On Haskell, Ruby, and Cards Against Humanity

http://www.drmaciver.com/2015/04/on-haskell-ruby-and-cards-against-humanity/

by luu • 11 years ago

61 61 11 years ago

malvosenior2• 11 years ago

Yikes, someone has a case of the Mondays!

I think playing around with the abstractions that Haskell offers can be very rewarding when you go back to your "regular language". It's just approachable enough that you can learn some really advanced techniques and dip your toes into category theory all through self study.

As for Ruby, I'm not proficient in it, but plenty of people have built amazing products using it and seem happy with their language of choice. Not sure why they should be bashed for poor programming when in the end, results are the only thing that matters.

A thing I value probably at least as much as humour is good API design.

Since he's throwing a lot of rocks, I'll say his house is looking pretty glass :) Slaying imaginary orcs isn't exactly the height of comedy.

DRMacIver4• 11 years ago

The difference is, I know when I'm not being funny.

rnalexander3• 11 years ago

Do you know that by invoking "Case of the Mondays" you're referring to a specific character from the movie Office Space ( https://www.youtube.com/watch?v=2AB9zPfXqQQ ) who, in making that comment and then chuckling to herself as she walked away, presented almost a perfect example of someone who thinks they are far more funny than anyone else does? You do get that right?

npizzolato2• 11 years ago

I chuckled too, probably because I had similar thoughts as the GP when reading the article. Sure, it was low hanging fruit, but I'd rather laugh at that than be someone who sticks their nose up because they're too good for games like Cards Against Humanity and pop culture references.

malvosenior2• 11 years ago

Yep I get it. That is the whole point of the joke after all.

eli_gottlieb• 11 years ago

Yikes, someone has a case of the Mondays!

Never say that, or I will burn the building down.

Zikes3• 11 years ago

People playing Cards Against Humanity don't merely think they are funny because they have the perfect card(s) to play, though the results themselves might be genuinely funny. It provides a backdrop against which more humor can be made. A perfect example is the somewhat recent TableTop episode with Aisha Tyler: https://www.youtube.com/watch?v=QCEqUn7If44

rnalexander3• 11 years ago

Yes, a perfect example of a person who does professional comedy being funny in a way that makes it look so easy that any rank amateur thinks they surely can do it as well (when they can't.)

Zikes3• 11 years ago

It is a perfect example of how much genuine humor can arise tangentially to the game. The author's argument is that CAH is not truly funny or that the players are not truly funny because it's just slapping a few naughty words together. The video serves to demonstrate that CAH can provide a solid foundation for spontaneous humor, professional status notwithstanding.

vezzy-fnord• 11 years ago

There's nothing intrinsically wrong with Haskell, but it does indeed attract lots of fundamentalists. That said, these overwhelmingly tend to be lower skilled or beginning. You learn some concepts like immutable state, referential transparency and static typing, at which point you become rabidly evangelical about functional programming... at the expense of all other paradigms. People complain about imperative blub programmers, but then they proceed to forget that concatenative, logic, array and message passing OO also exist. It's purely a functional-imperative dichotomy to some people.

And so you begin to shoehorn absolutely everything into the new magic potion you've learned. Erlang? How can I build concurrent and robust systems with it, it's dynamically typed and the syntax is scary! (Yes, people who somehow managed to learn Haskell syntax proceed to cry like babies over having to figure out some clause delimiters. The irony is enough to power the US steel industry for a year.) Mutable state and its evil becomes the next cargo cult, type systems means I no longer have to test my software, and other similar misconceptions abound.

But, I don't mind. Most tend to grow out of it eventually. You feel hardcore at first, but you're struck by humility once you learn more and more.

By the way, Cards Against Humanity is quite nice.

pron3• 11 years ago

but then they proceed to forget that concatenative, logic, array and message passing OO also exist.

