You have downloaded Google Refine 2.0 and are busy “reconciling” data.
Say you are one of the lucky ones and this is for your employer. 😉
Now you need to work on another project or even take some downtime.
So, how do you hand off maintenance of “reconciling” data?
A reconciliation on which your employer now relies.
You recognize the data to be reconciled but there are two problems:
- The raw data has implied properties you are using to reconcile the data. That means there isn’t anything for you to point anyone to as a basis for reconciliation. Just as well because:
- The rules for reconciling the data exist only in your head. So, the properties being implicit isn’t such an issue, the rules for handling it aren’t written down either.
Flatland identity as far as the eye can see.
If you had a defined set of properties (key/value pairs) as the basis for reconciliation, you could also say how to carry out the reconciliation.
And your data would be maintainable.
Best of luck with your downtime.
*****
PS: BTW, if you think documenting the names and locations of the data you are integrating counts as documentation, think again. What happens when new data comes along? Data your boss is going to expect to be integrated.
[…] Maintainability […]
Pingback by Maintainability, Auditability, eXtensibility – MAX Reconciliation « Another Word For It — November 17, 2010 @ 6:57 am