On Sat, 20 Oct 2018, Ørjan Johansen wrote:
> On Thu, 18 Oct 2018, Kerim Aydin wrote:
>
> > This is very weird phrasing to me. You can backdate ratification, so
> > possibly
> > better phrasing: "The following list is Ratified as being an accurate list
> > of
> > coin holdings for 18 Oct 2018". Maybe no big deal tho.
>
> I think I've quibbled in the past that the ratification rule is written such
> that it _isn't_ clear what happens when specifically constructing documents
> speaking about the past in order to ratify them.
>
> This is because what's a "minimal change" to the game state to make the
> document true is calculated for the time of document publishing, _not_ for the
> time of the referenced date, and so the more game state changes have happened
> between the times, the more the actual effect may be different from what you
> intuitively wanted to happen.
How is this different than self-ratification - when it self-ratifies a week
after publication, it ratifies the past condition as being true as of that
past date, correct? If not, we're really messed up.