[Home] [Groups] - Message: [Prev in Group] [Next in Group]

nu.kanga.list.mud-dev

3137: Re: [MUD-Dev] Population container

[Full Header] [Plain Text]
From: Wout Mertens <Wout.Mertens@rug.ac.be>
Newsgroups: nu.kanga.list.mud-dev
Date: Thu, 3 Jul 1997 17:31:26 +0200 (MET DST)
References: [1]
Organization: Kanga.Nu
On Tue, 1 Jul 1997 clawrenc@cup.hp.com wrote:

>    at 07:57 PM, Wout Mertens <Wout.Mertens@rug.ac.be> said:
> 
> >Well, the only feasible solution to both problems seems to me: Allow
> >only a certain number of instances max. to run loose in the area
> >controlled by the population container.  It would involve a slight
> >redefinition: The population container is a virtual container, that
> >contains both the real and virtual ('real' meaning virtual to us, of
> >course. This is getting philosophical :)) instances of the
> >population, and the model is calculating the numbers. 
> 
> I prefer something like:
> 
>   If a mobile has travelled a minimum of X unique rooms, during which
> time he has had available a minumim of Y alternate routes (ie
> different exits from a given room), and if during that time he has not
> passed or encountered any objects which either queried any of his
> methods, or redirected his IO (ie outgoing descriptions, entrance/exit
> strings etc), then that object can be de-instantiated.

Very nice. As long as the object is not something special (like a major or
a queen or so, I think this is a good solution

> >Which leaves me to think that maybe it should be called Population
> >Model and not Container... Or how a clever idea turns into a standard
> >solution :)
> >Ah well, it's not as if I'm not used it on this list :)
> 
> You put up the targets and we'll do our best to shoot them down or
> paint them a different colour.  Heck you guys have sandblasted enough
> of my own already.

Yup, great list :)

Wout.