[Freeciv-Dev] Re: Patch: Use original city id at client (Fixes some bugs)

[Thue <thue@xxxxxxx>]

After some consideration, I agree the orig_id way is preferable.

-Thue

On Thursday 21 December 2000 03:34, Marko Lindqvist wrote:
> On Tue, 21 Nov 2000, Thue wrote:
> > Yes, I hadn't read the patch properly. However, wouldn't it be simpler to
> > do things by using the private map in the server as I outlined than
> > introducing the orig_id? It seems everything can be solved that way. (ok,
> > the refounded city thing migth be a reason to remember the original city
> > ID)
> >
> > Though I am not completely decided I tend towards prefering to use the
> > private map for simplicity.
>
>  There is list of advantages I can see of each approach (over other).
>
>
>  private map:
>
>  - No client can tell whether city is destroyed and then refounded unless
>    area is unfogged at some point in between. (cheating clients)
>  - Simplicity
>
>
>  orig_id:
>
>    First four have any meaning only if clients are modified to somehow to
>  show if city is same as city previosly seen at this place. Of course
>  server can send all the following information to clients when
>  necessary, but figuring out when player has right to know
>  would be rather complicated (and memory hungry). It's easier to just let
>  player remember what he knows.
>
>  - If city once had a wonder, you can be sure it still has it.
>    Most players think this way anyway.
>  - Original owner (city.original) of the city never changes. Player
>    knows which cities grant partisans when conquered.
>  - Times diplomats have stolen technologies (city.steal) always rises,
>    never drops.
>  - Handling of other similar (future) variables.
>
>  - Compatibility with old clients. Only change required at client
>    side is removing of one LOG_ERROR message (Older clients might print it
>    every now and then)
>  - Patch exists already.
>
>
>  Well, I'm biased, but at least I left more abstract points out :)
>
>
>  So, do I continue working with my version or are you going to implement
> your own? (I'm too lazy to implement second solution to same problems)
>
>
>  Caz
>
> --