Homomorphisms of complexes via homologies (Q2583023): Difference between revisions

The main result of the paper gives a description of \(\roman{Hom}_{D^{b}}(T,S)\) as iterated pull-back of homomorphism spaces. More precisely, if \(R\) is a commutative ring, \(\Lambda\) is an \(R\)-algebra and \(T=(0\overset{\alpha^{0}}{} P^{1}\overset{\alpha^{1}}{}\cdots \overset{\alpha^{n-1}}{} P^{n} 0)\) and \(S=(0\overset{\beta_{0}}{} Q^{1}\overset{\beta^{1}}{}\cdots \overset{\beta^{n-1}}{} Q^{n} 0)\) are complexes of projective \(\Lambda\)-modules such that \(\roman{Hom}_{\Lambda}(\roman{Coker}(\alpha^{i}),\roman{Im}(\beta^{i}))= \roman{Hom}_{\Lambda}(\roman{Im}(\alpha^{i}),\roman{Coker}(\beta^{i}))= \roman{Hom}_{\Lambda}(\roman{Coker}(\alpha^{i}),Q^{i})= \roman{Ext}^{1}_{\Lambda}(\roman{Coker}(\alpha^{i}),H^{i}S)=0\) then \(\roman{Hom}_{D^{b}}(T^{i},S^{i})\) is the pull-back of the angle \[ \roman{Hom}_{D^{b}}(T^{i+1},S^{i+1})\to \roman{Hom}_{D^{b}}(\roman{Coker}(\alpha^{i+1}),H^{i+1}S[2])\leftarrow \roman{Hom}_{\Lambda}(H^{i+1}T,H^{i+1}S) \] for all \(i\in\{0,\dots,n-2\}\). Here \(T^{i}=(0\to \roman{Im}(\alpha^{i})\to P^{i+1}\overset{\alpha^{i+1}}{} \cdots \overset{\alpha_{n-1}}{} P^{n} 0)\) and \(S^{i}=(0\to \roman{Im}(\beta^{i})\to Q^{i+1}\overset{\beta^{i+1}}{} \cdots \overset{\beta_{n-1}}{} Q^{n} 0)\) and \(T^{0}=T\), \(S^{0}=S\). Moreover, the canonical homomorphisms \(\phi^{i}:\roman{Hom}_{D^{b}}(T^{i},S^{i})\to \roman{Hom}_{\Lambda}(H^{i+1}T,H^{i+1}S)\) are described. In the end of the paper the author gives suggestive examples which explain how to use the main theorem of the paper.