Right. I think, in general, PFP (pure-FP, i.e. Haskell and a few more obscure languages) sells itself as a solution to a problem and some people start believing it's the only solution to the problem, rather than one of many (and probably not the best one, at that). Not that Haskell's virtues aren't real -- they are -- but they come at significant cost, and are often available come much cheaper with other approaches. For example, referential-transparency-everywhere is just one of several solutions to the problem of mutable shared state, and, IMO at least, not among the best solutions. Even when it comes to provable correctness, which Haskell touts as a relatively unique feature (at least relative to mainstream languages) people forget -- or more likely, don't know -- that there are fully verifiable imperative languages out there (with their own serious problems -- correctness is never free), that have actually been used by the industry to write some very important, safety-critical software, and have been actually put to the test much more than Haskell.

In fact, one of Haskell's marketing advantages is that, after 20 years, it is still at the stage of selling the dream. Because it has hardly been used where its strengths are supposed to matter most (complex programs employing a lot of concurrency), it gets to sell its "promising initial results" without having to contend with too much disappointment, which invariably arrives once a technology has been well tested. It's the promising NBA rookie who got picked first in the draft but has never played a full quarter, getting people to say, "if only they'd get him off the bench...".

jcadam• 11 years ago

that there are fully verifiable imperative languages out there (with their own serious problem -- correctness is never free), that have actually been used by the industry to write some very important, safety-critical software, and have been actually put to the test much more than Haskell.

Ada is absolutely painful to work with as a developer (I used to work in Aerospace). But if you could get your code to compile, it stood a very good chance of just working.

dragonwriter4• 11 years ago

I would just like to point out that, for the number of time you have say, in different phrasing, that Haskell is not the best solution to the problems it aims to solve, you provide not one bit of argument or evidence to support that for any of the areas you say it about.

Anyway, lots of the reported production uses of Haskell are in exactly the areas it is supposed to be strong; the problem for sharing that experience is many of them are in internal proprietary systems, not publicly-available or open-source software, so external evaluators don't have the code -- or even, often, a chance to interact with the software.

davidism• 11 years ago

People walk away from a game of Cards Against Humanity thinking that they’re hilarious.

Which people? My friends and I walk away from a game of Cards Against Humanity thinking we had a fun time coming up with stupid combinations for an hour. I can also turn the author's previous sentence around and state that there are probably many people that get too invested in elves and orcs as well.

... this passes once you’re more familiar with the language and are using it for real things
... a large enough subset of the community never reaches that stage

This is true about pretty much every programming language (and probably generalizable to most non-programming things too). The author just happened to apply it to Ruby and Haskell for some reason.

If there's a message such as "programmers should be more aware of the language's strengths and weaknesses and shouldn't feel overconfident", it's lost in the paragraphs of vague complaints.

thoughtpalette• 11 years ago

Yea, I get his arguments about API design, but the CAH stuff was somewhat unwarranted and an overgeneralization.

aantix• 11 years ago

His CAH argument was weak at best. Does he hate iMovie too because the videos produced inflate the perception of his video editing abilities?

After playing a game of CAH, I don't think any of my friends have any delusions about doing standup or writing a comedic op-ed piece for the NY Times.

roneesh• 11 years ago

I'm not going to doubt that the author has reasons for saying what he's saying, but examples would have been nice.

For instance most API design is based off the standard Rails scaffold command, and an explanation about what's wrong about that would have been nice, and elevated this article from kind of a screed, into something informative and worthy of study.

zxdfghj4• 11 years ago

Given that most people are not very funny[0] but that most people like to believe that they are funny, it is probably a very profitable enterprise to sell someone an experience that lets them believe (even if just for an hour) that they are much funnier than they actually are.

[0] consider that 50% of people are less funny than average, and then reflect on how funny the average person is.

seba_dos12• 11 years ago

Where did you get your "50% of people are less funny than average" from? Do you know what average is?

cven714• 11 years ago

He got it from George Carlin.

zxdfghj4• 11 years ago

I know what the distinction between the mean and the median is, and I know which one is appropriate to use with quantity like "how funny someone is" (hint: you can take three people and point to the one who is in the middle on the funniness scale, but you can't add together how funny they are and divide it by three).

taeric2• 11 years ago

For some reason, the 50% of people are above/below average fallacy cracks me up every time I see it. Consider, how many people have more/less than the average number of fingers?

zxdfghj4• 11 years ago

You can meaningfully count up the total number of fingers that five people have between them, and divide it by five.

