From Newsgroup: comp.lang.prolog

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

On Monday, 6 November 2023 at 01:42:59 UTC+1, Mild Shock wrote:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

Go fuck yourself, spammer.

*Spammer Alert*

Julio
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

I got the puzzle from a discussion here,
which was about Prolog versus Theorem Proving:

The lion and the unicorn met PROLOG
Bruce D. Ramsey, 1986 - Free Access https://dl.acm.org/doi/10.1145/382278.382395

This gives immediately the following database:

yesterday(monday , sunday ).
yesterday(tuesday , monday ).
yesterday(wednesday, tuesday ).
yesterday(thursday , wednesday).
yesterday(friday , thursday ).
yesterday(saturday , friday ).
yesterday(sunday , saturday ).

lies(lion, monday ).
lies(lion, tuesday ).
lies(lion, wednesday).

lies(unicorn, thursday).
lies(unicorn, friday ).
lies(unicorn, saturday).

Now what?
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Picking up an idea by @jamesnvc , abstracting a pattern
behind lying_day_lie_or_nonlying_day_truth/3 into something
with more meta calls:

contrary(S, T) :- S, \+ T.
contrary(S, T) :- \+ S, T.

solve(D) :- yesterday(D, Y),
contrary(lies(lion, D), lies(lion, Y)),
contrary(lies(unicorn, D), lies(unicorn, Y)).

Here some benchmark:

/* SWI-Prolog 9.1.17 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 45,999,999 inferences, 2.719 CPU in 2.712 seconds (100% CPU, 16919540 Lips) true.

/* Dogelog Player 1.1.3, Java */
?- time((between(1,1000000,_), solve(_), fail; true)).
% Zeit 9282 ms, GC 0 ms, Lips 10019403, Uhr 06.11.2023 16:34
true.

/* GNU Prolog 1.5.0 */
?- between(1,1000000,_), solve(_), fail; true.
(10782 ms) yes

Quite amazing the speed of SWI-Prolog, here I can only beat GNU Prolog
with my new Dogelog Player for Java. Mostlikely an argument in favor of inlining
negation as failure somehow, as SWI-Prolog does. Only lets do it correctly!

Mild Shock schrieb am Montag, 6. November 2023 um 16:43:13 UTC+1:

I got the puzzle from a discussion here,
which was about Prolog versus Theorem Proving:

The lion and the unicorn met PROLOG
Bruce D. Ramsey, 1986 - Free Access https://dl.acm.org/doi/10.1145/382278.382395

This gives immediately the following database:

yesterday(monday , sunday ).
yesterday(tuesday , monday ).
yesterday(wednesday, tuesday ).
yesterday(thursday , wednesday).
yesterday(friday , thursday ).
yesterday(saturday , friday ).
yesterday(sunday , saturday ).

lies(lion, monday ).
lies(lion, tuesday ).
lies(lion, wednesday).

lies(unicorn, thursday).
lies(unicorn, friday ).
lies(unicorn, saturday).

Now what?

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Its also a test case where my sys_trans_allowed/1 is totally useless. But
I guess this is a problem of my sys_trans_allowed/1, maybe we can make an exception for \+ A where A is a variable and arrive at something useful?

If A is a variable, and if we would replace it by (A->fail; true) this would
be compiled with call(A), and thus give a correct translation of negation as failure again. Its a similar special case like call/1 has the special case that

at runtime call(A) with A variable throws an exception to avoid call/1
wrapping at infinitum and that the Prolog interpreter runs into nirvana.
So the revised sys_trans_allowed/1 would read as follows:

/* Version 2.0 */
sys_trans_allowed(V) :- var(V), !.
sys_trans_allowed(A) :- sys_trans_allowed2(A).

sys_trans_allowed2(V) :- var(V), !, fail.
sys_trans_allowed2((A,B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2((A;B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2((A->B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2(A) :- callable(A).

But I guess a little more is needed. Namely make (A->fail; true) cheap.

