antonio . dolcetta on Wed, 22 Nov 2006 08:59:38 -0700 (MST)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: [s-d] [s-b] Proposal: Conflict resolution


> ----- Original Message ----
> From: Andy Jones <shadowfirebird@xxxxxxxxx>
> To: spoon-business@xxxxxxxxx
> Sent: Wednesday, November 22, 2006 10:52:44 AM
> Subject: [s-b] Proposal: Conflict resolution
> 
> Proposal: "Conflict resoluton"
> 
> 1) Change rule 1-8 to add at the end:
> {{Every rule has an amendment date.  This is a date.  There can only
> be one amendment date per rule.}}
> 
> 2) Change all existing rules to add, as its amendment date, the date
> that the rule was last changed.
> 
> 3) Change rule 2-2; add the following at the end:
> {{Each rule created or amended by the proposal has its amendment date
> updated to the date the proposal passed.}}
> 
> 4) Add a new rule titled "conflict resolution":
> {{When two or more rules conflict, one rule "wins" and the rest are ignored.

What exactly is ignored in the "losing" rule, just the conflicting part or the whole rule ?
Usually in nomic this problem is handled by saying: "rule x takes precedence over rule y", this implies that any effects in rule y that are non-conflicting still happen.
What you are saying is basically "rule x exists and rule y does not count" which can lead to pretty unpredictable results IMHO. 

> 
> If one rule says it overrides the others, that rule wins; otherwise,
> if one rule is of a type that overrides the other rules' type(s), then
> that rule wins; otherwise, the rule with the oldest creation date
> wins.

You have not yet defined a type that rules can have (I suppose this goes together with your proposed immutable rule proposal)
Also you have not defined the creation date for a rule (if that is indeed a separate value from the amendment date)

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