A notion of transcendence degree for fields of formal power series?

531 Views Asked by At

There is an intuitive sense in which one would like to say that the field of formal Laurent series $k((z))$ over a field $k$ has "transcendence degree $1$". Of course, it doesn't have transcendence degree $1$ over $k$ in the usual sense (in fact I think the transcendence degree is countably infinite, by a cardinality argument). However, $k((z))$ is the completion of $k(z)$ for the $z$-adic absolute value, so it would be nice to say that it has transcendence degree one in some sense. Is there a way to make this precise, or is my intuition overly ambitious?

Remark: I think that it is true that $k((x,y)):=k((x))((y)) \cong k((z))$ as $k$-algebras, simply because both sides have the same (infinite) transcendence degree over $k$, in the usual sense. This doesn't bode well for my question, since one would certainly like to say that $k((x,y))$ has "formal" transcendence degree $2$ over $k$. However, there are natural topologies on both of these $k$-algebras (there is the $z$-adic topology on $k((z))$, and there are a few candidates on $k((x,y))$, for instance the topology for which two formal series are close if their difference is divisible by a high power of $x$ or $y$). If one takes these topologies into account, then I don't think they are isomorphic as topological $k$-algebras, so maybe my question can be rescued.