[tannhaeuser]

Sure, Erlang was prototyped on Prolog because Prolog has excellent built-in facilities for domain-specific languages: you can define new unary or binary operators along with priorities and associativity rules (you can use this to implement JSON or other expression parsing in like two lines of code, which is kindof shocking for newcomers, but comes very handy for integrating Prolog "microservices" into backend stacks), and you get recursive-decent parsing with backtracking for free as a trivial specialization of Prolog evaluation with a built-in short syntax (definite clause grammars) even.

But apart from syntax, Erlang has quite different goals as a backend language for interruption-free telco equipment compared to Prolog.

[rramadass]

In The Development of Erlang Joe Armstrong says "We concluded that we would like something like Prolog with added facilities for concurrency and improved error handling".

See pdf linked here - https://news.ycombinator.com/item?id=40998632

[derdi]

You're reading too much into that quote. This is in a section titled "early experiments". It was an initial goal.

There is a lot of historical connection to Prolog, due to the original implementation, and there are syntactic similarities and non-linear pattern matching and dynamic types and a general declarative vibe, but the actual end result of Erlang's evolution, despite the goal of "something like Prolog", is not very much like Prolog at all. Erlang is a functional language, not a logic language. Prolog is a logic language, not a functional language. General goals like in that quote can change over the decade-long development of a language.

[rramadass]

You are stating some obvious things which are not what is being argued here. I only gave the above quote because it is the conclusion that the Erlang designers came to after a lot of research/playing/implementing with other languages (read the full paper and others listed below). I had also stated above that "he conceived Erlang as Prolog-Ideas+Functional/Procedural+Concurrency+Fault-Tolerance." So we already know Erlang is a different language. Note that Robert Virding also states in the Erlog page "due to the fact that Erlang evolved from Prolog". So obviously Prolog was a huge influence on Erlang design (not to be confused with the fact that the first experimentation was done in Prolog) in addition to other systems/languages. The above paper also states "It was a strange mixture, with declarative features (inherited from Prolog), multi-tasking and concurrency (inherited from EriPascal and Ada) and an original combination of error handling mechanisms". The last came from AXE/PLEX and others.

Joe Armstrong wrote two papers The Development of Erlang (linked here https://news.ycombinator.com/item?id=40998632) and a longer A History of Erlang (pdf at https://www.labouseur.com/courses/erlang/history-of-erlang-a...). In addition to his thesis (pdf at https://erlang.org/download/armstrong_thesis_2003.pdf) they provide a fascinating study into what goes into the design of a language i.e. lots of messy experiments, shifting goals, inspiration/features from many different languages etc. until everything coalesces into a organic whole which is then validated by users. Reading the above two papers will give you a more complete picture of Prolog's influence on Erlang (in addition to others).

[derdi]

Thanks, I know that history, and nobody is denying the influence. The broader context are your claims about "a lot of commonalities here". There are commonalities, but they are shallow.

[rramadass]

That we have to disagree. Leaving aside the non-Prolog features, much of the Erlang syntax is definitely inspired by Prolog (mentioned in many papers including https://www.erlang.org/faq/academic). That was the reason the HN submission i had referenced which linked to a reference sheet comparing Prolog/Erlang/Elixir was so interesting (direct link https://hyperpolyglot.org/logic). In most cases the Erlang syntax is just a simplified version.

[derdi]

I mentioned the syntactic similarities (and more) above. Syntax is shallow.

[YeGoblynQueenne]

Yeah but to be honest, Erlang ended up being not something like Prolog at all.

I think Joe Armstrong was a user here and I interacted with him waaaay way back when I first joined. He's dead now :(

[rramadass]

See https://news.ycombinator.com/item?id=41013798

[YeGoblynQueenne]

I saw, as I saw your upthread quote from the paper above but I think you misunderstand it. The previous paragraph gives context to your quote:

  The main conclusion [5] was that declarative language
  programs for POTS were a lot shorter and easier to
  understand than imperative language programs. Un-
  fortunately the declarative languages lacked features
  for concurrency control and had poor error handling
  facilities.

  We concluded that we would like something like Pro-
  log with added facilities for concurrency and improved
  error handling. No such language existed at the time.

