dhis2-users team mailing list archive

[Lars Helge Øverland <larshelge@xxxxxxxxx>]

Hi Robin,

its a good question and actually something that has been discussed before.
You highlight some of the pros and cons in your email.

The reason why we haven't done it so far is simply that we want to avoid
the complexity that comes with validity dates instead of routine time
periods and data aggregation. If this is used with care it would work well
but the system must take handle the quirky cases as well. Example: if
someone enters 10 data sets with completely arbitrary validity dates during
a year, then what should be the yearly average?

So far we have been relying on scripts to move data over to new time
periods. The typical use case is population data (like you say), where you
have censuses being done every 3-10 years. In this case one would enter the
data for a year (month is often too granular), then apply a SQL script to
copy the data to coming years, taking a growth rate into account. This
works pretty well.

Data with validity dates is also bordering to what we call "events", or
"surveys", which you can capture through the "tracker" component. Here,
events are capture for a timestamp, and you can capture any number of
events during a timeline.


regards,

Lars