According to the New York Times, Mozilla will only fix about 20% of the blockers before Firefox 3.0 (AKA Minefield) is shipped to general users. Blockers are serious issues, not necessary bugs, that must be cleared before an app is released.
This means that 80% of the bugs will remain there, which translates to 700*0.8=560 bugs in the official version of Firefox 3.0.
Wow, Firefox 3.0 is surely very buggy and not usable by the time it is released to the public, is it not?
According to Reports of Firefox 3.0 overblown, most blockers are not serious problems. Some are just feature requests, minor enhancements or minor bugs that will appear on specific websites. They shouldn't affect the general usability of Firefox. The reason these issues got classified as blockers was because of the open source nature of the Firefox development. You see, everyone can contribute code and bug reports to an open source project, and so everyone can set the priorities of their requests. If you submit a bug report, of course you want it to be resolved fast and will therefore mark it as a first priority item.
This is the reason why Mozilla won't fix so many of the blockers.
From my experience, Minefield is already stable to used since Beta 1. I started using Minefield as my primary browser since the days of Alpha release-- and never turn back. Occasionally I got builds that contained serious regression bugs, but instead of switching to IE, I waited for the next build. It's hard to explain why or how Firefox is better than IE, I just feel that Firefox is way better than IE.
When you have built up a group of hardcore users who won't leave your software no matter what, even when it is sinking, you know that you have already succeeded in creating a brand.