About

Full Text available: http://www.russet.org.uk/blog/2012/04/three-steps-to-heaven/

Three Steps to Heaven

We argue that to make semantic publishing possible we need to:

make life better for the machine to
make life better for the author to
make life better for the reader

To begin at the beginning

The birth of kblog (http://knowledgeblog.org)

Taken from: http://en.wikipedia.org/wiki/File:La_Palmyre_041-crop.jpg

The problem with publishing

images/800px-Clock_in_Kings_Cross.jpg Taken from: http://en.wikipedia.org/wiki/File:Clock_in_Kings_Cross.jpg

Often time-consuming
Over a long time

The process

images/500px-LaTeX_cover.svg.png Taken from: http://commons.wikimedia.org/wiki/File:LaTeX_cover.svg

The process

images/500px-Adobe_PDF_Icon.svg.png Taken from: http://en.wikipedia.org/wiki/File:Adobe_PDF_Icon.svg

The process

images/500px-Adobe_PDF_Icon.svg.png

Again!

The process

images/500px-Microsoft_Word_Icon.svg.png Taken from: http://en.wikipedia.org/wiki/File:Microsoft_Word_Icon.svg

The process

images/500px-XML.svg.png Taken from: http://en.wikipedia.org/wiki/File:XML.svg

The process

images/HTML5_Logo_512.png Taken from: http://www.w3c.org/html/logo

The process

images/500px-Adobe_PDF_Icon.svg.png

Again, Again!

The process

images/wordpress-blue-xl.png

Web First
RSS
Trackbacks
Stats
Gravatars

Existing Workflow

Fitting in with peoples workflows
Fitting in with peoples tools
- Word/Email
- Word/Dropbox
- LaTeX/Versioning
- Asciidoc/Dropbox
The wordpress editor is lacking

Adding Semantics

Adding semantics is hard
The three steps to heaven limit us
Going to describe two exemplars

Maths (author)

The author adds [latex]e=mc^2[/latex] to their document
Which is rendered as:

e = m c^{2}

This works in any authoring environment
The author adds "here is maths" semantics

Maths (reader)

The reader gets a nicely rendered equation

Maths (reader)

Which is NOT image based

Maths (reader)

And has some additional features

Maths (reader)

Access the tex

Maths (reader)

Or transformed to MathML

Architecture

$images/architecture-mathjax.jpg$

Why TeX?

There is a W3C standard

e = m c^{2}

<math>
   <mrow>
     <mi>e</mi>
     <mo>=</mo>
     <mrow>
       <mi>m</mi>
       <msup>
         <mi>c</mi>
         <mn>2</mn>
       </msup>
     </mrow>
  </mrow>
</math>

Why [Shortcodes]?

XML is painful
Most of the tools escape it anyway
Short codes are simple and reasonably well supported
We support other syntaxes also

Why client-side rendering

MathJax did not need developing
It looks nice
And has added functionality

What have we achieved?

Authors can write TeX.
Readers get a nice view
Machines get a markup saying "maths is here"
- Can tell that markup is being used
- Markup can be converted to MathML
- Authors have (probably) already checked this works

Maths

What we have achieved adds limited semantics
But is useful to reader, author and machine

References

Academics love in text-citations (Lord, 2012)
Will describe two tools, kcite and greycite

References

Lord, P (2012) Academics love Citations. J.Unsupport.Assert

References (author)

Authors again insert shortcodes
[cite]10.100/100.1[/cite]

images/kblog-with-edam-with-more-references.png

References (reader)

Readers can see citation or direct link
Readers can change citation style

images/kblog-with-edam-with-numeric-listing.png

References (kcite architecture)

The Problem

Inserting links is painful
Fortunately, we can use the bibliographic metadata to help
Will demonstrate this after a brief interlude

Greycite

Originally, kcite could not support URLs
We needed a source for metadata
Now we have http://greycite.knowledgeblog.org
Developed by Lindsay Marshall, Computing Science, Newcastle

Greycite

This is greycite

Greycite

Add a URL

Greycite

URL results

Greycite

Scrolling to more detail

Greycite

Provenance (coins)

Greycite

Provenance (OGP)

Greycite

And other (hosted) blogs

Greycite

Although somewhat limited

Greycite

And elsewhere

Greycite

Sadly, some websites have no semantics that we can find

Greycite

Preservation

Greycite

Finally

Greycite

We can retrieve as JSON (or bibtex)

Citing Links

How does adding links work?

Citing Links

We want to cite this page

Citing Links

In an article, I am writing

Citing Links

First we take the URL

Citing Links

Query Greycite

images/emacs-with-edam-url-and-mxgreycite.png

Citing Links

Get back bibtex

Citing Links

So, we search

images/emacs-and-post-ref-with-regexp.png

Citing Links

And select

images/emacs-and-post-ref-with-reftex-dialog.png

Citing Links

And insert

images/emacs-and-post-ref-with-kurl-inserted.png

Citing Links

And publish

Citing with Emacs

Most of the tools are easy to connect

(defadvice reftex-format-citation (around phil-asciidoc-around activate)
   (if phil-reftex-citation-override
      (setq ad-return-value (phil-reftex-format-citation entry format))
     ad-do-it))

(defun phil-reftex-format-citation( entry format )
    (let ((doi (reftex-get-bib-field "doi" entry)))
    (format "pass:[[cite source='doi'\\]%s[/cite\\]]" doi)))

Or with CSL

Citation Style Language, supported by Zotero, Mendeley and others

<citation>
   <layout prefix="[cite]" suffix="[/cite]"
     delimiter="[/cite] [cite]">
      <text variable="DOI"/>
   </layout>
</citation>

What have we achieved?

Reader gets a nicely laid out bibliography
Machine gets a primary identifier
Author gets some reasonable tools

What have we achieved?

We have made both the primary ID and the metadata useful
We have made authors want them to be right
Since using kcite I have
- Complained about metadata to crossref and datacite
  - Both now supply JSON directly, with the same interface
- Got several crossref metadata records fixed
- Discovered three broken DOIs in one special issue
  - One linked from my website; 404 is not enough
- Fixed author metadata on my blog
- Removed broken characters from bio-ontologies titles
- Picked up http://dropbox.org ref when posting this paper
- It is only after we wrote a tool, that I fixed my metadata

What have we achieved?

In short, the total metadata in the world
is slightly better than it was before

Summary

We need to use metadata to improve everyones life
A balanced approach is necessary
A little semantics is a good thing

Acknowledgements

Robert Stevens, Manchester (kblog)
Lindsay Marshall, Newcastle (greycite)
Simon Cockell, Newcastle (kcite)
Dan Swan, (ex-)Newcastle (array express — see paper)
Georgina Moulton, Manchester (kblog content)
http://knowledgeblog.org
http://bio-ontologies.knowledgeblog.org
http://greycite.knowledgeblog.org