SMGR-CRTS-CONTRACTION test fails. about kenzo

[gheber]

Francis' comment:

It is not an actual failure. The point is the following. The small chain complex of the reduction is very small, it is the chain complex with only one generator in degree 0, nothing else, constructed by Z-CHCM.

The maps f and g of the reduction should not be the zero maps, but the canonical injection Z -> C_\ast(BG \times_\tau G) for g and the canonical projection in the other direction for f. The chain complex C_\ast(BG \times_\tau G) has only C_0 = Z also. Replacing the correct f and g by the zero maps generates the errors observed. Consider this a positive test of the error tracking process!