Mild Shock schrieb am Montag, 6. November 2023 um 16:47:40 UTC+1:

Picking up an idea by @jamesnvc , abstracting a pattern
behind lying_day_lie_or_nonlying_day_truth/3 into something
with more meta calls:

contrary(S, T) :- S, \+ T.
contrary(S, T) :- \+ S, T.

solve(D) :- yesterday(D, Y),
contrary(lies(lion, D), lies(lion, Y)),
contrary(lies(unicorn, D), lies(unicorn, Y)).

Here some benchmark:

/* SWI-Prolog 9.1.17 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 45,999,999 inferences, 2.719 CPU in 2.712 seconds (100% CPU, 16919540 Lips)
true.

/* Dogelog Player 1.1.3, Java */
?- time((between(1,1000000,_), solve(_), fail; true)).
% Zeit 9282 ms, GC 0 ms, Lips 10019403, Uhr 06.11.2023 16:34
true.

/* GNU Prolog 1.5.0 */
?- between(1,1000000,_), solve(_), fail; true.
(10782 ms) yes

Quite amazing the speed of SWI-Prolog, here I can only beat GNU Prolog
with my new Dogelog Player for Java. Mostlikely an argument in favor of inlining
negation as failure somehow, as SWI-Prolog does. Only lets do it correctly! Mild Shock schrieb am Montag, 6. November 2023 um 16:43:13 UTC+1:

I got the puzzle from a discussion here,
which was about Prolog versus Theorem Proving:

The lion and the unicorn met PROLOG
Bruce D. Ramsey, 1986 - Free Access https://dl.acm.org/doi/10.1145/382278.382395

This gives immediately the following database:

yesterday(monday , sunday ).
yesterday(tuesday , monday ).
yesterday(wednesday, tuesday ).
yesterday(thursday , wednesday).
yesterday(friday , thursday ).
yesterday(saturday , friday ).
yesterday(sunday , saturday ).

lies(lion, monday ).
lies(lion, tuesday ).
lies(lion, wednesday).

lies(unicorn, thursday).
lies(unicorn, friday ).
lies(unicorn, saturday).

Now what?

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Ok I was comparing apples with oranges when I looked
at SWI-Prolog and other Prolog systems. Since SWI-Prolog
creates a multi-argument index, which explains the speed:

?- jiti_list(lies).
Predicate Indexed Buckets Speedup Flags ============================================================================ user:lies/2 2 8 6.0
true.

Can I manually switch off this index for a fairer comparison?
Whats even the logic behind choosing this index. Thats
a quite an amazing choice sind the call of lies/2 is ground.

Mild Shock schrieb am Montag, 6. November 2023 um 17:03:16 UTC+1:

Its also a test case where my sys_trans_allowed/1 is totally useless. But
I guess this is a problem of my sys_trans_allowed/1, maybe we can make an exception for \+ A where A is a variable and arrive at something useful?

If A is a variable, and if we would replace it by (A->fail; true) this would be compiled with call(A), and thus give a correct translation of negation as failure again. Its a similar special case like call/1 has the special case that

at runtime call(A) with A variable throws an exception to avoid call/1 wrapping at infinitum and that the Prolog interpreter runs into nirvana.
So the revised sys_trans_allowed/1 would read as follows:

/* Version 2.0 */
sys_trans_allowed(V) :- var(V), !.
sys_trans_allowed(A) :- sys_trans_allowed2(A).

sys_trans_allowed2(V) :- var(V), !, fail.
sys_trans_allowed2((A,B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2((A;B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2((A->B)) :- sys_trans_allowed2(A), sys_trans_allowed2(B). sys_trans_allowed2(A) :- callable(A).

But I guess a little more is needed. Namely make (A->fail; true) cheap.
Mild Shock schrieb am Montag, 6. November 2023 um 16:47:40 UTC+1:

Picking up an idea by @jamesnvc , abstracting a pattern
behind lying_day_lie_or_nonlying_day_truth/3 into something
with more meta calls:

contrary(S, T) :- S, \+ T.
contrary(S, T) :- \+ S, T.

solve(D) :- yesterday(D, Y),
contrary(lies(lion, D), lies(lion, Y)),
contrary(lies(unicorn, D), lies(unicorn, Y)).

