This post is for those who would like to negate / support the importance of providing test case while raising a defect.
It might be given in summary or in any other apt field of defect data base.
I support it as I believe that there should be a basis for raising a defect and test case id is the basis for any defect.
I like having a field in the defect tracking system called test number. My testers would fill it out if they know an existing test covers a particular defect. It would not be required on submission because defects do not always come from QA and do not always originate from test cases or have an existing test case. In the heat of a project it would be unlikely that team members would have time to update every test case.
At the end of the project I gather up all the defects that do not have numbers and assign them to team members to add a test or verify an existing test covers the defect.
Some customers do not have the money to buy test case management software and this is a good manual process. I would keep the test number in a unique field so that it can be easily searched in the system. It would be too painful to search for the value within a summary.
Testing is not an art (that is unless your definition of art includes breaking other "art").
The only basis for raising a defect you should need is that the software does not match the spec, or generates an error, or otherwise misbehaves. Test cases are only a way for us to be more methodical in finding these defects.
Including the test case in the bug report is useful for tracking purposes and to assure your tests will catch it if it comes back or pops up in another area, but to me it doesn't lend any more weight to whether something is or is not a bug.