"Something like Prolog" here reefers to the declarative features found in Prolog as well as other declarative languages, left unnamed in the quote, which make programs "a lot shorter and easier to understand thatn imperative language programs".

But that's where the "commonalities" you mention in your OP, between Prolog and Erlang, end. Similarly, CSS and XML are declarative but that's where their "commonalities" with Prolog (and Erlang) end.

I'm insisting with this because I'm concerned that your comment promulgates a common misconception about the, like you say, "commonalities" between Prolog and Erlang. These end with syntactic similarities and misunderstanding this can cause some disappointment to people trying to go from one to the other. I've seen a similar misunderstanding arise about "commonalities" between Prolog and, e.g. Haskell - they both have weird, arcane syntax and immutable data structures, but that's all. Or, think of C and javascript: both Algol like syntax and in fact the first js compilres must have been written in C, but that's all. etc.

There is much, much more to Prolog than the declarative syntax.

[rramadass]

First, my other comments for reference; https://news.ycombinator.com/item?id=41013798 and https://news.ycombinator.com/item?id=41015406

I had explicitly stated above that "he conceived Erlang as Prolog-Ideas+Functional/Procedural+Concurrency+Fault-Tolerance." Here i am not making an equivalence between Prolog and Erlang but emphasizing the inspiration that Prolog provided for the design of Erlang (that is what is to be understood from all the links to the papers and comparison charts given in earlier comments). That inspiration is in the syntax of pattern matching, Atoms/Variables, Module/Export directives etc. Similarity of Syntax is very important for learning/understanding new languages since your cognitive load is decreased dramatically. A good example is C++ to Java/C# where the similarity in syntax (though the runtime object model is very different) is what was crucial to the widespread adoption of the latter. It is in this sense that Prolog to Erlang comparisons should be understood.

Your arguments of grouping/comparing widely dissimilar languages are somewhat disingenuous. A much better side-by-side comparison of languages is https://hyperpolyglot.org/ where the author has tried to group by intended functionality and historical development. I think this is a good way to do it.

[YeGoblynQueenne]

>> It is in this sense that Prolog to Erlang comparisons should be understood.

What I'm saying is that this is the entirely wrong thing to focus on, i.e. syntactic similarities. Prolog is a language of the logic programming paradigm, one of the first ones. Its syntax is that of (a restriction of) the first order predicate calculus. Its programs are First Order Logic theorems. Its interpreter is an automated theorem prover. The motivation for Prolog is the ability to program a computer using the syntax and semantics of First Order Logic, to be able to prove the truth values of statements in a formal language automatically, with a computer. It has nothing to do with Erlang, a functional programming language designed to program telephony switches. Any similarity is superficial: Erlang expressions are not definite clauses; everything in a Prolog program is a definite clause, modulo punctuation. Knowing Erlang will not help you learn Prolog just because the syntax looks similar. You can forget about that right now- that's the misconception I'm trying to correct. Don't encourage people to try to understand Prolog that way because you will only cause them pain.

Take the site you link to as an example. It tries to bodge together Prolog with Erlang and Elixir in a "side-by-side reference sheet" that includes rows for "assignment" and "parallel assignment" using =/2. That's the unification predicate! Prolog does not have assignment! Just imagine the suffering of a novice programmer trying to use their knowledge of assignment, in Erlang or any language, to understand the following:

  ?- X = a, X = b.
  false.

That's just setting up programming students to fail, to fail to understand, to fail to learn- and to only succeed in blaming Prolog for being a stupid, painful language that is hard to learn. Of course it's hard to learn! If you go around telling people that they can learn it more easily if they know Erlang!

>> Your arguments of grouping/comparing widely dissimilar languages are somewhat disingenuous.

No, the point is that they are widely dissimilar and that you won't understand that if you just stop at the syntax. Like the site you link to, where someone clearly made an effort to memorise syntax but completely failed to understand semantics.

Semantics shmantics! The attitude I see here is the one that Peter Norvig criticises in "Teach yourself programming in 10 years": try to find a shortcut around the hard stuff so you don't have to use the brain. You don't learn anything that way.