Here some benchmark:

/* SWI-Prolog 9.1.17 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 45,999,999 inferences, 2.719 CPU in 2.712 seconds (100% CPU, 16919540 Lips)
true.

/* Dogelog Player 1.1.3, Java */
?- time((between(1,1000000,_), solve(_), fail; true)).
% Zeit 9282 ms, GC 0 ms, Lips 10019403, Uhr 06.11.2023 16:34
true.

/* GNU Prolog 1.5.0 */
?- between(1,1000000,_), solve(_), fail; true.
(10782 ms) yes

Quite amazing the speed of SWI-Prolog, here I can only beat GNU Prolog with my new Dogelog Player for Java. Mostlikely an argument in favor of inlining
negation as failure somehow, as SWI-Prolog does. Only lets do it correctly!
Mild Shock schrieb am Montag, 6. November 2023 um 16:43:13 UTC+1:

I got the puzzle from a discussion here,
which was about Prolog versus Theorem Proving:

The lion and the unicorn met PROLOG
Bruce D. Ramsey, 1986 - Free Access https://dl.acm.org/doi/10.1145/382278.382395

This gives immediately the following database:

yesterday(monday , sunday ).
yesterday(tuesday , monday ).
yesterday(wednesday, tuesday ).
yesterday(thursday , wednesday).
yesterday(friday , thursday ).
yesterday(saturday , friday ).
yesterday(sunday , saturday ).

lies(lion, monday ).
lies(lion, tuesday ).
lies(lion, wednesday).

lies(unicorn, thursday).
lies(unicorn, friday ).
lies(unicorn, saturday).

Now what?

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Just for the fun of it, also benchmarking,
was expecting same behaviour as contrary2:
contrary5(S, T) :- \+ (forall(S,T), forall(T,S)).¨
Not so good because forall/2 isn’t handled as a “control construct”, whereas (\+)/1 is handled as a “control construct” and inlined.
Maybe this is the reason:
/* contrary5 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 65,999,999 inferences, 2.812 CPU in 2.813 seconds (100% CPU, 23466666 Lips) true.
So contrary is the XOR, i.e. ~(A ↔ B) == ~((A → B) & (B → A)).
What can a constraint programming system do to the Lion and Unicorn
example. Are there not special SAT solvers who are good in XOR?
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Could ChatGPT, the automatic programming Robot,
help develop a solution with a SAT solver for XOR in mind?
My idea currently, use 3 variables X,Y,Z for weekday:
X,Y,Z weekday
0,0,0 monday
0,0,1 tuesday
0,1,0 wednesday
0,1,1 thursday
1,0,0 friday
1,0,1 saturday
1,1,0 sunday
Or maybe somebody, a human, did it already this way.
Mild Shock schrieb am Samstag, 11. November 2023 um 18:33:50 UTC+1:

Just for the fun of it, also benchmarking,
was expecting same behaviour as contrary2:

contrary5(S, T) :- \+ (forall(S,T), forall(T,S)).¨

Not so good because forall/2 isn’t handled as a “control construct”, whereas (\+)/1 is handled as a “control construct” and inlined.
Maybe this is the reason:

/* contrary5 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 65,999,999 inferences, 2.812 CPU in 2.813 seconds (100% CPU, 23466666 Lips)
true.

So contrary is the XOR, i.e. ~(A ↔ B) == ~((A → B) & (B → A)).
What can a constraint programming system do to the Lion and Unicorn
example. Are there not special SAT solvers who are good in XOR?

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

What the heck, one can even write ultrafast
parsers with SAT (some bit encoding)?