You cannot meaningfully add together how funny five people are and divide it by five. However, you can put five people in order of how funny they are, and point to the middle one.

The only appropriate concept of average for ordinal data (e.g. how funny someone is) is the median, from which it follows by definition that 50% of people are less funny than average.

PS - if it is true that it "cracks you up" every time you hear "the 50% of people are above/below average fallacy", I suggest that you might be one of the 50% of people who have a below-average sense of humour ;)

dragonwriter4• 11 years ago

consider that 50% of people are less funny than average

Assuming that average here is an arithmetic mean (which is the most common use, but how are we quantifying "funniness" to warrant such a measure?), this would only be true if we assume a symmetric distribution with no individuals at the average.

This also assumes that objective funniness is even a meaningful concept; funniness is obviously subjective, and I would suspect that people tend to voluntarily socially engage with people that they find pleasant to be around -- part of which is having compatible sense of humor. So, the people you are voluntarily associating with (i.e., the people you are like to play CAH with) are likely to be disproportionately from the upper end of the funniness distribution from your subjective perspective -- and are likely to be among the upper end of the distribution in how they perceive your funniness, as well.

zxdfghj4• 11 years ago

I don't think that the arithmetic mean is appropriate here (see my replies to two sibling comments) but rather the median.

Your second point is a good one - clearly, funniness is subjective. On the other hand, I don't think my argument relies on the precise quantitative definition of funniness!

wyager• 11 years ago

What is the point of this article? "Most people aren't funny and actually suck at programming"? What does the author hope to achieve with this?

skywhopper• 11 years ago

I'm not sure I follow the parallels between these three things. There's really no explanation. How does Haskell's obtuseness compare to the ease of creating shocking jokes with Cards Against Humanity? What mechanisms are shared by Cards Against Humanity and Ruby which make it easy to create amusing juxtapositions in CAH and encourage pretty-but-poor APIs in Ruby?

I'd be a lot more interested in hearing why the author thinks the Sequel and Sinatra APIs are good, and how they differ from more typical poor Ruby APIs, and what features of Ruby the author thinks encourages these poor APIs.

What languages or language communities encourage better practices, and what makes them different from Haskell and Ruby, and how does that tie back to the Cards Against Humanity analogy? By the end of the piece, I was left wondering what the point was to begin with.

TheGRS• 11 years ago

Ah, the old elevating yourself by putting down some popular things trick. Never ceases to get some unwarranted attention.

freyrs3• 11 years ago

The way this problem manifests in Haskell is in how incredibly clever it makes you feel to get something done in it. Haskell is different enough from most languages that everything feels like an achievement when writing it. “Look, I used a monad! And defined my own type class for custom folding of data! Isn’t that amazing?“. “What does it do?” “It’s a CRUD app”.

The fallacy here is that type classes, folds and monads are not esoteric structures within the context of Haskell, and using them to structure regular business is pretty mundane. If someone feels a sense of achievement from doing this, they're probably very new to the language. The author of the article has baked in a value judgement about the "normal" way to structure a CRUD application that precludes alternative approaches to structuring application logic.

You could just as well argue that laying out object class hierarchies and using inheritance is "[solving problems] that have nothing to do with what you’re actually trying to achieve."

jbooth• 11 years ago

A lot of people do argue that about big object-class hierarchies and inheritance, and they've been declining since the mid-2000s at least in mainstream OO languages like Java and C#.

skj• 11 years ago

You could just as well argue that laying out object class hierarchies and using inheritance is "[solving problems] that have nothing to do with what you’re actually trying to achieve."

Definitely, 100%.

DRMacIver4• 11 years ago

You could just as well argue that laying out object class hierarchies and using inheritance is "[solving problems] that have nothing to do with what you’re actually trying to achieve."

Could, indeed semi-regularly do. I actively dislike solving problems with inheritance.

I also actively dislike people missing the point. The problem is not "using programming to solve problems". The problem is letting the novelty of the programming you're using to solve problems con you into thinking you're doing something more clever than you actually are.

iand6752• 11 years ago

I think you're equally missing the point– for serious Haskell programmers, using a monad to structure computation for a CRUD app is nothing to write home about. Having personally written quite a lot of CRUD code with Haskell, I have never thought "Gee, I wrote this with a monad. I deserve a back pat."

