I'm attempting to wrap my head around spectral sequences, so constructed a really basic example to apply the definitions and go through the motions. My filtered chain complexes are:

$F_2C_*: 0 \rightarrow \mathbb{Z} \xrightarrow{i_1} \mathbb{Z_1}\oplus \mathbb{Z_2} \xrightarrow{p_2} \mathbb{Z} \rightarrow 0 $

$F_1C_*: 0 \rightarrow 0 \rightarrow \mathbb{Z}_2 \xrightarrow{p_2} \mathbb{Z} \rightarrow 0 $

$F_0C_*:$ The zero complex.

Here $i_1$ and $p_2$ are the inclusion and projection maps, and the sub-indices on the $\mathbb{Z}s$ are just for labeling purposes.

With $Z^r_{p,q}$ defined as $\{x \in F_pC_q : \partial x \in F_{p-r}C_{q-1}\}/F_{p-1}C_q$, the differential is meant to restrict to well-defined maps $d^r:Z^r_{p,q} \rightarrow Z^r_{p-r,q-1}$.

But in my example the map $Z^1_{2,2} \rightarrow Z^1_{1,1}$ does not appear to be well-defined:

$Z^1_{2,2} := \{x \in \mathbb{Z}\oplus \mathbb{Z} : p_2 x \in \mathbb{Z} \}/\mathbb{Z}_2 \cong \mathbb{Z}_1$,

$Z^1_{1,1} := \mathbb{Z}$.

However $[(a,b)]$ and $[(a,b')]$ in $Z^1_{2,2}$ get sent to $b$ and $b'$ in $\mathbb{Z}$ respectively, even though $[(a,b)] = [(a,b')]$.

I cannot see where I've made a mistake, but I'm pretty sure something's not right. Where have I gone wrong?

Given a map $A \xrightarrow{f} C$ with $B \subset A$ and $D \subset C$, in order for the quotient map $A/B \xrightarrow{\bar f} C/D$ to be defined we must have $f(B) \subset D$.

But it this case we have a map $(Z_1 \oplus Z_2) / Z_2 \xrightarrow{\bar p_2} Z/\{0\}$, but $p_2(Z_2) = Z_2$ which is not contained in $\{0\}!$


$Z_{p,q}^r$ isn’t a subquotient just a subobject. $$Z_{p,q}^r := \{x \in F^pC_q: dx \in F^{p-r}C_{q-1}\}$$ check eg wikipedia.


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.