Long live SMT. Long live meta-programming

This article describes why and how I wrote absolut; a Rust
crate for automating a common pattern of constructing
SIMD byte-wise lookup tables. I learned about this technique
while studying the Parsing Gigabytes of JSON per Second
paper by Geoff Langdale & Daniel Lemire.
https://fuzzypixelz.com/blog/absolut/
Mild Shock schrieb am Samstag, 11. November 2023 um 19:40:21 UTC+1:

Could ChatGPT, the automatic programming Robot,
help develop a solution with a SAT solver for XOR in mind?

My idea currently, use 3 variables X,Y,Z for weekday:

X,Y,Z weekday
0,0,0 monday
0,0,1 tuesday
0,1,0 wednesday
0,1,1 thursday
1,0,0 friday
1,0,1 saturday
1,1,0 sunday

Or maybe somebody, a human, did it already this way.
Mild Shock schrieb am Samstag, 11. November 2023 um 18:33:50 UTC+1:

Just for the fun of it, also benchmarking,
was expecting same behaviour as contrary2:

contrary5(S, T) :- \+ (forall(S,T), forall(T,S)).¨

Not so good because forall/2 isn’t handled as a “control construct”, whereas (\+)/1 is handled as a “control construct” and inlined.
Maybe this is the reason:

/* contrary5 */
?- time((between(1,1000000,_), solve(_), fail; true)).
% 65,999,999 inferences, 2.812 CPU in 2.813 seconds (100% CPU, 23466666 Lips)
true.

So contrary is the XOR, i.e. ~(A ↔ B) == ~((A → B) & (B → A)).
What can a constraint programming system do to the Lion and Unicorn example. Are there not special SAT solvers who are good in XOR?

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

Ok, I was using a SAT solver in a little unorthodox way.
Used it as a compiler for a formula, so that I got:

yesterday(0,0,0,1,1,0).
yesterday(0,0,1,0,0,0).
yesterday(0,1,0,0,0,1).
yesterday(0,1,1,0,1,0).
yesterday(1,0,0,0,1,1).
yesterday(1,0,1,1,0,0).
yesterday(1,1,0,1,0,1).

two_liars(T1,T2,T3,Y1,Y2,Y3) :-
T1=\=Y1,
T1=:=(T1/\Y2) xor (T1/\Y3) xor (Y2/\Y3),
Y1=:=(T2/\T3) xor (T2/\Y1) xor (T3/\Y1),
T1=\=(T1/\T2) xor (T1/\T3) xor T2 xor (T2/\T3) xor T3,
Y1=\=(Y1/\Y2) xor (Y1/\Y3) xor Y2 xor (Y2/\Y3) xor Y3.

solve_for(T1,T2,T3) :-
yesterday(T1,T2,T3,Y1,Y2,Y3),
two_liars(T1,T2,T3,Y1,Y2,Y3).

Works fine:

/* SWI-Prolog 9.1.17, optimise=false */
?- time(solve_for(T1,T2,T3)).
% 16 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
T1 = 1,
T2 = T3, T3 = 0 ;
% 10 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
false.
--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

But I guess I always miss launching SWI-Prolog with optimise=true,
and the version I am using has default optimise=false.
Because I get these varying results:

/* SWI-Prolog 9.1.17, optimise=false */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% 22,999,999 inferences, 2.016 CPU in 2.016 seconds
(100% CPU, 11410852 Lips)

/* SWI-Prolog 9.1.17, optimise=true */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% 9,999,999 inferences, 1.203 CPU in 1.199 seconds
(100% CPU, 8311687 Lips)
Whereas in Dogelog Player I have now practically always
optimise=true, which then gives me the following:

/* Dogelog Player 1.1.4, JDK 8 */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% Zeit 1892 ms, GC 0 ms, Lips 13213582, Uhr 13.11.2023 23:54

Mild Shock schrieb am Dienstag, 14. November 2023 um 00:11:37 UTC+1:

Ok, I was using a SAT solver in a little unorthodox way.
Used it as a compiler for a formula, so that I got:

yesterday(0,0,0,1,1,0).
yesterday(0,0,1,0,0,0).
yesterday(0,1,0,0,0,1).
yesterday(0,1,1,0,1,0).
yesterday(1,0,0,0,1,1).
yesterday(1,0,1,1,0,0).
yesterday(1,1,0,1,0,1).

