Been meditating lately on how best to unblock folks in my community who are trying to find information; and the same for their communities. The common cry heard is that data is hidden, or there’s too much data without metadata indicating how current or stale it is; or who owns it and where to follow-up. “Why isn’t data more discoverable?”
Here’s my general thinking about moving toward a better situation:
1.All data should be consumable. Expose the data via APIs or semantic queries. Apply authentication or other systems to protect privacy where it’s needed; but don’t apply it if it’s not needed. Goal is perfect awareness, i.e. all data is available to explore.
2.Let individuals or groups build apps to suit their data exploration needs; but avoid these needs getting hardwired into APIs or data structure. Every few years there’s talk percolating around single semantic models or taxonomies or single sourcing data repositories. Isn’t the internet already providing this? History is littered with new ‘single’ efforts; unfortunately these usually decay quickly if left to a few [or one] individual to maintain.
3.Social connections create bridges across disconnected data. I have immense faith in human beings connecting with each other, and sharing what they believe between each other to be the data that matters. Therefore let the data surface organically by enabling communities to form around data, then let community interaction transform it into more useful data. Back to data discovery; the community effect is generally very good at self-policing incomplete or inaccurate metadata, and retiring out-of-date or redundant data.
4.Apps built for desktop and mobile devices should be the same app. The context is physically, experientially different but the data and the app that presents it should be the same. Without giving too much credit to any one company for moving in this direction; consider the Facebook app and how it presents the same data across desktop and mobile devices. There are other examples in the print world; where books and magazines consistently publish the same data without redesigning [if you will] the app that presents the data.
5.Apps should cross intranet/extranet boundaries securely and seamlessly. Individual apps can determine rights and permissions for their users exploring data. Apps should therefore enable users to cross boundaries into intranet domains similar to how banking and retail [shopping] securely proxies this access for their customers. There’s likely some data exploration that is incomplete without this cross boundary access.
This post will be the beginning of a series examing all five points above in more detail.