Comment 1 for bug 1012377

Agreed, seeing the previous manual tests while one or more shell tests are running doesn't make sense. However, I'm not clear on the solution you propose.

You first suggest "tests should run in their own window", do you mean a popup window? If so, should a new popup window open and close for each shell test or should one popup window refresh itself as long as there are shell tests running?

You also suggest "refresh/restore active window", what should it contain?

While answering these questions, I'd appreciate if you could consider the two extreme use cases where you have few shell tests that take a lot of time or you have many shell tests that take little time. So, for a popup window that opens and closes, the user experience might just look like windows flashing by if the tests take little time.