This is one argument I don't buy. He talks about the Mort programmers never being able to understand Lisp. I am willing to bet, any Mort programmer who looked at this C# would not be able to understand it. I'm sure they could take advantage of it, but write it themselves, I don't think so. And anyone who could write this, can grasp Lisp themselves.

I program in C#, and lambdas and delegates are nice (I can't use expression trees because the company I work for standardizes on .NET 2.0). But the system is too gnarly, still verbose as fuck, and nowhere near as nice as Lisp (as Krishnan freely admits). This just increases the need for sophisticated developer tools and makes things more obscurantist, not less. C# is just another variation on Java, i.e. a way to drag Elvis types 10% closer to Lisp without depriving them of their oh so precious ALGOL syntax.

Fuck the priesthood, seriously.

    [a, b, c] -> '(a b c)

    f(x) -> (f x)

Some syntactic sugar:

    {
        f(x)
        g(x)
    } ->
    (progn
        (f x)
        (g x)
    )

http://docs.plt-scheme.org/algol60/index.html

    destructuring-bind([a b &rest c] some-list lambda([x] +(a b x)))

    loop(for i from 1 to 10 collect 1+(i))

    with-current-buffer(get-buffer-create("foo")
      save-excursion(
        insert(text)
        buffer-substring-no-properties(point-min() point-max())))

Lisp gets it right by excluding sugar from the language kernel, but a "batteries-included Lisp" should include it in a standard library.

http://arclanguage.org/item?id=8172

The syntax is inspired by McCarthy's in his paper from 1958. It works, but i don't find it useful.

My point is: ITS JUST SYNTAX! Syntax doesn't define the language, just the look. I do like a nice, clean, concise syntax though.

Hell, I even hate doing any complex shell scripting because I can never remember the difference between [ ] and [[ ]] and ( ) when using conditionals.

Write the memorizer? That's basic functional programming, but what does that have to do with Lisp? You could write it in JavaScript:

    function memoize(f) {
        var memory = {};
        return function memoized(arg) {
            return arg in memory ? memory[arg] : (memory[arg] = f(arg));
        }
    }

A lot of ways, languages are like religions - they're part of social structure, and part of people's identity, and they tell you how to pray, when to fast, how to allocate and free memory, how to traverse the call stack, etc. But when you think about it hard enough, it's clear that all of these decisions are better made by individuals, not committees or institutions.

1. http://piumarta.com/software/cola/

This is another reason we should be grateful for Clojure. It will have a .Net implementation soon, too.

1. This only works if you're willing to seclude yourself in the Microsoft Yurt and never leave. Technical benefits aside, history has shown MS is not afraid to abandon projects. It's a dangerous position to take, in my opinion.

And I sincerely doubt a port of Clojure to .NET would result in a 0-difference result. We're back to the exact same problem that plagues other lisp implementations right now, which is to say we haven't really bought much.

2. It's not that different from how things are in the C world, save that you'd share more of your runtime infrastructure. My job has been full of instances where an executable interacts with a diverse set of languages. One example I work on daily includes Erlang, Ruby, Prolog, and C++ all in one space, calling around (in a structured way, of course). The implication that you cannot make modern languages communicate without a single umbrella runtime is demonstrably false.

http://blog.n01se.net/?p=41

http://clojure.org/todo

As a rule, things move pretty fast in Clojure-land. And in the beginning it was dual-platform, so it's not really a new idea either.

He makes the extremely important point that the technology has failed if ordinary people can't use it to get their own work done. However, I've personally seen a project where the aim was essentially to let non-programmers program, and the result was horrendously messy. In large part, I think that's because a lot of people ended up developing that had no understanding of the underlying concepts, and as a result the output was extremely hacked together and unintuitive. I think it's great when people realize that the whole point of most software is to enable regular users to complete a task without worrying about the internal computation, but it's just as important to realize the need to _understand_ what is happening.