If we're talking about things that make me smug as a Haskell developer, I'd probably say that I feel most accomplished when I figure out how to encode more invariants into the type system so that the compiler helps me make fewer errors. I still stand by Haskell as being the best language for that that's not just a university research language.

serve_yay• 11 years ago

One of the nice things about writing JavaScript all the time is there's never really a chance to congratulate yourself for using some elegantly-designed beautiful gem of a language. My colleagues and I are constantly ribbing JS, in a good-natured way. I like using it but I can't pretend it's not ugly in some ways.

bvanslyke2• 11 years ago

I think that's exactly it. We use js so we can unite against a common enemy & build solidarity among the team :-)

Bahamut• 11 years ago

This is an insipid article - how people feel should have no weight when objectively evaluating a language. If someone enjoys writing in a particular language, that is fine. If someone enjoys a game of CAH, that is also fine. There's nothing wrong with it, and it is stupid to attribute more to it.

pcarolan• 11 years ago

I think you need a hug.

drostie• 11 years ago

I'm probably guilty of this, as a guy who is learning Haskell. FizzBuzz can be just

    main = putStrLn $ unlines $ map fizzer [1..100] where
        fizzer n | n % 15 == 0  = "FizzBuzz"
                 | n %  3 == 0  = "Fizz"
                 | n %  5 == 0  = "Buzz"
                 | otherwise    = show n

but I think the first time I tried to solve it in Haskell I had something more like:

    import Data.Monoid (Monoid, mconcat)
    import Data.Maybe (catMaybes)

    specialcases :: (Monoid y) => [x -> Maybe y] -> x -> Either x y
    specialcases rules x = case catMaybes (map ($ x) rules) of
        [] -> Left x
        xs -> Right (mconcat xs)

    ifDivBy :: Integer -> x -> Integer -> Maybe x
    ifDivBy n x m = if m `mod` n == 0 then Just x
                                        else Nothing

    main = putStrLn $ unlines $ strings where
        strings = map (either show id) cases
        fizzer = specialcases [ifDivBy 3 "Fizz", ifDivBy 5 "Buzz"]
        cases = map fizzer [1..100]

This is basically thinking about an imperative problem the "wrong way around": starting with the logical distinction between "we're in a special case" and "we're in the normal case" and building everything else around it. (Also, who is ever going to use that Monoid instance? Replace `id` with `concat` and you can just remove `Data.Monoid` altogether -- and if you really wanted it generic, taking an explicit `y -> y -> y` and using `foldl1` would be more flexible anyway.)

bvanslyke2• 11 years ago

Ah! I had this same(-ish) idea but couldn't figure out how to write it. And when I realized how much time I was spending on Fizz Buzz forced myself to stop.

Aushin• 11 years ago

Can someone explain to me what "real things" are in the context of the statement 'once you are using [insert language] for real things'?

Is there a a difficulty barrier for considering an application real? A lines of code requirement? Is it like pornography and I'll know it when I see it?

ianamartin• 11 years ago

So basically, some people are douchebags; therefore Cards Against Humanity, Ruby, and Haskell all suck.

Got it.

Holy non-sequitur, Batman.

MrZongle2• 11 years ago

"I don’t particularly hate it for the obvious reasons of middle class white kids sitting around a table giggling at how transgressive they are being with their jokes about race. I don’t approve, and I would probably avoid it for that reason alone, but it’s frankly so banal in this regard that it probably wouldn’t be enough to raise my ire above lukewarm. What I really hate about it is that it’s a game about convincing yourself that you’re much funnier than you actually are."

Wow. There's a lot going on there, up in that ivory tower.

bshimmin• 11 years ago

It's an incredible feeling to read an article where all you can think the whole way through is how little you'd like to go for a pint with the author.

(Yes, I like Ruby, I like to poke Haskell occasionally, and I've only played Cards Against Humanity once, but it resulted in my mother-in-law combining "a girl's best friend" and "three dicks at once".)

QuantumRoar• 11 years ago

But I do find a fairly large subset of their communities really annoying in how much they exemplify this behaviour. They’re too caught up in congratulating themselves on how good they are at something to notice that they’re terrible at it, and it makes for some extremely frustrating interactions.

