if rules are set up for a process like this one, following these rules is a major benefit in itself - even if you personally don't agree with these rules. not only does it guarantee a standard quality; it also shows if the rules work or not. ignoring the rules won't make them better.

as for the current rules and the bugtracker issue:
i think the current QA rules are reasonable. you might be tempted to argue if the lack of a valid bug tracker really should be a blocker.... yes, i think it should be. either the project does have a tracker; then it's so easy for the author to include it that he ha no excuse at all not to do it. or the project really doesn't have a tracker - well, then it mustn't be released in extras. you cannot just release software without letting ppl know how to report bugs.

so i don't really see the benefit if dropping this requirement. i also don't see why testers should kind of "silently ignore" the fact that it's missing, hoping that a tracker will be added in the next release or the one afterwards. if the author doesn't do it when he first gets his package in testing, he most likely never will.

there's a very clear documentation that says which criteria have to be met for a package to be promoted to extras. if these criteria aren't met, authors shouldn't try to push their work into -testing in the first place. and testerst shouldn't vote those pakages up based on nothing else than "cool application, works great!". that's not enough.