Re: [s-d] RFC: Rule Tag

[Daniel Lepage <dpl33@xxxxxxxxxxx>]

[subject changed from "RFC: The Grid" to "RFC: Rule Tag" for accuracy]

On Nov 29, 2006, at 12:30 PM, shadowfirebird@xxxxxxxxx wrote:

>> I think it looks like a lot of fun, however I'm worried that mobility
>> will be hard. You allow players to move from rX to rY if and only if
>> rY defines a term that is used in rX; this creates a directed graph
>> on which the players can move. Glancing over the current rules, it
>> seems to me that everyone will eventually end up trapped in R1
>
> It's not as bad as that.  You can move from rX to rY if rX defines a
> term in rY +or+ rY defines a term in rX.  So all paths go both ways.
> And in any case you always have the option of dropping out of the game
> (moving to null) and then choosing a new starting RTL.

Oh, ok. I misread that.

Actually, I don't think I like letting players move to null. If the  
rules change and you get trapped, then you'd better find a way to  
make a new rule for you to travel to before somebody repeals your  
prison.

>> I don't know how hard it would be implement this, but what about
>> connecting the rules in an undirected graph where rX connects to rY
>> if they have at least three words of five or more letters in common?
>
> I think that would be too easy.  And not nearly as elegant: having
> navigation  depend on the spelling rather than the actual rules.

This is true. OTOH, it makes it a lot easier to have a wikibot crawl  
the rules and draw the pathways. But that might not be too important.

I'm also a bit worried that we'll have ambiguity regarding which rule  
"defines" a word. What if the full definition of a term is spread  
across multiple rules? I can't think of an example offhand of how  
this would happen, but I'm not willing to accept it as impossible  
either.

Both problems could be solved by adding an extra field to each rule  
that lists the objects defined by that rule and the objects mentioned  
by the rule. This would also make it easy to "turn off" certain words  
if we felt they were overly connective (for example, "player",  
"nweek", and "Game Object" will all connect to a vast majority of the  
rules). The field would still be determined by the content of the  
rule (we wouldn't add "blue frog" to the list, unless the rule  
actually mentions blue frogs).

A subcommittee/minister to keep those up-to-date would be helpful.

>> Or perhaps we could bring back Keywords, where each rule has a list
>> of keywords telling what it relates to (for example, r2 might have
>> the keywords "Proposals", "Voting", "Players"), and you can move to
>> rules that share keywords with your current rule.
>
> I thought of that, and it's doable.  We could have other ruletag-only
> sections for each rule, too.  "this rule deals with
> judgements....there is a blue frog here".  But I think we should leave
> that as an add-on.  I'd rather get the basics working without it.
> Besides, I like the 1-1 correspondance between the ruleset and the
> gameboard; the idea that to change the gameboard, you actually have to
> change the ruleset - at least for navigation....

Heh... I can see all sorts of random rules being made just to make  
tag pathways.

Also, a rule could influence the players on it:
{{
__Springboard__
Any player who occupies this rule at the beginning of an nweek may  
make three moves instead of one this nweek.

This rule mentions blue frogs.
}}

>> We could set up an even weirder system... For example, maybe when
>> somebody's score tops 100, the player with the *second highest* score
>> wins.
>
> Eurgh.  what would the winning strategy be?  A little too random  
> for my tastes.

In the case of your subgame, the winning strategy would be to get a  
lot of points, and wait for someone else to do the same, and then try  
to position yourself so that you're forced to give them 30 points. I  
don't think it would be too random.

Another variation would be to separate the end condition from the  
victory condition. Perhaps each player has two scores, one in normal  
points, and one in coolness points. As soon as a player has more than  
100 normal points, the player with the highest coolness wins.

-- 
Wonko


_______________________________________________
spoon-discuss mailing list
spoon-discuss@xxxxxxxxx
http://lists.ellipsis.cx/mailman/listinfo/spoon-discuss