bug#15117: 24.3.50; doc of `(forward|backward)-*': state return value

[Andreas Röhler]

Am 11.02.2014 12:16, schrieb Nicolas Richard:
> Andreas Röhler <andreas.roehler@easy-emacs.de> writes:
> > Am 11.02.2014 05:01, schrieb Stefan Monnier:
> > > Says you.  For me (and most definitions of side-effects I've ever seen),
> > > a side-effecting function is something like a function where the calls
> > > can't simply be replaced by their return value.
>
> > Can't see a logic link between a possible return value and the notion of 
> > side-effect.
> > It all depends from the purpose of the function.
>
> I guess this is a slight misunderstanding. The notion of side-effect is
> well defined in CS: « a function or expression is said to have a side
> effect if, in addition to returning a value, it also modifies some state
> or has an observable interaction with calling functions or the outside
> world. »
> (http://en.wikipedia.org/wiki/Side_effect_%28computer_science%29)

WP is often good, sometimes bad, sometimes just an agent of CIA, NSA and their 
satellites.
Saying in WP is no proof at all.

If you read "in addition to returning a value": functions must not return a 
value, but might have side-effects

> The notions of "mostly-unintended effect" (like so: (defun
> my-point-at-eol (progn (end-of-line) (point)))) or "multi-purpose
> functions by means of global variables" (e.g. the use of 'org-ts-what in
> org-at-timestamp-p -- which is kind of scary if I may say) are only
> special cases of side effects.

It's  also not about unintended side-effect, also also about intended.