As the holidays are rapidly approaching, we thought we would give you some real life examples of how great the arXiv API is to share with your family. We knew that you were dreading explaining to Grandma what Atom over HTTP is, and thanks to several intrepid developers you don’t have to! Now you can send her a link to this post and she can discover for herself all the cool things people are already doing with the API.

A Mediawiki Extension by Bill Flanagan

The first in the list came out only a couple of days after the API. OpenWetWare’s Bill Flanagan has extended Mediawiki to enable arXiv magic words. In case you are not familiar with Mediawiki, it is the software that runs Wikipedia, so making an extension to it is extremely interesting given Wikipedia’s popularity.

Magic words are a convenient shortcut when editing a wiki article. Say you want to reference a particular arXiv article in a wiki page. You could manually type in details of the article, or even create a link to an arXiv abstract page. Now with Bill’s extension it is much easier. All you have to do is type something like

ARXID 0711.3008

and Bill’s system will insert a link to more information for you. Of course behind the scenes Bill’s extension is calling the arXiv API which is why we are telling you (and Grandma) about it.

Bookends Reference Manager by Jonathan Ashwell

Jonathan Ashwell of Sonny Software has already integrated an arXiv search into his Bookends reference manager software. If you don’t know Bookends, you should because it can really save you lots of time when you are writing those papers and theses, especially if you are citing a lot of arXiv articles.

Bookends is a reference manager. Let’s say I want to write a paper (to post on the arXiv) and I need to cite several arXiv papers. Most people would typically type in citation information into a BibTex file and reference that file within their Latex manuscript. That means you have to go to arxiv.org, search for papers, and then manually type stuff. That sounds like a lot of work. With Bookends, you can search the arXiv within your reference manager itself. If you find a paper you like, simply import it into your reference library. You can even attach PDF’s to these reference library entries so you don’t have to worry about forgetting where you put that PDF. Then it is just a few clicks to generate a BibTex file containing the papers you want. For the record, Bookends can format citations in a variety of formats, and it can also search over other literature services like PubMed.

Nature Scintilla Search by Alf Eaton

From the Nature blog

Scintilla is an aggregator—of science weblogs, news stories and
publication databases—but it works in a slightly different way from the
existing online RSS readers that cover the whole internet. For a start,
the sources are manually selected, and only related to science, so there
shouldn't be any trouble with spam when searching for stories. Also
there's no 'unread items' count, so you don't have to feel like you have
lots of reading to catch up on. Browse the site, add sources to your
collection, and visit your 'Read' page on Scintilla whenever you're
looking for some juicy science stories to read.

Thanks to Alf, we can add arXiv to the list of publication databases that Scintilla works with. If you are a Scintilla user, this is great news, and if you aren’t, maybe you should be! For more information, visit the Scintilla home page.

And of course we love it when you talk about us. Thanks to the PhysMathCentral blog for keeping track of our progress. As always, comments are welcome, even from Grandma!

Links

• OpenWetWare’s Bill Flanagan
• OpenWetWare
• Mediawiki wiki engine
• Wikipedia
• Bookends Reference Manager Software
• PubMed
• Nature’s Alf Eaton
• Nature’s Scintilla Service
• PhysMathCentral Blog

This week has been real busy for the arXiv API team. We spent a lot of time introducing ourselves to the community, and it looks like there are more murmurs about us on the web (see Jacques Distler’s blog and Life on the Lattice). We love to hear that the community is excited, so keep on talking!

Amongst all the excitement, we also got a chance to make some improvements to the API. The API is now returning arXiv subject classification.

Every arXiv e-print has a primary subject classification, and can be cross-listed into secondary subject classifications. For an example, go to

http://arxiv.org/abs/0710.5276

and look for the ‘Subjects:’ line. You will see that this e-print falls into two categories: cond-mat.mes-hall and cond-mat.mtrl-sci. The primary category (in bold) for this e-print is cond-mat.mes-hall.

Now the API will return ALL arXiv categories as Atom elements. For example, these categories would be included as:

<category term="cond-mat.mes-hall" scheme="http://arxiv.org/schemas/atom"/>
<category term="cond-mat.mtrl-sci" scheme="http://arxiv.org/schemas/atom"/>

There is a new extension element in the xmlns:arxiv=”http://arxiv.org/schemas/atom” that repeats the primary category information, and is called primary_category. In this example, it looks like

<arxiv:primary_category xmlns:arxiv="http://arxiv.org/schemas/atom" term="cond-mat.mes-hall" scheme="http://arxiv.org/schemas/atom"/>

The documentation has been updated, and you can find more information in the user’s manual

In addition, all the code examples have been updated in the documentation, so you can start playing!

Links

• API User’s Manual
• Jacques Distler on the arXiv API
• Life on the Lattice on the arXiv API