School of Haskell Goes Beta

http://fpcomplete.com/school-of-haskell-goes-beta

by dons • 13 years ago

179 69 13 years ago

hosh• 13 years ago

Lady's Illustrated Primer, ha!

I've been working my way through "Learn You a Haskell For Great Good". I worked my way up around functors and monads. Then I stopped for a while. This new School will be interesting :-)

tunesmith2• 13 years ago

Yeah, this will be interesting. I've also been going through Learn You A Haskell, making an Anki deck as I go. Slow work, but Anki will help me retain everything I'm learning.

ehamberg• 13 years ago

Slightly off-topic: How much do you break down concepts when using Anki to learn a programming language? I use Anki for tonnes of stuff, but haven't used it for learning programming languages/APIs.

monkeyfacebag• 13 years ago

As excited as this makes me, I am way more excited to hear about the "full blown Haskell IDE" they mention. Is this a cloud service?

efnx2• 13 years ago

It may be leksah, which is currently available on hackage, though I've never been able to get it to compile (I haven't tried that hard)...

mlinksva• 13 years ago

Not likely leksah http://fpcomplete.com/designing-the-haskell-ide/

FPguy6• 13 years ago

FP Complete is currently developing a Haskell IDE which will be exposed as a cloud service.

Gregg Lebovitz talks about it here: http://www.youtube.com/watch?v=IM_OSykXKxM

skwosh2• 13 years ago

Hope this goes into a lot of depth...

My main issue with existing books and tutorials is that I'm left in the dark about the more interesting/advanced parts of Haskell.

Some of the areas I would like to have a better understanding:

- More advanced kinds of monads e.g. Logic, Continuation

- Monad Transformers

- Arrows

- Rank-N types

- GADTs

- Category theoretical ideas e.g. Bananas and Lenses

- Type derivatives and TypeClass abuse c.f. Conor McBride

- Control patterns like Iteratees, Generic Zippers c.f. Oleg Kiselyov, Chung-Chieh Shan

I have an intuitive grasp of the above, but to me Haskell is more about programming with types than anything functional.

On the practical side, it seems like Haskell excels as a language processor, and parsing/compilation would be a great way to explore how to structure certain kinds of applications (like web servers, graphics pipelines, etc).

Please don't make it too real-world, I already have bash ;)

dysoco• 13 years ago

I'd say you should take a look at research papers for those topics: But it's true, we are lacking advanced Haskell books.

skwosh2• 13 years ago

For me, there's a pretty big gap between the available educational materials and research papers. My understanding of the academic stuff is mostly intuitive. It isn't until I read a paper like Monadic Parser Combinators [1] or sigfpe's blog [2] that it really becomes concrete.

I find that for a lot of higher-level ideas much of the heavy lifting isn't necessarily through application of functional ideas per se, but through more sophisticated type gymnastics.

Because expressions are often written in short-hand (to expose structure), as well abbreviated and underspecified (to get the compiler to do the work), making sense of academic code can be a challenge.

In another life I'd have dedicated more time to learning the fundamentals (of Programming Language Theory and Category Theory), because I find the whole field insanely facinating. Maybe I'm looking for a short-cut but there isn't one?

[1] http://eprints.nottingham.ac.uk/237/1/monparsing.pdf [2] http://blog.sigfpe.com/

josh-j• 13 years ago

If you're reading this and understand those areas well, then consider donating some time towards improving the advanced track on the Haskell wikibook (http://en.wikibooks.org/wiki/Haskell).

gtani3• 13 years ago

Everybody has a different list of a dozen intermed/advanced topics ;-}

http://www.reddit.com/r/haskell/comments/17gavl/what_you_con...

http://stackoverflow.com/questions/5778436/what-haskell-topi...

And most of those are fast moving topics these days but they're well covered in blogs, the Cafe list, stackoverflow, they just need to be collated/cataloged. At a minimum some wiki farming would address a lot of those. Unfortunately, I emailed the wiki contact a couple times about pitching in and never got response.

___________________

There's a few books you could look at, tho they're more building out what's in RWH, not up:

draft by Snoyman https://github.com/mezzohaskell/mezzohaskell

Bird's Functional algorithm pearls http://www.amazon.com/Pearls-Functional-Algorithm-Design-Ric...

http://haskell.cs.yale.edu/wp-content/uploads/2013/01/HSoM3....

