User blog comment:KthulhuHimself/Terminal array notation./@comment-25601061-20151028175208

My proposal for definition of <0<(PP<#>)1> (described in similar way as KthuluHimself does):

So, we could have  <0<( <1> P<1>)>1> arrays,  <0<( /<1> P<1>)>1> arrays, heck, even  <0<( <1><1> P<1>(#))>1> arrays. In this matter, we could extend the P<#> operator to DTaN and SdTaN arrays, for example, have a  <0<(P<0<1>1>  {1}P<1>  )1> array, or even a  <0<(P<0<1>1>  {1}P<1>(1)  P<0<1>1>  {1}P<1>  )1> array. And as such, we could even extend it to ultradimensional arrays! And to note the superdimension, instead of using P<#>, we use a higher order notation: PP<#>!

I think that's it. Is this definition right?