“Integrity is not a conditional word. It doesn’t blow in the wind or change with the weather.”
John D. MacDonald.
So you have found the holy grail of the OSS world and delivered the perfect OSS product. Hats off to you!
But you can’t stop there can you? Is the data in your OSS perfect? If it is, that’s holy grail number two.
But you’re still not finished are you? Holy grail number three is defining the process (or processes) that keep the data perfect.
Holy grails one and two are probably impossible:
- For HG#1 every user will have a different opinion of perfection that will evolve over time; and
- For HG#2 you’ll have legacy data, data that isn’t tracked (eg passive devices with no programmatic interface), near-realtime data with different polling windows, etc
To be honest, HG#3 is also impossible but with clever process designs that have data integrity feedback loops built in, you can at least ensure that daily activities are pushing data ever closer to the asymptote of perfection.