101–110 of 289 posts
Girard has some commentary scattered about his writing. The search algorithms for logic programming are simply slow, it's a very interesting idea in programming languages, but there's a reason it's not widely used. > PROLOG, its misery. Logic programming was bound to failure, not be- cause of a want of quality, but because of its exaggerations. Indeed, the slogan was something like « pose the question, PROLOG will do…
Earlier quoted context omitted.
Prolog and Datalog example (they are identical in this case) % Facts parent(john, mary). parent(mary, ann). parent(mary, tom). % Rules ancestor(X, Y) :- parent(X, Y). ancestor(X, Z) :- parent(X, Y), ancestor(Y, Z). % Query ?- ancestor(john, X). The Prolog code looks identical to Datalog but the execution model is different. Prolog uses depth-first search and backtracking, which can lead to infinite loops if the rules…
> Prolog uses depth-first search and backtracking, which can lead to infinite loops if the rules are not carefully ordered Is this an issue in practice? Most languages can create programs with infinite loops, but it's easy to spot in code reviews. It's been over a decade since I encountered an infinite loop in production in the backend. Just wondering if the same is true for Prolog.
Prolog variables can have two states at runtime: unbound or bound. A bound variable refers to some value, while an unbound variable is a "hole" to be filled in at a later time. It's common to pass an unbound variable into some call and expect the callee to bind it to a value. This can cause problems with infinite recursion where you intend to write a call that binds some variable, but the way you've structured your program, it will not actually bind it. So the callee ends up in the same state as the caller, makes a recursive call hoping its callee will bind the variable, and down the infinite recursion you go. With experience you can definitely spot this in code review. You'll also catch it in testing, if you test properly. But it's different enough from other languages that learners struggle with it at first.
Another source of (seeming) nontermination is when you ask Prolog's backtracking search to find an answer to some query in a search space that is too large and may not contain an answer at all, or only an answer that is impracticably far away. This is also sort of Prolog-specific since in other languages you rarely write the same kind of optimistic recursive search. This is harder to spot in code review since it's really application-specific what the search space looks like. But again, you test. And when in doubt, you direct and limit the search appropriately.
Earlier quoted context omitted.
Do you have any papers comparing Scryer with other prolog systems (like SWI-prolog or SICStus prolog) performance-wise ?
There are some benchmarks here of SWI Prolog's benchmark suite on diffrent Prolog systems by Jan Wielemaker the SWI Prolog author: https://swi-prolog.discourse.group/t/porting-the-swi-prolog-... He finds Scryer performs worse, which he does comment on, he also explains some tradeoffs and historic choices in SWI's design which affects its performance. I think I have seen the author of Scryer saying that's not surprisi…
With compliments to your prof ;), interest in Prolog just now is recovering from a year-long focus on W3C's RDF/SPARQL. TBL surely had an itch to scratch with regards to logical knowledge representation dating back even longer than the web [1]. But Prolog has broader applicability not only in logical/knowledge graph querying, but also in solving all kinds of discrete combinatorical optimization problems. Or, as the Q…
The problem w/ OWL is that everybody wants to work with first-order logic + math, but Gödel proved it isn't decidable. For instance if I wanted to express financial regulations or business rules inside a bank or other business I'd need to use math: for instance to express the conditions for reserve requirements or approving a loan. OWL is best thought of as a set of templates for generating first-order logic rules th…
He did no such thing. He proved undecidable problems exist in any system powerful enough to be useful. That doesn’t make those systems useless, though.
Prolog, and Constraint Programming especially are great to have in your toolbox. I’ve done research in the field for years, and my job in the industry today is writing Prolog. There are real issues with Prolog: - no proper module nor package system in the modern sense. - in large code bases extra-logical constructs (like cuts) are unavoidable and turn Prolog code into an untenable mess. SWI prolog has single-sided un…
Prolog, and Constraint Programming especially are great to have in your toolbox. I’ve done research in the field for years, and my job in the industry today is writing Prolog. There are real issues with Prolog: - no proper module nor package system in the modern sense. - in large code bases extra-logical constructs (like cuts) are unavoidable and turn Prolog code into an untenable mess. SWI prolog has single-sided un…
Maybe it's just me but I see a lack of a package manager as a massive, massive pro. I can't stand how seemingly every language has a package manager which requires it's own installation and you have to learn how to use THAT thing and then you need some library off github that does some minor task really well but you can't just download the fucking code, you have to import it via, idk, the Fork-Lyft manager which requ…
Earlier quoted context omitted.
