I plan to give a talk on Steenrod algebra in a student seminar. but there are some questions that I didn't find an answer to, and it seems to me that I'm missing something:

  1. if the algebra of cohomology operations $\mathcal{A}\left(E\right)$ is isomorphic to the cohomology algebra $E^*\left(E\right)$, why it isn't (graded) commutative?

  2. the right and left units (source & target homomorphism) on the dual $E_*\left(E\right)$ are induced by the morphisms of spectra $\mathbb{S}\wedge E\to E\wedge E$ and $E\wedge\mathbb{S}\to E\wedge E$. why they are usually distict? and why they coincide in the classical cases ($H\mathbb{F}_p$)?

and another related question:

how by "correcting" the non-commutativity of cup product (in spaces or chain complexes) the stability is automaticly "corrected"?

  • 3
    $\begingroup$ The multiplication on $E^{\ast}(E)$ is not the cup product, it's composition. $\endgroup$ – Qiaochu Yuan Aug 25 '14 at 17:14

For part 2: The two "units" are very nearly always distinct. You should regard it as an accident that they coincide for $H\mathbb{F}_p$.

In fact, if $k$ is a field which is not a prime field, then the two units are distinct. In this case, $$ \pi_0 (Hk\wedge Hk) \approx k\otimes_{\mathbb{Z}} k, $$ and in homotopy the units are the two evident maps of $k$ into the factors of the tensor product. If $k$ is not $\mathbb{F}_p$ or $\mathbb{Q}$, then $k\otimes k\not\approx k$, so the units are different.

| cite | improve this answer | |
  • 2
    $\begingroup$ I think it is worth while to note that even in the case of $E=H\mathbb{F}_p$, the two maps of spectra $E\rightarrow E\wedge E$ are different, although they induce the same map in homotopy. $\endgroup$ – user43326 Aug 26 '14 at 15:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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