Checklist for Technological Innovation in Data Science
Lately I’ve been thinking about how one decides to adopt new technology or change the way one works. There’s a lot of push for Data Scientists to keep up with the latest and greatest innovation in IT or analysis (e.g. change to git
rather than staying with svn
) but there’s rarely any objective analysis or thought put into whether or not a Data Scientist should invest the time and energy in doing this. Therefore I decided to put together a list of principles that I hold and use a a yard stick when presented with such situations. These are presented below.
-
The new tool or method should be cheaper than the one it replaces.
-
It should be at least as fast as the tool or method it replaces.
-
It should do work or produce work that is clearly and demonstrably ‘better’ than the one it replaces.
-
It should be repairable and maintainable by people inside the organisation, provided that they are provided with the necessary tools.
-
It should not replace or disrupt anything good that already exists, and this includes people and community relationships.