two_liars(T1,T2,T3,Y1,Y2,Y3) :-
T1=\=Y1,
T1=:=(T1/\Y2) xor (T1/\Y3) xor (Y2/\Y3),
Y1=:=(T2/\T3) xor (T2/\Y1) xor (T3/\Y1),
T1=\=(T1/\T2) xor (T1/\T3) xor T2 xor (T2/\T3) xor T3,
Y1=\=(Y1/\Y2) xor (Y1/\Y3) xor Y2 xor (Y2/\Y3) xor Y3.

solve_for(T1,T2,T3) :-
yesterday(T1,T2,T3,Y1,Y2,Y3),
two_liars(T1,T2,T3,Y1,Y2,Y3).

Works fine:

/* SWI-Prolog 9.1.17, optimise=false */
?- time(solve_for(T1,T2,T3)).
% 16 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
T1 = 1,
T2 = T3, T3 = 0 ;
% 10 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
false.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog

So how to use a SAT solver as a compiler. First model the problem
with library(clpb) as Boolean formulas.

lying_day(A, B, C, D, M) :- M = (
~A* ~B* ~C*D+
~A* ~B*C* ~D+
~A* ~B*C*D+
A*B* ~C* ~D+
A*B* ~C*D+
A*B*C* ~D).

two_liars(T1, T2, T3, Y1, Y2, Y3, ((M1#M2)*(M3#M4))) :-
lying_day(0, T1, T2, T3, M1),
lying_day(0, Y1, Y2, Y3, M2),
lying_day(1, T1, T2, T3, M3),
lying_day(1, Y1, Y2, Y3, M4).

Then look at the normalform:

?- two_liars(T1,T2,T3,_,_,_,M), sat(M).
sat(1#T2*T3#T2*_A#T3*_A#_A),
sat(T1=:=T1*_B#T1*_C#_B*_C),
sat(T1=\=_A),
sat(T1=\=T1*T2#T1*T3#T2#T2*T3#T3),
sat(_A=\=_A*_B#_A*_C#_B#_B*_C#_C).

Take the normalform, reorder the equations a little bit, fewer
variables first, and rewrite them into ISO core standard bitwise
operations, and then take profit.

Mild Shock schrieb am Dienstag, 14. November 2023 um 00:12:16 UTC+1:

But I guess I always miss launching SWI-Prolog with optimise=true,
and the version I am using has default optimise=false.
Because I get these varying results:
/* SWI-Prolog 9.1.17, optimise=false */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% 22,999,999 inferences, 2.016 CPU in 2.016 seconds
(100% CPU, 11410852 Lips)

/* SWI-Prolog 9.1.17, optimise=true */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% 9,999,999 inferences, 1.203 CPU in 1.199 seconds
(100% CPU, 8311687 Lips)
Whereas in Dogelog Player I have now practically always
optimise=true, which then gives me the following:

/* Dogelog Player 1.1.4, JDK 8 */
?- time((between(1, 1000000, _), solve_for(T1,T2,T3), fail; true)).
% Zeit 1892 ms, GC 0 ms, Lips 13213582, Uhr 13.11.2023 23:54
Mild Shock schrieb am Dienstag, 14. November 2023 um 00:11:37 UTC+1:

Ok, I was using a SAT solver in a little unorthodox way.
Used it as a compiler for a formula, so that I got:

yesterday(0,0,0,1,1,0).
yesterday(0,0,1,0,0,0).
yesterday(0,1,0,0,0,1).
yesterday(0,1,1,0,1,0).
yesterday(1,0,0,0,1,1).
yesterday(1,0,1,1,0,0).
yesterday(1,1,0,1,0,1).

two_liars(T1,T2,T3,Y1,Y2,Y3) :-
T1=\=Y1,
T1=:=(T1/\Y2) xor (T1/\Y3) xor (Y2/\Y3),
Y1=:=(T2/\T3) xor (T2/\Y1) xor (T3/\Y1),
T1=\=(T1/\T2) xor (T1/\T3) xor T2 xor (T2/\T3) xor T3,
Y1=\=(Y1/\Y2) xor (Y1/\Y3) xor Y2 xor (Y2/\Y3) xor Y3.

solve_for(T1,T2,T3) :-
yesterday(T1,T2,T3,Y1,Y2,Y3),
two_liars(T1,T2,T3,Y1,Y2,Y3).

