I have sort of an underlying philosophical theory that open-source projects that start out as attempts to clone a popular commercial package will ultimately fail. The reasoning goes like this: even if you achieve a good clone, the commercial project will move along and you will not be able to keep up. Your desire to catch a user-base that is resistant to change will actually encourage you to not keep up, as at first this looks like a kind of success, but then a few years down the road you will realize that you are terrible behind.

As an example, look at LibreOffice. It is widely marketed by OSS enthusiasts as "just like Office" and "just as good as Office," but it struggled to meet feature parity while remaining stable early on. Then, MS transitioned Office to the Ribbon UI. This was absolutely hated by a fairly large group of users, and it became a marketing point, at least in my circles, for LibreOffice that it had not made such a transition. This pulled in more users and looked like a success story. Years later, though, Office users have adapted to the Ribbon UI and discovered its advantages, and when you compare Office to LibreOffice, LibreOffice seems hopelessly out of date.

I think KDE is another example of this phenomenon as it was commonly seen as the best way to transition from Windows. But as a project it seriously struggled to retain interest after Microsoft continued to evolve their desktop environment in big ways, and now that I think they've successfully divorced themselves from any "clone of Windows" objectives they're still not really able to make up for the lost steam.

For just a last example, I would point to The GIMP, which has successfully navigated the transition from "a kind of clone of Photoshop" to "a kind of clone of a very old version of Photoshop."