bug with a[,1]:,1 ?

[tangentstorm]

in ngn:

  a:!10; a[,2]:,50; a
0 1 50 3 4 5 6 7 8 9

in oK:

   a:!10; a[,2]:,50; a
(0
 1
 ,50
 3
 4
 5
 6
 7
 8
 9)

[JohnEarnest]

ngn/k's behavior is also consistent with k3.

[razetime]

is this being worked on? I'd like to solve it.