Re: Metatables cannot have metatables?

[Sir Pogsalot <sir.pogsalot@...>]

On Sat, Jan 4, 2014 at 2:24 PM, Sean Conner <sean@conman.org> wrote:

It was thus said that the Great Sir Pogsalot once stated:

>
> The point I was trying to make that while reading the docs is always a good
> idea, a language that is consistent with its behaviors is self-explanatory
> -- and from the perspective of the user, cleaner (imo). :-)

  Given the Maxwell's Equations of Software [1] one can start writing
software to calculate pi to a million digits.  Of course, Maxwell's
Equations of Software (translating it to a more modern dialect of Lisp is
left as an exercise for the reader) don't mention anything at all about
numbers, but that's okay, because what we have is a very consistent
language, and thus, building up arthmetic is trivial [2].

  Although it's not very fast.  And it does take an inordinate amount of
memory.  But it *is* consistent!

  -spc (Even Haskel had to slip in non-consistent functions [3] in order
        to be useful ... )

[1]     http://www.michaelnielsen.org/ddi/lisp-as-the-maxwells-equations-of-software/

[2]     http://copperthoughts.com/p/set-theory-and-lisp/

[3]     Non-pure---i.e. a function that has side effects.

I don't think implying that it would take a lot of resources to effect this change is a fair argument -- after all you can already recursively __index tables.  The worst that happens is a stack overflow (fault of the user/scripter).  Conceptually, be consistent and let people shoot their feet :p