Відео

Thank you for this video. It has been very helpful in understanding Prolog.

Very useful, thanks!

Prolog is not definitely your average programming language... and I like it.

The intro aged very well.

Hello Markus, could you create a video about CHR? You're the best resource for learning Prolog. Thanks!

Great video as ever. I don't have `volatile` in my scryer-prolog operator table. Is that something specific to your environment?

This was the best prolog intro I had! Also this emacs config is great, gonna try both of 'em soon, damm as a nvim guy you kinda stole my heart there.

How are representing the AST on the sides of iur videos? Btw... Thanks alot for this whole channel, this is awesome and I thank you for that. People like you (teahcers) are the people that contribute the most with the world.

what is the software you use for these videos? You seem to evaluate prolog code and then print a tree is it emacs?

Enlightening as always. I’m inspired to rewrite a bunch of code tomorrow.

A says: Either ... or. To me this is a xor, not an or. Then you also have the solution A=0, B=1. goal: sat((~A # B) =:= A), labeling([A,B]).

Why not "is" instead of #=

The problem is for some boards, like 5x5, there is no circuit, but there is a complete path.

If you create the new state "out of the board", you must start and end in out of border. The first move from out of board can go to any position and the last move can go from any position to out of board, the condition is: if all squares are not empty you go to out of board. It will, at least I hope, create all possible paths. In graph representation, is a new node that is connected to all nodes, and it is the start point of the circuit. What do you think?

6:00 - using clpz disunction/conjunction constraints 7:00 - foldl/3 explanation 14:13 - most general query example 15:08 - predicate overgeneralization example, explanation, and tradeoff explanation 16:50 - pairs library, pairs_key_value/3 predicate, label/1 predicate 18:50 - use of lambda to project away bindings 21:29 - core relations and conventions in constraint programming 21:35 - labeling/2 guarantees 26:42 - first example of thrashing with visualization 27:40 - example of ff options for labeling/2 27:47 - second example of thrashing with visualization 29:10 - discussion of weak propagation properties of clpz conjunctions and disjunctions 32:06 - base8 representation of chess board 33:15 - second example of weak propagation 35:27 - introduction of tuples_in/2 and all_different/1 37:46 - practical example of tuples_in/2 and all_different/1 used in a relation 38:29 - bounds consistency 39:02 - domain consistency 41:07 - third example of thrashing 43:51 - defining model as in "modeling a task" 45:05 - discussion of mental model of declarative programming 48:53 - overview of clpz combinatorial constraints 49:00 - tuples_in/2 49:07 - global_cardinality/3 49:20 - circuit/1 50:00 - circuit/1 explanation and demonstration 51:40 - representing knight's tour as a graph 52:37 - base N coordinate n_x_y_x/4 relation development and intuition 53:47 - knight_move/3 state transition / successor relation rewrite 56:08 - n_tour/2 development and rewrite 56:11 - knight_graph/4 predicate development and intuition 1:00:05 - practical usage of lambda with foldl/3 and clever inline toplevel query for iterative development 1:01:49 - discussion of operational constraings using circuit/1 1:04:22 - "Ready? Go". Successful knight tour. 1:05:10 - observations 1:08:01 - developing a readable output 1:08:59 - another example of foldl/3 usage 1:09:40 - example of nth1/3 usage 1:11:45 - usage DCGs in to develop format string 1:18:14 - recap of modeling and discussion of mental model for Prolog and conclusion *Note* for curious non-prolog developers: The name/number syntax indicates to other Prolog programmers predicate/arrity. If you have never worked with predicates before, it is easiest to categorize them as "Prolog functions" (but you will also soon see that the differences between them are great enough that the phrase "Prolog function" would make most Prolog developers instantly cringe -- I'm just mentioning this to help you orient yourself to this comment if the syntax is unfamiliar) *Note1:* DCGs stands for Definite Clause Grammars, Markus has another excellent video on them. He also has a **great video just on string formatting in Prolog.

I think that most of the arguments about readability (for foldl or the final definition of knight_move) are not really convincing (people not used to functional programming have no idea what foldl does for instance). I also think that saying that there is no disjunction in an equality constraint with an absolute value… is only true if one knows how the solver is programmed ;) For the rest, didn't watch everything, but it seemed pretty nice !!! 👍🏼

Any hamiltonian path can be extended into a cycle by adding an additional vertex that connects the start and end of the path, so I imagine a way to model all knights tours is to add a (N*N + 1)th vertex that connects to all of the squares and consider the vertices next to it on the path as the ends of the tour.

I'll be following a course on Prolog at uni next semester, and I can already tell this'll be a very useful channel. Thank you for the content you put out!

Finally! I'm always waiting for these.

