Confluence databases is a great feature but ...
I have only worked with it on a very small scale and limited, but there are already a number of very annoying hickups. Seems to me to be more of a beta version to be honest. Here are the first / biggest / most important problems that I encountered, and for the time being I have limited myself to the most important basic functionalities
I promote the use of confluence databases in our confluence environment but still experience quite a few issues
Nice workaround
Just hope that the atlassian team comes with a solution.
A small refinement to this: if I expand the new entry, I can even add the title first. Although the entry disappears from the view, the expanded view of the entry stays open and I can continue filling in the rest of the fields.
Seconding the stuff about adding new rows when searching, filtering or sorting!
All really good points and experienced them also. Many users complain to me that when they add rows they just disappear. I then tell them to scroll to the bottom. Often, we find multiple empty rows at the bottom as people tried repeatedly to add a row and not realising they were created off screen.
Suggestion to Confluence: Just add the new row below or above the current position in the database. The user can refresh the database if they want so to show it ordered correctly again. Or something else, but whatever is the least interruptive to workflow. You don't want to be jumping around all over.
That first point really tripped me up when I was doing some testing. It took me a bit to figure out why it wasn't generating an empty row after I'd filtered the database. Of course, it was creating a new row, but if I'm using a filtered view on a page because those entries are related to the page content, a user would logically expect to be able to add new rows related to that content right there.