There it is, in the last paragraph. I was wondering what you are trying to say. But I guarantee you, it's not Ruby or Haskell making them like that. It's rather something intrinsic that feeds on accomplishment, however little it might seem to you. On the other hand, there's also a lot of mathematicians, bloggers and even roleplayers who display quite similar behavior.

Just imagine the pride of the young mathematicians when they prove something difficult for the first time without the help of a text book, when the professor could solve this in his sleep. Or think of the blogger that spreads the word on his latest article, when a philosopher or writer thinks it doesn't say much about anything. Or the roleplayer boasting about his skill for acting out his character, when an actual actor thinks its laughable.

Many people, irrespective of what they're doing, display this kind of behavior. So if these people annoy you so much, avoid them or help them.

Don't tell them that it's frustrating to interact with them. I guess you wouldn't have enjoyed it either if you just started out and someone would have told you that all your accomplishments until now are worth nothing and you shouldn't tell or show anybody what you are doing since it's embarrassing for the community.

Nobody is good at everything, I know a thousand things that I don't know or which I would execute terribly. But I know that I can still reduce this number. And if it takes me a whole week to write a program that is an abomination and it would've been a matter of hours for you, I'll still tell it to everybody since I'm proud of however little it is that I can accomplish.

On the other hand, I won't laugh at you and I won't tell you that it's frustrating to talk to you if you say something about the LHC or black holes or semiconductor physics when you barely have an idea about quantum field theory, general relativity and solid state phyiscs. I'll either avoid you or - what is usually the case - I'll help you understand it.

npizzolato2• 11 years ago

They’re too caught up in congratulating themselves on how good they are at something to notice that they’re terrible at it, and it makes for some extremely frustrating interactions.

I'd rather surround myself with people who are generally happy about things -- those who laugh a lot when playing games like Cards Against Humanity and celebrate achievements, even if they're small -- than those who spend their time writing angry rants about how they're so much funnier and better at programming than everyone else.

Maybe the OP tried to make a more important point somewhere in the article, but it was surely lost to me among the rant.

dragonwriter4• 11 years ago

I don't particularly hate it for the obvious reasons of middle class white kids sitting around a table giggling at how transgressive they are being with their jokes about race.

Why would that be an obvious reason to hate it: its not like middle-class white kids are the only people who play Cards Against Humanity. (The few times I've played it, the only one of those descriptors that was accurate of the group as a whole was "middle-class" -- the groups were racially diverse and included an age range of something like mid-20s to early 50s.)

What I really hate about it is that it’s a game about convincing yourself that you’re much funnier than you actually are.

Er, no. The whole thing is about creative juxtaposition from limited options. Its not really about how funny you are, its about how funny things taken in the wrong context can be. Like Mad Libs, its about how funny the result is, and while there is definitely more deliberate input from the participants, its not mostly about how funny the participants are, because the participants are so sharply constrained that even if they were the world's greatest comedian, they wouldn't be able to show it.

People walk away from a game of Cards Against Humanity thinking that they’re hilarious.

Yeah, I don't see any evidence that this is particularly true.

And the thing is, that the amount that the author here doesn't get CAH is echoed in the amount he doesn't see to get the things he uses it as an analogy for.

There are some exceptions (Sequel and Sinatra for example), but by and large Ruby APIs seem almost actively designed to promote bad software.

Its interesting that the author can name notable counter-examples to the effect he is describing, but can't be bothered to name even a single notable supporting example.

Sure, lots of stuff written in Ruby or any language is bad. But where is the evidence that Ruby disproportionately (either inherently or through community dynamics) promotes bad design? There's no substance here.

Ruby is full of people who think they’re doing good API design, but they’re not. The problem is that in Ruby everyone is designing “DSLs”.

That's not even a coherent complaint. There is essentially no difference between an internal DSL and an API.

The focus is on making the examples look pretty.

I don't see a lot of evidence that this is true.

