Re: where do I think wrong

On 14-03-14 13:26, Roelof Wobben wrote:
>> Date: Fri, 14 Mar 2014 12:03:03 +0100
>> > From: [email protected]
>> > To: [email protected]
>> > Subject: Re: [SWIPL] where do I think wrong
>> > 
>> > On 14-03-14 07:54, Roelof Wobben wrote:
>>>> > >>     half_or_full_siblings(X, Y) :-
>>>>> > >> >         (   same_father(X, Y),
>>>>> > >> > 	    X \== Y
>>>>> > >> > 	;   same_mother(X, Y),
>>>>> > >> > 	    \+ same_father(X, Y)
>>>>> > >> > 	).
>>> > >  
>>> > > Oke, I think this is what im searching for.
>>> > > One rule for full and half siblings. 
>>> > >  
>>> > > But I do not understand why "is not equal"  is used.
>> > 
>> > Because you don't usually want to say that a person is his or her own
>> > sibling.
> No, that was is this one: X /== Y. 
> 
> Im talking about this one :  	
> ;   same_mother(X, Y),  <- have the same mother and
> \+ same_father(X, Y)  <- is not equal to the same father.
> ).
> 
> What if father is divorced and marry another female.
> Then the father is the same but the mother not.

That's what the first branch of the "or" is for, there you check for
same_father(X, Y).

In the second branch same_father(X, Y) is negated just so you don't get
the same solution twice, as Richard explained.