Tom Coates of Plastic Bag (and now Yahoo!) has long been reliable for his occasional thoughtful long-form essays about the nature of the web as a medium. Back in February he posted one called Native to a web of data. Then a few weeks ago Christina Wodtke posted about it to the IAI list, summing up the points thusly:
- Look to add value to the Aggregate Web of data
- Build for normal users, developers and machines
- Start designing with data, not with pages
- Identify your first order objects and make them addressable
- Use readable, reliable and hackable URLs
- Correlate with external identifier schemes
- Build list views and batch manipulation interfaces
- Create parallel data services using standards
- Make your data as discoverable as possible
Then Margaret Hanley followed up, writing:
Tom and I worked together at the BBC on the project that brought our thoughts together on this.
We created PIPs (Programme Information Pages) that has now been rolled out to create pages on Radio 3 and Radio 4.
As a quite technical information architect form a data perspective, I enjoyed myself immensely, looking at the XML, identifying how to break source data to make it human readable/ sensible and the “highlight” getting 5 new fields into a BBC-wide database. But I also did user testing and research and worked with a fabulous designer on creating the pages.
What struck me, since I moved into a more traditional UX/IA role in an agency, is that there are very few of us who enjoy the level and granularity of data into interface or applications. My team look aghast at the thought that they would look at XML and identify how to create applications out of it.
I think this is the growth area of IA especially in the world of mash ups, but it does require a love of detail and the vision to see how it can be grown, merged and manipulated. It is the ultimate in collision between Big and Little IA.