[Freeciv-Dev] Re: Open patches? (PR#720)

[Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx>]

On Fri, Oct 12, 2001 at 05:02:27PM +0200, Davide Pagnin wrote:
> >On 2001-10-11 11:41:02, Gregory Berkolaiko wrote:
> >> Comment on patch:
> >> The model is a little bit complicated but I prefer one variable to
> >> Raimar's two.  How about providing smooth transition from size 1 to
> >> fulltradesize by removing *3/2 altogether?
> >
> >Well... That's the exact same as having Raimars approach and not
> >touching the lower bound variable... :) If you had to modify two
> >variables to get any useful result that would be a problem but that's
> >not the case here...
> >
> >I like the idea and favor Raimars solution... Flexibility is nice... :)
> >Defaults should match current behaviour though...
> >
> >/Daniel
> 
> Raimar solution is more flexible, but put another 2 server variables in.
> I think that server variables are too much and if you do a show command
> you need a 80 lines x-terminal to see every variable...
> 
> Indeed, if we chase flexibility, a better solution can be to implement
> 2 new government variables and put them in governments.ruleset.
> (Perhaps government.mintradesize and government.fulltradesize)
> With this approach we lose the possibility to change this in a easy
> way within the server, and this implies a less flexibility, I know,
> but in return of this we have 2 less server variables and
> furthermore, we can set different mintradesize and fulltradesize
> for different form of government. 
> I think that adjusting this variable depending on government type, 
> is an excellent idea, don't you think?

AFAIK the main reason for server variables are that they are settable
on public servers. But I agree that government specific limit may be
nice. I don't know if these are needed however.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 1 + 1 = 3, for large values of 1