Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2003:
[Freeciv-Dev] Re: Union vs Research Treaty
Home

[Freeciv-Dev] Re: Union vs Research Treaty

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: "Per I. Mathisen" <per@xxxxxxxxxxx>
Cc: freeciv-dev@xxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Union vs Research Treaty
From: Raimar Falke <rf13@xxxxxxxxxxxxxxxxx>
Date: Wed, 18 Jun 2003 16:52:27 +0200

On Wed, Jun 18, 2003 at 09:33:08AM +0000, Per I. Mathisen wrote:
> I and Raimar discussed this on IRC, and came to the conclusion at that
> time that there was another idea that might just as good: Adding a new
> treaty, like Shared Vision, that implements a Research Treaty in the
> following way:
> 
> In tech dialog: Show two panels, one for player and one for RT. There
> should be a slider to control how many research points you devote to your
> own research and how much to RT. You can change the RT's research item and
> goal, but not to a tech that any other RT member is researching or already
> has. You get info about all other RT members' techs, current research and
> research goal. The cost of the tech being research is the average of all
> members'. (If there are tech restrictions, then all restrictions apply.)
> 
> An obvious problem is that treaties are not transitive, thus you can get
> multiple RTs at the same time, maing it necessary it have multiple RT
> panels. The interaction between the various research panels would be
> difficult.
> 
> (One possible solution is this: The RT is transitive, meaning you get RT
> with all those who have an RT with the player you get an RT with. In order
> to agree to an RT, you must first have 'contact' or embassy with all those
> players and they must at least be Neutral towards you. However, in
> previous discussions, the idea of transitive treaties did not receive a
> warm welcome.)

> Having slept on this, I found some additional problems: Using default
> rules, you usually want to concentrate on one research path in any case
> (Republic, Steam), thus splitting of research points won't be used or will
> be used only be newbie players to their disadvantage (further increasing
> divide between good and bad players along the fault line of good players
> intentionally not using the full feature set of the game).

The splitting can be used for cheating on your RT members. Instead of
adding all your research to the RT pool you use 5% for your own
research and aim at secondary techs to get better navel units or
happiness wonders. This will be impossible if you see the total
research points of the other members and their contribution towards to
RT pool.

> The interaction
> between the two research panels is going to be messy. Since the two
> targets and goals must be mutually exclusive for all players in the RT, it
> will be very easy to disrupt the RT setup unintentionally unless you
> remove both RT's researched tech and RT's goal tech from the player's
> panel, in which case this will easily lead to confusion. Instant effects
> provided by the RT researched tech have nowhere to be applied. Popup of
> the research dialog must flip to correct panel. This all requires quite a
> bit of client changes.
> 
> Finally, I will not have time to implement it, while the Union patch is
> already mostly complete. So I believe that while an interesting idea, I do
> not believe it would be worth the effort or additional complexity.

The reason for this different idea is that the Union is in essence a
separate research entity. With a pool in which the members contribute
and the cost for the research. However the way in which the Union
implements this idea is not very obvious. Neither for the player nor
in the code changes. So I would like to make this explicit in the case
that two players want to put bulbs into a shared pool and get both get
the benefit. I also think that this shouldn't be limited to team
members. I think that RT in this case is similar to the pooling you
see in the industry now ("AMD and IBM cooperate on development on 65nm
technology").

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "Premature optimization is the root of all evil."
    -- D. E. Knuth in "Structured Programming with go to Statements"



[Prev in Thread] Current Thread [Next in Thread]