Let's go

norayr.am/papers/Dijkstra%20-%20Efficient%20String.pdf

Rust and eveything written in it is garbage.

Can I close the clause?

I'm starting your online book, and I'm starting with just your videos (one after another). I've read some Prolog books before, but I want to learn to read and write Prolog like a Prolog programmer would read and write Prolog, so stuff like mother_child(Mother, Child) instead of isMother(X, Y).

this is like existential philosophy 303

This was so helpful! Thanks

Amazing video. Very fun topic!

Nice to see the progress since this video was made. When trying the example at 3:45 in this video, there is no longer a redundant choice point and the query returns only one result.

Very well illustrated.

All these years on the web and one day, *boom*, someone else recommending Dvorak! 😬

I have been working for some time with the Prolog programming language and have recently installed Scryer Prolog 0.9.3. I am currently working my way through the book "Prolog programming in depth" by M. Covington. The best source for learning Prolog that I have found so far. Your wonderful video series on Prolog supports the whole thing. So far I have been able to use Scryer Prolog to solve all the problems I have encountered. There's just one thing I've noticed. The predicate 'listing/0, listing/1' does not seem to work. With Gnu P. and SWI P. the predicates work. Unfortunately not for Scryer. My question therefore relates to the current status of the predicate 'listing', or what am I doing wrong here.

The year of the Prolog programming language!.. next year will surely be the one... I'll stick with C.

Interesting that the redundant choice-point is no longer there in up-to-date scryer-prolog.

Phenomenal series of videos. Perpetual Prolog beginner here. I get the arguments against the cut, but how would you then handle this very simple example? We want to get a thing. Getting it is failure-prone and expensive; we'll retry a couple of times and then give up (fail) if we can't get it, as we conclude that it's somehow ungettable. We really don't want to get it more than once. I would write without hesitation: get_thing(NumAttempts, Thing) :- NumAttempts > 0, go_get_it_and_pay_high_price(Thing), !. get_thing(NumAttempts, Thing) :- NumAttempts > 0, OneFewerAttempts is NumAttempts - 1, get_thing(OneFewerAttempts, Thing). I understand this eliminates logical possibilities -- which might be an issue if, say, thing changes over time in relevant ways. But in practice, let's say we just want the first success, not all successes, and we don't want a problem later in the program to trigger a backtrack over our get_thing/2. How would you solve this?

Supplement to former comment: Same n_factorial program executes in GNU Prolog 1.5.0.

man, I would like to learn plot the tree like in minute 1:14, did you have a vídeo of configuration of make trees?

Hi guys , Do anyone have idea how to integrate prolog in networking?

I have a question about the general structure of the N-Queens problem. I've implemented code similar to that presented in your previous N-Queens video. The heuristic I use is a very simple first-fail. Typically, my solution converges within 10,000 recursive function calls when iterating through board sizes. However, beyond a board size of 80, it loses this ability. Why is this? Additionally, is there a more suitable heuristic for this problem, and are there specific claims that can be made about such a heuristic? Also, what is the upper limit on the twice-reordered list heuristic combined with first fail? In the video, you demonstrate quickly solving for 120, but what about 150 or 200? Thank you.

Excellent Work!

Nice video! However I think you are (unconsciously) performing a magic trick while posing the problem and writing safe_queens/1. By searching for a 1D list, instead of a 2D structure, you are doing optimization while defining the problem to be solved. The idea to choose a 1D list as a representation is of course relatively obvious. But it is arbitrary. The underlying observation also holds for each row. We could further minimize the search space, by searching for a permutation of the numbers 1..N. This (in my opinion) works against the point you are making in the first half of the video. Sure, one can try different labeling strategies, while not modifying the representation. But tinkering with the *logic* part is equally effective.

Video starts as a trivial N-Queens problem, but then went into search optimization and introduced good way of thinking about performance of Prolog programs. Nice!

The prolog I know comes from watching your videos and scrolling trough "the power of prolog" website. Thank you sharing all :)

Amazing video. Gotta get myself more seriously into prolog or something like it :)

Fascinating, as always! Elucidated with expertise.

Another awesome video, thank you for your wonderful work.

A highly awaited sequel!!

Finally!!! Thanks Markus! Please don't make us wait so long between videos :)

Hello i had a question i can not solve yet: What happens if aº(bºc) = (aºb)ºc != aºbºc, what kind of mathematical structure is this, is it still a group?

Forget to the link to the paper d1.awsstatic.com/whitepapers/Security/Reachability_Analysis_for_AWS-based_Networks.pdf

AWS uses the optimized datalog solver soufflé to statically analysis network connectivity. One use is to prove security invariants like the DMZ network cannot access the management interfaces of any device.