The Content Mine website – how we create it. And the community can edit and contribute.

We are now about 6 weeks into The Content Mine project and have now released our website (http://www.contentmine.org). In the spirit of living a web-friendly life this is a living object which is planned to be:

• easy to update and maintain
• re-usable
• communal and collaborative.
• scalable

© Raimond Spekking / CC BY-SA-3.0 (via Wikimedia Commons)
To do that we have taken a novel approach to creating the site. We want the material to be easy to edit and create, with potentially lots of contributors. That’s not always easy if you have to have login access to the website.
The best software is often on collaborative FLOSS software sites. That’s because it’s had hundreds of years of knowledgeable users and developers. So I turned to Github and its wiki.  A wiki is an excellent tool to develop one’s thoughts as the structure evolves as our insight develops. So I started off with a list of the most important things that I thought we would need and put them on the first page of the Wiki (https://github.com/petermr/contentMine/wiki/ContentMining) which looks/ed like:

 
This is how you see it after an initial edit. It’s very functional, with lots of editing icons, etc. The blue phrases are links to other pages or external pages. I created about 100 pages on Sunday – some are stubs but most have text and links to other pages. And the value is that we are building up a structured resource. It’s a set of pages that can be re-used for tutorials, reference and, we hope, additions by volunteers.
However to make it more like a normal web page Mark MacGillivray and his Cottage Labs colleagues have created software for transferring Github content to a standard website. It can be automated so that, for example, we can update the website from the wiki every midnight. Here’s the same page:

 
(The picture is RNA from some of Ross Mounce’s Openly extracted phytotaxa scraping.). Mark’s done a great job in almost no time. That’s partly because CL are  very smart and partly because CL build re-usable code. And it’s easy to change the look-and feel.
Most people hate keeping websites up to date,  but I like wikis.  So I’ll be adding more pages which will help to explain content mining, and create re-usable resource.