I’m all for making examples look pretty, but when you focus on that to the exclusion of the underlying semantics, what you get is the Cards Against Humanity effect all over again. You’re not doing good API design – you’re doing something that has the appearance of good API design, but is in reality horrifically non-modular and will break in catastrophic ways the first time something unexpected happens, but the lovely syntax lets you pat yourself on the back and tell yourself what a good API designer you are.

That's a nice generality, but where is the evidence or argument supporting the idea that this is somehow more common of a problem in Ruby than other languages?

The way this problem manifests in Haskell is in how incredibly clever it makes you feel to get something done in it.

I don't feel particularly clever getting something done in Haskell. I occasionally am amazed at how succinct the solution is compared to what it would be in some other language, or how in solving a particular problem I've also created something that could be reused that would solve a more general class of problems, or how I didn't have to use boilerplate template code with fill-in-the-blanks to solve a particular case of well-known general problem as I would in many other languages, but was instead able to leverage an existing more general solution because of Haskell's more powerful abstractions.

But that's more about thinking that the designers of the Haskell library and language were clever than that I was. And its part of the process of learning new languages/libraries that have different strengths than those you are familiar with. I remember doing similar things (different details, of course) with pretty much every language other than my first.

spoiler• 11 years ago

I'd like to chip in about Ruby's API/DSL designs.

Most DSLs (at least the ones I've worked with, or made), are simply a convenience, and eventually to make the code more expressive.

Behind this fancy DSL, there is an API, the DSL itself isn't really an API, it's just a "frontend" for that API.

I'll pull a simple and shitty example out of my ass[1]:

    game = Game::CardsAgainstHumanity.new do
      fun 9000
      rounds 10

      must_involve_drinking!
    end

which could be a DSL for doing something like:

    game = Game::CardsAgainstHumanity.new()
    game.configuration(:fun_level, 9000)
    game.configuration(:round_count, 10)
    game.set_game_flag(:drinking, true)
    game.set_game_flag(:drinking_optional, false)

Like I said, this is a shit example. The API behind this could be way more complex and allow for much greater fine-tuning of stuff than the DSL would, but the DSL exist for stuff that's common.

Also, one would have to go out of their way to hide the actual API behind the DSL, which is very unlikely and due to Ruby's nature, one could always expose the API anyway (through the C API, as a last resort).

Also, I'd like to add little bit of something about the Haskell part, even though I've never written one line of Haskell, but this isn't really about Haskell:

"Look, I used a monad! And defined my own type class for custom folding of data! Isn’t that amazing?“. “What does it do?” “It’s a CRUD app”

I bet everyone did the same or something of that sort in <pick your language> when they were starting out with it, or exploring some part of it that they were previously unfamiliar with. This is excellent, actually. Why? Because they'll learn the benefits and pitfalls of that approach. They'll be wiser and better programmers tomorrow because of that.

--- [1]: No pun intended. Also, not looking for a real example out of principle, because I feel like the author was too lazy to do some research; this is just my opinion. Frankly, this feels like click-baiting.

heurist• 11 years ago

I don't enjoy CAH either. I have had some incredibly funny people in my life before and was able to play off their antics to have genuinely fun and creative conversations. CAH is that without the creativity (which is where the fun comes from for me). My current social group loves it so I have to sit through games occasionally... I can stave off boredom by finding combinations with as many layers as possible, but people usually don't notice. I think my preference is more geared toward improv, so I'm considering trying that.

I don't know Haskell but I've started trying to learn it a couple times. Haskell is close to how I think about problems, not sure how close yet, but I can tell it would be a great tool if I took the time to learn it. I'd probably learn a lot about how to think about problems if I did use it, and that's probably what happens for other people. It's such a major paradigm shift from the 'usual' methods that it's raised to an exalted status. If you try to solve problems without using state and with static types you can get to almost the same place regardless of language, it's just harder to implement. I've been working in Java 6 with a team firmly rooted in OO the last few years and often wish I had monads...

karmakaze• 11 years ago

This ties in with other things I've heard about Rubyists. DSLs are a thing, a good thing, but recently mistook for superficial syntax and not deep constructs. A bunch of parens interspersed with letters is a good DSL if their component meanings follow sensible lines. On the Haskell side, I used to bump into it, but seems much less so.

On Haskell, Ruby, and Cards Against Humanity

Related Stories