User blog comment:DrCeasium/new hyperfactorial array notation/@comment-5150073-20130415141127/@comment-5529393-20130416182114

The actual definition of BEAF for pentational spaces and above is a bit up in the air, as Bowers was rather vague and did not define his arrays formally beyond dimensional arrays. But the answer to your question is yes, if Ikosarakt1 has his druthers. He's insisting that any definition of BEAF satisfies f(X) & b has f(b) non-1 entries.

Regarding the second part of your post, you are mistaken. For example, take

{b, p (2) 2}.

This is equal to

{b, b, b, ... b (1) b, b, b, ... b (1) ... (1) b, b, b, ..., b}

where there are p rows of p values of b. This is obviously not the same as {b, p (1) 2} = {b, b, b, ..., b}. Perhaps you should review the definition of the Extended Array function.