Arc Forumnew | comments | leaders | submitlogin
Opinion Poll - Arcists thought on Prolog
1 point by fail 6129 days ago | 2 comments
The paragraph, "This is one reason the source code of Arc itself is so short, incidentally. I did the same thing to it. But my first priority was making applications shorter, not the language. There are features, most notably Prolog-style pattern-matching, that seem to promise great savings in length, but turn out only to be useful for writing a few basic sequence operations like append, remove, and so on. Prolog is a great language for writing append; after that it's all downhill." makes me wonder,

What are those who are interested in Arc's thoughts on the language Prolog?



3 points by kennytilton 6129 days ago | link

I actually made a brave effort to use Prolog for serious development, and Paul pretty much nailed it on the head: it just does not scale, mostly because its all logic programming all the time. That can make normally direct things rather painful. OTOH, I am using PAIP Prolog embedded in my big CL application and getting huge mileage out of it -- for a tightly defined problem that does indeed benefit from Prolog's strengths. Basically Prolog's mistake (iff it wanted to be an all-porpoise language) was not offering other paradigms.

-----

3 points by chaos 6129 days ago | link

My thoughts on Prolog:

Why, oh why, didn't I take Werner Hett's (aka Mr. Ninety-Nine_Prolog_Problems) Prolog/logic programming course when I had the opportunity. :/ Maybe then I could actually judge the language.

When used in small doses (and I think that's how it would be used, when embedded in other languages) it can be incredibly elegant, readable and short.

-----