It's one thing to take an approach like Apple Automator (a much beloved automation environment), but it's another entirely to say, "General Purpose Software should be within the grasp of the 'average person.'" It's not clear that the goal is even reasonable! A lot of software, to do what it does, requires the use of complex mathematics, algorithms, and cryptography that frequently even the implementors only vaguely understand. Manipulating those in an abstract and correct fashion is difficult in the extreme (e.g., the morass of timing attacks that have plagued modern implementations of cryptographic protocols). We can barely make software with smart people!

Yeah, you are exactly right. Programming languages need to be designed for the very best programmers -- they are the ones that need the productivity boost. For me, not understanding the language is never a problem. Making something "easier to understand" will not help me write software that goes together faster or ends up more reliable. It often has the opposite effect; Java's lack of ... everything, for the sake of simplicity, really slows me down. Comparatively, learning a new concept is much easier. If I start using a library that uses Arrows, I read the Arrow paper, spend a few days playing with it, and then I have one more tool that will help me be more productive every day for the rest of my life. Making a language that doesn't have arrows saves me reading a paper about them, but it wastes my time every day for the rest of my life.

So, "simple" programming languages -- do not want. I am not saying that you should start your "intro to computer programming" class with a long lecture about category theory, but I am saying that you should never hide things from people that want to know about them.

In terms of theory, I found the "Applicative Programming with Effects" paper and the "Typeclassopedia" very helpful.

http://www.soi.city.ac.uk/~ross/papers/Applicative.html

http://haskell.org/sitewiki/images/8/85/TMR-Issue13.pdf

I started doing a real implementation of these data structures in Perl this week. (I say "real" because people have "added Monads to Perl" before, but they didn't really add Monads, they just added a "programmable semicolon".) Anyway, in doing so, I saw in great detail the relationships between the various types (specifically "liftM2 ($)" and "<*>"), and tried to generalize things as much as possible. The result was something like Arrows, and that helped me understand the "why" in addition to the "how". Writing test cases was also helpful to me; using these esoteric stuctures in "real code" helped me build up the intuition needed to advance farther.

Anyway, it took me three readings of the Applicative Programming with Effects paper to get Applicative Functors, and after that, Arrows were pretty easy. So I recommend that paper, and lots of poking around in ghci.

To bring this back on topic, yeah, it would be hard to teach this stuff to Joe Average. But a language that allows it is still important to develop. (I also learned firsthand that language design decisions to make it difficult to implement certain features. Implementing the equivalent of "instance Monoid b => Monoid (a -> b)" is almost impossible to do cleanly, due to the polymorphism of "mempty"; it's defined in Haskell as "mempty _ = mempty", and in Perl, we discard the type information before we have a chance to pick the "second" mempty.)

While I know that there are a lot of horror stories (and I can tell a few), some of the most productive "programmers" I've ever met used MS Access or Excel.

If you don't consider advanced usage of these applications as a form of programming, then you are illustrating the point: It is not only possible to "dumb down" programming, it is inevitable. So much so that you haven't even noticed it happening.

I'm talking about "general purpose software," which encompasses a huge volume of space. Rare is the person who can competently address the wide variety of problems found in the world of software. Excel and Access, in my view, are perfect examples of paring down the acceptable problem space to something reasonable and making tools for people to solve those specific problems naturally. You can't write a 3d engine in excel, but that's okay, because the tool is not for that.

But this argument breaks down when you try to take it to a general-purpose software kit, which has to handle anything and everything. In this wide, unshielded world of hard problems you have to have the most capable and powerful tools available. Deliberately limiting tools in this environment is like saying, "Our river pilots have lots of experience handling barges, so it'd be counter productive to train them on oceangoing vessels, we'll just build bigger, sturdier barges for ocean travel." Yes, sailing on the open ocean requires more powerful and complex tools, but you appreciate those tools when the ocean gets rough.

