Advantages of using _argument states

["Nikolai Weibull" <now@xxxxxxxx>]

What are the advantages and disadvantages of using _argument states
over writing functions that generate the matches, i.e,

 '*-chmod[change mode of file]:mode:->modes'

versus

'*-chmod[change mode of file]:mode:_modes'

(or perhaps __modes to not clutter the Type-completion namespace)?

As far as I can tell, the only difference (and an obvious one at that)
is that states result in a big case-statement but you avoid extra
functions, whereas writing a function adds an extra function, but
saves us from the big case-statement.

I've gone over the documentation a bunch of times and I've been
looking through completion definitions without seeing any point to
using states over functions.

Am I correct in my assessment that they're basically two ways of doing
the same thing and the difference between them is the one I wrote
about above?

Also, writing a function makes it easier to turn that into a Type
completion later on, if need be.

(By the way, I'm writing a completion definition for mkisofs and growisofs.)

Thanks!

 nikolai