Peter Eisentraut wrote:
>
> Tom Lane writes:
>
> > (a) a bug *tracking* system is not the same as a bug *reporting*
> > system. A tracking system will be useless if it gets cluttered
> > with non-bug reports, duplicate entries, etc. There must be a human
> > filter controlling what gets entered into the system.
>
> Letting any user submit bug reports directly into any such system is
> certainly not going to work, we'd have "query does not use index" 5 times
> a day. I consider the current *reporting* procedure pretty good; web forms
> are overrated in my mind.
>
> What I had in mind was more a databased incarnation of the TODO list. I
> mean, who are we kidding, we are writing a database and maintain the list
> of problems in flat text. The TODO list has already moved to the
> TODO.detail extension, but we could take it a bit further.
>
I maintain my todo items for my projects in a postgres database. But
there are a lot of issues to consider there too:
- a table of projects (or topics)
- a table of todo items with synopsis, full description, ...
- a table of versions (item is planned to be solved in version, x.x.x,
actually solved in y.y.y)
- a table of developers
- assign table (projects -> developers, items -> developers)
- change type: bug,doc,rfe (request for enhancement),idc (internal
design change), ...
- change state (accepted, evaluated, fixed, rejected, incomplete,
committed, ...
- severity or priority of each item, project
- search functionality
Regards
Wim