Works fine:

/* SWI-Prolog 9.1.17, optimise=false */
?- time(solve_for(T1,T2,T3)).
% 16 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
T1 = 1,
T2 = T3, T3 = 0 ;
% 10 inferences, 0.000 CPU in 0.000 seconds (0% CPU, Infinite Lips)
false.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


For those that thought the Lion and Unicorn riddle has something to do
with natural language understanding and ChatGPT, I have a new problem.
SAT+CAS for Problems in Quantum Foundations
A SAT Solver + Computer Algebra Attack on
the Minimum Kochen–Specker Problem
Brian Li, Curtis Bright, Vijay Ganesh - Oct 1, 2023 https://nikolajbjorner.github.io/ShonanArtOfSAT/program.html
Who can show the Kochen-Specker Paradox via SWI-Prolog?
Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


Ha Ha, board fired Sam Altman of OpenAI.
What now? What will become of ChatGPT?

Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


Comedy on the horizon and lots of tech penny stocks to buy!
Before you criticize the board at OpenAI, walk a mile in their shoes:
VCs Congratulating Themselves 👏👏👏 https://twitter.com/VCBrags/status/1725976543585784022
LoL
Mild Shock schrieb am Sonntag, 19. November 2023 um 03:57:40 UTC+1:

Ha Ha, board fired Sam Altman of OpenAI.
What now? What will become of ChatGPT?
Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


Fuck around find out. Guess who is a board reporting
to? Well the stock holders, and they are reinstantiating
the board every year. So if stock holders want Sam
Altman, stock holders will have Sam Altman, they simply
send the board into /dev/null and create a new one.
BTW: Interesting information about the board,
didn't think they have relation to Sam Bankman-Fired,
and hence its rewarding to listen to TiffanyFong:
s Sam Altman BACK as CEO of OpenAI?
Board Resigning & “Effective Altruism” Ties [UPDATE] https://www.youtube.com/watch?v=LXDWy5-bc-c
Yeah, get your finger off EA, see FTX fiasko.
Mild Shock schrieb am Sonntag, 19. November 2023 um 04:08:02 UTC+1:

Comedy on the horizon and lots of tech penny stocks to buy!
Before you criticize the board at OpenAI, walk a mile in their shoes:

VCs Congratulating Themselves 👏👏👏 https://twitter.com/VCBrags/status/1725976543585784022

LoL
Mild Shock schrieb am Sonntag, 19. November 2023 um 03:57:40 UTC+1:

Ha Ha, board fired Sam Altman of OpenAI.
What now? What will become of ChatGPT?
Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to
deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


Some actors are messing real hard with OpenAI.
My prediction, the only goal is to dissolve OpenAI.
Take the nomination of Emmett Shear. It shows
how unprepared and desperate the board is.
Mostlikely he will not accept the nomination, since
we wont hear anything of OpenAI anymore in the
future. Well he might lead its liquidation. But It
was just a start-up, a short episode, not the
beginning of a new behemoth.
Mild Shock schrieb am Montag, 20. November 2023 um 02:24:29 UTC+1:

Fuck around find out. Guess who is a board reporting
to? Well the stock holders, and they are reinstantiating
the board every year. So if stock holders want Sam

Altman, stock holders will have Sam Altman, they simply
send the board into /dev/null and create a new one.
BTW: Interesting information about the board,

didn't think they have relation to Sam Bankman-Fired,
and hence its rewarding to listen to TiffanyFong:

s Sam Altman BACK as CEO of OpenAI?
Board Resigning & “Effective Altruism” Ties [UPDATE] https://www.youtube.com/watch?v=LXDWy5-bc-c

