article of the future from Elsevier

[matei]

Today I received an Elsevier link to their vision of the "article of
the future". An example can be found at
http://www.articleofthefuture.com/S0022314X08001856/

Have a look through, and note the various annotation and referencing
systems that makes the work more accessible to the reader.

However the vision falls short in addressing the reproducible research
aspect that is currently lacking in research papers.

There are however positive features that I see in their examples,
which can be exploited to permit reproducible research: the inclusion
of interactive figures, the ability to download the figure material in
various formats, or access online datasets for analysis. See for
example,
http://www.articleofthefuture.com/S0020025510002756/#datarepo-item1

These features have been implemented by James in various ways, and
what I have seen from him looked more promising than what Elsevier
envisions.

James, when would you have a simple bare bones example and source code
we can bite our teeth on to prepare a simple pdf document with source
code listings and interactive analysis tools. Like the one you showed
me for the corner problem some time ago?

Matei

[James Quirk]

Matei,

On Fri, 17 Jun 2011, matei wrote:

> Today I received an Elsevier link to their vision of the "article of
> the future". An example can be found at
> http://www.articleofthefuture.com/S0022314X08001856/
>
> Have a look through, and note the various annotation and referencing
> systems that makes the work more accessible to the reader.
>
> However the vision falls short in addressing the reproducible research
> aspect that is currently lacking in research papers.

As Churchill might say: "reproducible research is not the end. It is not
even the beginning of the end. But it is, perhaps, the end of the
beginning." By that I mean one should not get carried away as the RR
movement seems with mechanical reproducibilty. For it says
absolutely nothing about numerical accuracy and should not
be confused with the reproducibility of a physical experiment.

>
> There are however positive features that I see in their examples,
> which can be exploited to permit reproducible research: the inclusion
> of interactive figures, the ability to download the figure material in
> various formats, or access online datasets for analysis. See for
> example,
> http://www.articleofthefuture.com/S0020025510002756/#datarepo-item1
>
> These features have been implemented by James in various ways, and
> what I have seen from him looked more promising than what Elsevier
> envisions.

Thanks, but there are storm clouds ahead. Adobe have abandoned
/RichMedia annotations in Linux Adobe Reader, which sadly undermines
the PDF machinery I've developed. This was done purely on economic
grounds.

>
> James, when would you have a simple bare bones example and source code
> we can bite our teeth on to prepare a simple pdf document with source
> code listings and interactive analysis tools. Like the one you showed
> me for the corner problem some time ago?

The source code for the climategate example is embedded in the PDF.
To access it type:

amrpdf -unpack all climategate.pdf

This will produce a directory:

climategate.pdfWorkDir

and inside that you'll see a file:

AMRGI1.pdf

which you can also unpack. It contains the barebone script
you requested. You will not, however, be able to run the
script as it uses library routines that are only in my
developer's version. But that should not prevent you from
viewing the source with amrgi. The bit where it builds
a document is not too onerous for someone who already
uses LaTeX.

You might also consider dissecting the VKI notes:

amrcp vki/lecture.1
amrcp vki/lecture.2

which can be whittled down to:

Latex2eHead {
amrpdf = yes
}
fold::latex { body of document
fold::amrita { get figures
amrcp -run -ex Chp2/my.script -rtn fig1,fig2 -> grid,schlieren
}
&latex::fig([width=200pt]{*grid})
&latex::fig([width=200pt]{*schlieren})
}
LatexTail
Latex

where amrcp is discussed in the notes.

James

>
> Matei
>
>