Douglas Crockford: Monads and Gonads (YUIConf Evening Keynote) 

I watched this video with already an idea of what monads are, and so I may have a different perspective than others about this. This being a more than 10-year-old video also changes how it sounds when watching it. I don't think this is a boring presentation. It is quite an old-fashioned way to present perhaps, but it does provide code examples. I can see what I have learned in Kotlin about monads completely reflected in this presentation even in current times and that makes it a good presentation in my view and I am very happy to have watched this.

Watching this in 2018 feels like this talk was given decades ago :'D

Well, the curse is real.

This is the problem with every video I've seen about monads. They all start with: 'We're going to do a better job at explaining monads than every other video on youtube.', but then proceed to make the same fundamental mistake. Instead this is what they should do: 1) show a concrete example of the problem sttement, in the language that the viewer already knows, 2) show the viewer what the typical way to solve the problem, in the way that the reader would likely solve the problem on their own, 3) now tell them what the downside is with their solution and finally 4) show the viewer how monads solves the same problem without the downside.

this man is a prophet - brings the message for the future, and only few can hear him.

Excellent explanation. For the part where Crockford discusses the MayBe monad, and passing that into the MONAD constructor (mins. ~28) he left out a critical code slide to actually illustrate that the MONAD receives such a modifier parameter. It would look something like var MONAD = function (modifier) ... further down if (typeof modifier === 'function') { modifier(monad, value); } in order to actually invoke it. Otherwise very clear and simple explanations.

Suggesting a slight change. The coloring should distinguish scopes. For example unit and value exist in different scopes ( at 18:00 ), but they're the same color in Doug's example.

Weird reasoning; "coloring is for kids, I'm a grown up...but I would like to have coloring..." If there was something like a "ReasoningLint", this reasoning would not pass ;-)

Someone please build Douglas Crockford's context coloring in vim. I'll give high fives!

Look! Someone did build it. Search for "JavaScript-Scope-Context-Coloring" on Github (yt won't let me post an URL)

Context coloring: github.com/jacksonrayhamilton/context-coloring

the intro history stuff about Fortran kind of resembles Ruby. no curly braces, usage of variables without declaration, etc.

At 51mins, no other language that can do that ? Man you can do that in D and it is dead easy ! I'm pretty sure that many other languages allow that as well.

Interested in the Actor Model: channel9.msdn.com/Shows/Going+Deep/Hewitt-Meijer-and-Szyperski-The-Actor-Model-everything-you-wanted-to-know-but-were-afraid-to-ask ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-7erJ1DV_Tlo.html

1.) The monad is the structure relating the objects, not the objects themselves as he asserts. 2.) His Maybe "monad" isn't the Maybe monad. Maybe can usefully distinguish between Nothing, Just Nothing and Just (Just a), but introspects on the values it is given breaking parametricity and cannot. This has big implications w.r.t. storing nulls in containers. 3.) He spends much of his time arguing against some straw man who says you must learn all of category theory to learn how to use a Monad.

Not to mention Boost has primatives that will perform parallel map over vectors also.

"Go to yahoo and then google for Monad"... hehehe I love how "staple" has become Google for our society

Agree: this statement is wrong. parMap in the widely used `parallel` package in Haskell exists for exactly this purpose.

Is the enqueue function supposed to have a return statement? return queue?

FTR, The IO "loophole" in Haskell is more rigorous than described. E.g. you can't just do IO anywhere you want.

I have attempted to take his code and put it in a library and extend it to be more useful. github.com/scottburch/functional I think it has a simpler explanation since I don't go into the details of the code itself. Any feedback is welcome.

Has the ...args "splat" feature been implemented in newer versions of ECMAScript, then? (Or maybe he's joking?) Edit: He's talking about it around 24:11

You can use the ideal file system API with promises right now. npm i fs-promise

Memory consumption really is an issue in Javascript. Especially when creating monads which necessarily close over vast number of closures, hamstringing the garbage collector. This will very quickly become an issue on long running client-side apps and especially server applications which must service hundreds of thousands of customers within the same "huge" ram space.

The talk is great. The Q&A is not worth the time, tho.

What does he say at 20:00? Eke mono?

OK I understand monad .. and I can not explain what it is!

"go to yahoo and google for it"

Fully agree.

first time i saw a talk from crockford i thought he just screwed up :P.

Anybody build that context colorer yet? lol

.. it is still cursed, obviously (:

"Ajax Mondad"? Seriously? That's Method Chaining.... I don't want to live on this planet anymore....

really interesting video. It would be even better without the stereotyping of Spanish speaking people. You can learn Spanish to talk to business partners, to talk to collaborators across the world, not only to "your workers". You were just shy from saying "you can learn Spanish to talk to your maid or your gardener." Just saying.

Ladys look away, DOG BALLS! 38:20

Not sure if Crockford is trolling...

47:50 "Trying to save memory in Javascript... I can't imagine a bigger waste of time" I can, though. Sure, "premature optimization is the root of all evil", but a lot webapps, sites, and HTML5+css+js applications seriously need some love. It's both lazy and ethically questionable to keep pushing all the load to hardware requirements. Lazy because part of your job is to make things that work as intended now, not things that will probably work as the should in 2 years. Ethically questionable, because you put an arbitrary economic barrier to access. "Doesn't work in your computer? Buy a better one". Why? It's just a stupid gallery page that should work on any 10 years old computer! What that way of working does is to push the cost from the company making software to the user buying hardware.

that doesn't bode well :p

LOl!

Awkward distracting stereotype at 11:50

I couldn't listen to the full talk: he is too annoying..

I really hope this smug, sloppy presentation style doesn't catch on. Crockford may have had something useful to say about implementing monads in Javascript, but it is lost in a fog of arrogance and misrepresentation.