Yeah, get your finger off EA, see FTX fiasko.
Mild Shock schrieb am Sonntag, 19. November 2023 um 04:08:02 UTC+1:

Comedy on the horizon and lots of tech penny stocks to buy!
Before you criticize the board at OpenAI, walk a mile in their shoes:

VCs Congratulating Themselves 👏👏👏 https://twitter.com/VCBrags/status/1725976543585784022

LoL
Mild Shock schrieb am Sonntag, 19. November 2023 um 03:57:40 UTC+1:

Ha Ha, board fired Sam Altman of OpenAI.
What now? What will become of ChatGPT?
Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree.
They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114

From Newsgroup: comp.lang.prolog


So what needs to be done? Foremost move Chat AI
out of the computing cloud claws. Make it runnable
on the Edge, make it runnable by simply copying a
LLM on a 1 TB NVMe. Make it runnable by simply
starting your 1'000 core Handy or Laptop. No API
subscription nonsense. So we are just at beginning,
we are still at square zero !!!
This Breakthrough AI Chip is BIG Trouble for Apple & Intel Stocks https://www.youtube.com/watch?v=nJQulJ_gBjo
Mild Shock schrieb am Montag, 20. November 2023 um 12:49:40 UTC+1:

Some actors are messing real hard with OpenAI.
My prediction, the only goal is to dissolve OpenAI.
Take the nomination of Emmett Shear. It shows

how unprepared and desperate the board is.
Mostlikely he will not accept the nomination, since
we wont hear anything of OpenAI anymore in the

future. Well he might lead its liquidation. But It
was just a start-up, a short episode, not the
beginning of a new behemoth.
Mild Shock schrieb am Montag, 20. November 2023 um 02:24:29 UTC+1:

Fuck around find out. Guess who is a board reporting
to? Well the stock holders, and they are reinstantiating
the board every year. So if stock holders want Sam

Altman, stock holders will have Sam Altman, they simply
send the board into /dev/null and create a new one.
BTW: Interesting information about the board,

didn't think they have relation to Sam Bankman-Fired,
and hence its rewarding to listen to TiffanyFong:

s Sam Altman BACK as CEO of OpenAI?
Board Resigning & “Effective Altruism” Ties [UPDATE] https://www.youtube.com/watch?v=LXDWy5-bc-c

Yeah, get your finger off EA, see FTX fiasko.
Mild Shock schrieb am Sonntag, 19. November 2023 um 04:08:02 UTC+1:

Comedy on the horizon and lots of tech penny stocks to buy!
Before you criticize the board at OpenAI, walk a mile in their shoes:

VCs Congratulating Themselves 👏👏👏 https://twitter.com/VCBrags/status/1725976543585784022

LoL
Mild Shock schrieb am Sonntag, 19. November 2023 um 03:57:40 UTC+1:

Ha Ha, board fired Sam Altman of OpenAI.
What now? What will become of ChatGPT?
Mild Shock schrieb am Montag, 6. November 2023 um 01:42:59 UTC+1:

Please solve this with logic:

When Alice entered the forest of forgetfulness, she did not
forget everything, only certain things. She often forgot her
name, and the most likely thing for her to forget was the day
of the week. Now, the lion and the unicorn were frequent
visitors to this forest. These two are strange creatures. The
lion lies on Mondays, Tuesdays, and Wednesdays and tells
the truth on the other days of the week. The unicorn, on the
other hand, lies on Thursdays, Fridays, and Saturdays, but tells
the truth on the other days of the week.

One day Alice met the lion and the unicorn resting under a tree. They made the following statements:

Lion: Yesterday was one of my lying days.
Unicorn: Yesterday was one of my lying days.

From these statements, Alice, who was a bright girl, was able to deduce the day of the week. What was it?

P.S.: Please no DC Proof solutions where animals outside
of the forest appear because of Russell Paradox.

--- Synchronet 3.20a-Linux NewsLink 1.114