CIL 2007 – Mashups, remixing info & making data browsable – Karen Huffman and Derek Willis

Couldn’t get to a power or get my laptop out in this crowded session, so had to use treeware and now is the first chance to type this up. This session was Monday morning, so sorry for the delay.

Karen Huffman spoke of her experiences at National Geographic (NG). Web 2.0 is what we are already in, it’s where our users are and where our desktops are going. Adapt, adopt or die. We envision Web 2.0 in different ways, but are all still figuring it out and have different ways of applying the same solutions.

NG has RSS and podcasting hosted externally. Started with RSS feeds posted on a simple web page, then using Magpie RSS and php, they started streaming RSS to the homepage. Started with current content and recorded it as podcasts, then educated their users. Relaunched podcasts into iTunes and NG is now rated at No. 8. Have a staff news area which includes RSS feeds from NG blogs.

They use Newsgator for intranet news feeds instead of the external Bloglines. Newsgator can give mobile access to feeds which staff can access via their Blackberrys.

They are investigating Google gadgets to enhance their website. One gadget they are using on the their intranet is a Word Press plugin which shows a thumbnail image of the commenter on their blog comments. Simplified wiki page creation with a widget that gives them a “Create a new project” button. Also investigating Mind mapping software, including Gliffy and Mind Manager. They are also using Google Maps and Google Earth mashups in their Women Explorers wiki and BioBlitz projects as well as mapping out recipients of NG grants.

Lessons learned:

  • need the right people on board
  • work in the white spaces
  • collaboration
  • understand organic culture
  • prototype ideas, keep it simple
  • communicate
  • adapt if the strategy doesn’t match needs

Derek Willis from spoke on creating browsable data with Django.

Information gap includes:

  • what doesn’t make it into the news
  • the data you don’t use
  • what you can do about it

Can search it and searching is fun, but what happens when you don’t have a clear search term – are your users good searchers?

Django users a python web framework, takes data and puts it on the web. It is open source and automated as much as possible. More info and software available at Presently django is used to run – a database of crime reported in Chicago and several Washington Post projects including the congressional votes database, Faces of the fallen and Recipe Finder. It is mainly browsable, although it can be free text searched or browsed/searched by category.

Need the Python script language on a web server, a database such as MySql. It runs on all operating systems and is free. Advantages include control over your data, using your data is easier, has built in admin interface, plus it supports syndication, generic views, authorisation, forms, file uploads and is used for about 15 smaller newspapers around the US.

Things to do to make it happen:

  • become or find a geek
  • scavenge for hardware
  • think about your data

Give your users the value in the information you already have.