I’m a little puzzled about how to fill vectors into matrix elements… Consider the two attempts below: the first is unsuccessful, the second is successful.

```
julia> # Unsuccessful strategy
julia> a11 = [1,2];
julia> a12 = [1,-2];
julia> a21 = [1,2,1];
julia> a22 = [1,2,1];
julia> Matrix{Vector}([a11 a12; a21 a22])
ERROR: MethodError: Cannot `convert` an object of type
Int64 to an object of type
Array{T,1} where T
Closest candidates are:
convert(::Type{T}, ::AbstractArray) where T<:Array at array.jl:533
convert(::Type{T}, ::T) where T<:AbstractArray at abstractarray.jl:14
convert(::Type{T}, ::LinearAlgebra.Factorization) where T<:AbstractArray at D:\buildbot\worker\package_win64\build\usr\sh
are\julia\stdlib\v1.4\LinearAlgebra\src\factorization.jl:55
...
Stacktrace:
[1] setindex!(::Array{Array{T,1} where T,2}, ::Int64, ::Int64) at .\array.jl:825
[2] copyto!(::Array{Array{T,1} where T,2}, ::Array{Int64,2}) at .\multidimensional.jl:962
[3] Array{Array{T,1} where T,2}(::Array{Int64,2}) at .\array.jl:541
[4] top-level scope at REPL[34]:1
julia> # Successful strategy
julia> A = Matrix{Vector}(undef,2,2)
2×2 Array{Array{T,1} where T,2}:
#undef #undef
#undef #undef
julia> A[1,1] = a11;
julia> A[1,2] = a12;
julia> A[2,1] = a21;
julia> A[2,2] = a22;
julia> A
2×2 Array{Array{T,1} where T,2}:
[1, 2] [1, -2]
[1, 2, 1] [1, 2, 1]
```

Question: is there a way to fill the matrix with vector elements using the first notation? [The first notation is more convenient as an argument in function calls.]