Prolog and Datalog example (they are identical in this case) % Facts parent(john, mary). parent(mary, ann). parent(mary, tom). % Rules ancestor(X, Y) :- parent(X, Y). ancestor(X, Z) :- parent(X, Y), ancestor(Y, Z). % Query ?- ancestor(john, X). The Prolog code looks identical to Datalog but the execution model is different. Prolog uses depth-first search and backtracking, which can lead to infinite loops if the rules…
> Prolog uses depth-first search and backtracking, which can lead to infinite loops if the rules are not carefully ordered Is this an issue in practice? Most languages can create programs with infinite loops, but it's easy to spot in code reviews. It's been over a decade since I encountered an infinite loop in production in the backend. Just wondering if the same is true for Prolog.
As an anecdote, a long time ago for a toy project switching two rules order got the runtime to finding all solutions from ~15mn to a around the second (long time, memory fuzzy...). The difference was going into a "wrong" path and wasting a lot of time evaluating failing possibilities, vs. taking the right path and getting to the solutions very quickly.
So in practice even if Prolog is declarative to get good results you need to understand how the search is done, and organize the rules so that this search is done in the most efficient way. The runtime search is a leaky abstraction in a way ;)
It's not an issue limited to Prolog, many solvers can be helped by steering the search in the "right" way. A declarative language for constraint problem like MiniZinc provides way to pass to the solver some indication on how to best search for example.
Also, most modern Prolog support tabling, which departs from strict DFS+backtracking and can help in some cases. But here too, to get the best results may require understanding how the engine will search, including tabling.
Earlier quoted context omitted.
There are some benchmarks here of SWI Prolog's benchmark suite on diffrent Prolog systems by Jan Wielemaker the SWI Prolog author: https://swi-prolog.discourse.group/t/porting-the-swi-prolog-... He finds Scryer performs worse, which he does comment on, he also explains some tradeoffs and historic choices in SWI's design which affects its performance. I think I have seen the author of Scryer saying that's not surprisi…
So SWI appears to be more performant, it has an open license, so as per the GGP's claim regarding Scryer in the post above, it must not be ISO-compliant?
Some default settings in SWI are not ISO-compliant (for example, it uses a string type that does not exist in ISO). But these are minor things that won't usually trip you up when feeding it ISO code. You can set flags to get it to conform in the way you want. And you should set flags whenever you want your ISO Prolog programs to be portable, because the standard is very lax and leaves a lot of things implementation-defined. But it specifies the flags to get implementations into the state you want.
Earlier quoted context omitted.
There are some benchmarks here of SWI Prolog's benchmark suite on diffrent Prolog systems by Jan Wielemaker the SWI Prolog author: https://swi-prolog.discourse.group/t/porting-the-swi-prolog-... He finds Scryer performs worse, which he does comment on, he also explains some tradeoffs and historic choices in SWI's design which affects its performance. I think I have seen the author of Scryer saying that's not surprisi…
So SWI appears to be more performant, it has an open license, so as per the GGP's claim regarding Scryer in the post above, it must not be ISO-compliant?
"Once I saw this guy on a bridge about to jump. I said, "Don't do it!" He said, "Nobody understand me." I said, "What's so special about you?"
He said, "I'm a computer guy." I said, "Me too! Desktop, tablet, console, smartphone?" He said "Desktop, mostly", I said "Me, too! Mac, Linux or Windows?" He said, "Any, I'm a programmer." I said, "Me, too! which style? OOP, Imperative, Functional, Logic, Array, Stack" He said, "Logic." I said, "Me, too! What subset? Answer Set Programming, Abductive Programming, Prolog, Datalog?" He said, "Prolog." I said, "Me, too! Conformant with the ISO/IEC 13211-1:1995 (core) standard term syntax for the period character or non-conformant extention decried by members of the ISO/IEC JTC1 SC22 WG17 working group?"
He said, "SWI Prolog 7" I said, "Die, heretic!" And I pushed him over."
- https://news.ycombinator.com/item?id=26624442
or read more seriously here:
- https://www.complang.tuwien.ac.at/ulrich/iso-prolog/SWI7_and...
Girard has some commentary scattered about his writing. The search algorithms for logic programming are simply slow, it's a very interesting idea in programming languages, but there's a reason it's not widely used. > PROLOG, its misery. Logic programming was bound to failure, not be- cause of a want of quality, but because of its exaggerations. Indeed, the slogan was something like « pose the question, PROLOG will do…
This is a great quote and sadly true. What text is this from?