You can, a guy has done it. I'm only being pedantic because this is cool if you haven't seen it:

http://www.gamasutra.com/view/feature/3563/microsoft_excel_r...

So what? The goal of turning every user into a programmer is perfectly reasonable. Almost all users of the early systems and computers were also programmers. Even today, business users make use of Excel/Access and some macros to do some programming (even though it can be seen as basic).

Don't you remember a time when operating systems came with programming languages and compilers and actively promoted them to the user?

Oh, and HyperCard. There was another great example of "user's" programming. I've also seen some hideously ugly GIMP script-fu scripts but they worked and did the job of the "user" that wrote them.

In the end, the distinction between a user and a programmer is artificially created by software that doesn't view anybody else as smart enough to enter the monastary of the programmer.

I think the trouble with Lisp (and Scheme) is that there's so much cruft and inconsistency beneath the veneer of simplicity (see http://tnovelli.blogspot.com/2009/08/lisp-crisis.html). Lisp is a great language to study; I just wish it was more practical and popular. It sucks having to choose between awkward (Lisp) and inflexible (everything else). :-(

PLT Scheme is competitive in every way with Python and Ruby, I don't know why people keep ignoring it.

It doesn't have to be that way: Clojure has abstract sequence types, with cons/car/cdr wrappers in case you want them.

Please refer to: http://doc.plt-scheme.org/guide/for.html

(defun memoize (fn) (let ((cache (make-hash-table :test #'equal))) #'(lambda (&rest args) (multiple-value-bind (val win) (gethash args cache) (if win val (setf (gethash args cache) (apply fn args)))))))

  (defun memoize (fn)
      (let ((cache (make-hash-table :test #'equal)))
      #'(lambda (&rest args)
          (multiple-value-bind (val win) (gethash args cache)
          (if win val
              (setf (gethash args cache) (apply fn args)))))))

  static Func<TParam, TReturn> Memoize<TParam, TReturn>(Func<TParam, TReturn> func)
  {
      var memoDict = new Dictionary<TParam, TReturn>();
      return arg =>
      {                
          TReturn result;
          if (!memoDict.TryGetValue(arg, out result))
          {
              result = func(arg);
              memoDict[arg] = result;
          }
          return result;
      };
  }
  

    (defun memoize (fn)
      (let ((cache (make-hash-table :test #'equal)))
        (lambda (&rest args)
          (multiple-value-bind (val win)
                               (gethash args cache)
            (if win
              val
              (setf (gethash args cache) (apply fn args)))))))

Second I don't look for { or ( but for the block indentation, so a { on a single line does not give me enough visual clues.

The 'return arg =>' form is not obvious to me. What does it do? Where dies TReturn come from, TParam?

Why is there }; and } ?

We could argue for days about these sort of issues, and there are passionate ideas on either side of the fence.

  function
  {
    code;
    code;
  }

  function
    code;
    code;

Have you never programmer outside of lisp? I've never programmed in C#, but if 'arg =>' is the beginning of lambda, then it's because the definition of that line is return EXPRESSION;.

And in that case EXPRESSION = 'arg => { code }'. That's where the ';' comes from.

  var square = x => x * x;

  var square = x => { return x * x; };

The typical/recommended/easy/practical solution is to just provide another overload, pack the args into an object array, and then call the original overload; completely ignoring the problem.

Alternatively, you could do some scary things with reflection...

Even so, with the caveat that you have to learn Common Lisp fist, I think the CL version is infinitely more clear and elegant. All that type information is redundant in the C# version, since in this case all you want to say about the types is that you don't care what they are.

multiple-value-bind is among the longest function name in CL, and doing something that python does much better with syntax alone, perhaps Python is the new Lisp.

Aha. WTF? Ok, if I know what the code means, than I can read it. But that's only the case for C# and not for Common Lisp.

So your statement that The C# one looks very long and ugly is entirely subjective.