The (geographic) gatherer currently has a timeout option that is measured relative to the latest expected granule. This is useful, but not for all situations. Consider:

Current situation

Consider a granule duration of 180 seconds, a timeliness of 5 minutes, for a certain area:

• [10:03] Granule covering 10:00 - 10:03 arrives. Gatherer expects 10:00-10:03, 10:03-10:06, 10:06-10:09, 10:09-10:12, 10:12-10:15, 10:15-10:18.
• [10:06] Granule covering 10:03 - 10:06 arrives.
• Nothing happens.
• [10:23] Timeout detected, gatherer sends message covering 10:00-10:06, with data that are 17-23 minutes old.

Desirable situation

With a „monitor for silence“ set to 7 minutes:

• [10:03] Granule covering 10:00 - 10:03 arrives. Gatherer expects 10:00-10:03, 10:03-10:06, 10:06-10:09, 10:09-10:12, 10:12-10:15, 10:15-10:18.
• [10:06] Granule covering 10:03 - 10:06 arrives.
• Nothing happens.
• [10:13] Silence detected, gatherer sends message covering 10:00-10:06, with data that are 7-13 minutes old.

The monitor for silence would bring the data to the users 10 minutes faster.

There is a tradeoff between fast and comprehensive, but that is up to the user to decide.