Choice of memory layout for VectorOfArray

From the docs:

A[i] # Returns the ith array in the vector of arrays
A[j, i] # Returns the jth component in the ith array
A[j1, ..., jN, i] # Returns the (j1,...,jN) component of the ith array

which presents itself as a column-major matrix with the columns being the arrays from the vector.

What is the reasoning behind this memory layout? Can one use a tranposed memory layout, i.e. each component is represented by a column? What are the performance implications when trying to access A[j1, :] for max(i) >> max(j), i.e. timeseries data?

Perhaps the answers to those questions should be included in the docs.

Start by reading the VectorOfArray indexing documentation and the surrounding implementation to determine how its current memory layout is represented. Document the rationale, compare the proposed transposed layout, and explain the performance implications for time-series access; the issue names no specific file or test to run.