[thisAintNoGame]

That depends on field. In most areas of computer science, conferences are more important than journals (journals generally just publish extended conference papers).

[SeanLuke]

I am not aware of any CS field where conferences are in general more important than journals. Though there are a few conferences whose rigor approaches journal publication.

I think the standard ordering is: books, chapters, journal articles, strongly refereed conferences, lightly refereed workshops, unrefereed works, dissertation. And numbered! Good grief.