4
$\begingroup$

The first non-trivial example of a uniform algebra which comes to mind is the disc algebra $A(\mathbb{D})$. In a similar manner one can define its relatives $P(U)$ and $R(U)$, where $U$ is any region on the complex plane and $P$, $R$ stand for the closures, of respectively polynomials and rational functions on $U$.

What if we go beyond the case when spectrum is contained in the complex plane? Rudin proved, that if $K$ is scattered, then there are no non-trivial uniform subalgebras of $C(K)$.

Are there any classical or canonical examples of uniform algebras with non-metrizable spectra or is the theory of uniform algebras just a daughter of complex analysis?

Forgive me if this question seems to be to vague.

Addendum: A uniform algebra is a closed subalgebra of a commutative C*-algebra $C(K)$ which contains constant functions and separates points in $K$.

$\endgroup$
3
  • $\begingroup$ It might help if you include the definition of a uniform algebra. Just click the "edit" button to add it. $\endgroup$
    – MTS
    Mar 13, 2012 at 20:31
  • $\begingroup$ Contra MTS, I think that people who might be interested in this question, or able to answer it, know full well what a uniform algebra is. No harm done, in any case. $\endgroup$
    – Yemon Choi
    Mar 13, 2012 at 22:00
  • $\begingroup$ Yemon, sure. I mean that all of them are $C(\Omega)$s. $\endgroup$ Mar 13, 2012 at 22:14

1 Answer 1

1
$\begingroup$

Here are some scattered thoughts: I'm not sure if they really answer what you seem to be driving at.

Complex analysis still gives you examples with nasty maximal ideal spaces, e.g. $H^\infty(\Omega)$. Even when $\Omega$ is the open unit disc, the spectrum is non-metrizable and not at all straightforward to understand (there ought to be some discussion in Gamelin's book).

$L^\infty$ has non-metrizable spectrum, although its structure is in some sense not as mysterious as that of $H^\infty$.

Note that every Banach space $E$ embeds as a closed, complemented subspace of a uniform algebra. Namely, take the canonical map from $E$ into $C(B)$, where $B$ is the closed unit ball of $E^*$ equipped with the weak$*$-topology, and let A be the closed subalgebra generated by the image of $E$. (This is a theorem of Milne; I learned of it from an article of Gamelin and Kislyakov, of which a preprint version can be found online.)

$\endgroup$
1
  • $\begingroup$ I think you meant "every Banach space E embeds as a closed COMPLEMENTED subspace of a uniform algebra." $\endgroup$ Mar 13, 2012 at 23:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.