[Freeciv-Dev] Re: (PR#2223) cma: server knowledge of island != client knowledge

[Christian Knoke <chrisk@xxxxxxxx>]

On Thu, Jan 09, 2003 at 03:44:26AM -0800, Per I. Mathisen via RT wrote:
> 
> On Thu, 9 Jan 2003, ChrisK@xxxxxxxx via RT wrote:
> > > Please note that we have to same situation with the effects. There may
> > > be an effect which affects all players of an island. Or doesn't such
> > > effect exists.
> >
> > I don't think so. Wonders are player-related.
> 
> The _current_ wonders are, yes, but it will be possible to define
> island-only effects once gen effects code is in. Also, isn't our civ2
> ruleset wrong on this point? I seem to remember that some wonders were
> island-only.

Yes, some are island-related (as J.S.Bach's), but the question was:
restricted to 1 island but effective for all players on that island.
This won't happen. And even if it will (some time), I think the effect
should be based on the server's calculation, and not on the client's.

> > But I also think, if you have
> > mananged to send a caravan to a city, you are allowed to know, whether it's
> > the same island or not. And if you've managed to build Bach's, you are
> > allowed to know which cities are on that island and effected by Bach's.
> 
> Why? That doesn't make any sense to me.

If I have 2 cities, with undiscoverd tiles between, I can guess whether
it's actually the same island or not. In most cases, I will be damned sure.
When I send a caravan to the other city, I will know, but this information
is not big, because the uncertainty was not big before. This information
is not revealed by a trick (like a hacked client), but by distinct player
action.

So I would KISS and let the server decide.

Christian

-- 
Christian Knoke     * * *      http://www.enter.de/~c.knoke/
* * * * * * * * *  Ceterum censeo Microsoft esse dividendum.