When a "Kudo" is giving within the Tempo Newsfeed, where is that data

Currently there is no out of the box method of aggregating the data of Kudo's. I could definitely submit an enhancement request on your behalf for this feature.

Just for your question not to remain fully answered: kudos are entries in the [dbo].[tp_feed_entry] table with a NULL feed_id and of a specific type. As of Appian 7.2 the type for kudos is 8 but obviously since there's no public API nor out of the box to query Kudos it wouldn't be safe to rely on this design to run a simple query since it can be subject to change in future versions.

I meant "*just for your question not to remain partially answered"

Thanks for the responses...and yes, you read my mind Eduardo. Thanks for pointing me in the right direction with the tables the data is stored in. We may create a query against that table in the meantime, but watch for any added system funtionality in future Appian releases. Thanks!

One last question though, I do not see any dates associated to the feed entries. Is there another associated table that stores these dates? Also, how would I retrieve all of the associated comments to the initial "kudos"? Thanks!

Actually the goal of my answer was for just question not to remain unanswered but we definitely don't recommend accessing the primary data source for several reasons, some examples are: if you expose it you won't be able to avoid people updating it and possibly creating inconsistencies, also since there is no public API yet you can't guarantee the structure will remain the same which is not maintainable, I would consider exposing it too risky and definitely not a best practice.

For your last question, which I'd hate to leave unanswered as well, the answer is: each row in the tp_feed_entry table has a created_ts which is a "long" used to represent the date when the entry was created. For example, if my row has a value of 1381895363259 in created_ts and I run this simple line in Java

System.out.println(new java.sql.Timestamp(1381895363259L));

I get: 2013-10-16 14:49:23.259

This is just an example but there are already functions in MySQL, SQL Server, etc that convert a long into a timestamp. Again keep in mind I am not recommending exposing/accessing the primary data source for what I've explained above but this would the answer to your last question.

Thank you Eduardo. Your input is much appreciated. We will look at a different solution, which makes me sad because we would have loved to use the system Kudos functionality.