
> [4] http://lists.w3.org/Archives/**Public/public-lod/2013May/**0023.html<http://lists.w3.org/Archives/Public/public-lod/2013May/0023.html>
> [5] https://github.com/csarven/**linked-research<https://github.com/csarven/linked-research>
>
> -Sarven
>
>