JA Alonso has another book of exercises (in Spanish http://www.cs.us.es/~jalonso/publicaciones/Piensa_en_Haskell...

also the FP in Scala (Morris, Chiusano, Bjarnason) may include some haskell code, I haven't read yet, but definitely covers a lot of these hot topics

http://manning.com/bjarnason/

eru7• 13 years ago

> Unfortunately, I emailed the wiki contact a couple times about pitching in and never got response.

Have you tried contacting the Haskell cafe instead? I am sure someone there will be able to get through.

FPguy6• 13 years ago

We welcome people who have written (or plan to write) Haskell tutorials and exercises to post them on the School of Haskell, if they find its features useful. By now you've probably seen the Active Haskell demo in the beta screencast video? https://haskell.fpcomplete.com/beta

We also welcome good organizers (like yourself perhaps?) to write tutorials that consist partly or entirely of links to other materials. The role of editors and organizers is sometimes undervalued, but not by us!

eru7• 13 years ago

About the top of your list: don't worry about Monads. Learn Applicatives first and foremost.

tel2• 13 years ago

Why do you say that? Applicatives are perhaps easier and open up Foldable/Traversable, but Monads are just everywhere. As the de facto default functor sequencing method, I'd recommend people start there.

yarou• 13 years ago

Signed up. :) This is my nth attempt at learning Haskell, but I have a fairly good handle on most of the basics.

hexonexxon• 13 years ago

alas waiting list

doubt i'll switch from scheme but still interested

gtani3• 13 years ago

I think it was opened up again, I got in

FPguy6• 13 years ago

Sorry about the short wait -- we've had literally thousands of applications in the first day, and we are letting people on by groups of 100 (and that step size itself will increase).

Ixiaus3• 13 years ago

Any Haskell web-application framework is awesome - Yesod in particular as it has the most active community and written material for it.

I've lately been working on an (unpublished) project using Haskell+Yesod+Fay+Clay and it's remarkable how productive I am; in that most of my time is spent figuring out my types, writing the code, then...it's done? In Python I run into a lot of programmer related bugs that Haskell's rigid type system prevents. It can't prevent logic/flow "bugs" but it does keep a lot of other bugs out that would normally have me spending potential future hours fixing/debugging.

It's also not just the type safety that increases my productivity, it's also...Haskell. Abstraction in Haskell can make for very concise and correct programs.

Fay and Clay are particularly awesome too.

Haskell definitely has a learning curve, there was a steep curve for me and I already had significant functional programming experience (Erlang & Scheme). I'm really happy to see these guys taking that on!

pekk3• 13 years ago

The correct way to use Python is to write unit tests. Don't blame your tools when you don't know how to use them correctly.

batgaijin• 13 years ago

Thank you for your time, enlightened one.

efnx2• 13 years ago

What it comes down to is that Haskell's type system eliminates whole categories of bugs because of the language. Having the correct way to use the language outside of the language itself, as an addition, in the case of Python or Ruby or (insert interpreted language here) makes for lots of extra debugging time.

chongli• 13 years ago

In other words: do the job that the compiler would have done for you if you'd been using Haskell!

DanWaterworth• 13 years ago

Types and tests give completely different assurances. Tests give you assurance that a particular code path works at least some of the time. Types give you assurance that an invariant is true all of the time.

gnuvince• 13 years ago

Really hoping to see that application, I've been meaning to pick up Yesod, but school and all that are getting in the way.

laureny• 13 years ago

> In Python I run into a lot of programmer related bugs that Haskell's rigid type system prevents.

Isn't your productivity gain coming from using a statically typed language rather than Haskell specifically? It feels like any statically typed language is a leg up from Python :)

freyrs3• 13 years ago

> Isn't your productivity gain coming from using a statically typed language rather than Haskell specifically

I think it's a bit more than that, Haskell has a much more sophisticated type system than quite a few of the popular statically typed langauges ( C, Java, C# ). Static typing is only the start, to be really useful we need modern type inference methods and a concise way of expressing generics. Both Scala and Haskell do this very well.

mightybyte• 13 years ago

Purity is a big factor too, and Haskell is the only pure language I know of that is mature enough for production use.

codewright• 13 years ago

Are we still spreading that canard where type errors are somehow some kind of bugbear in Python?

It's not a profound part of writing code in Python for anybody that's spent more than a few weeks with it.

merijnv• 13 years ago

It depends on what you mean by statically typed language. If you mean other stuff like SML/Ocaml, I agree. If you include C/C++/Java/C# in your definition "statically typed", then I beg to differ. I have considerable experience in C, Java and, to a lesser extent, C++. And all in all I consider their type systems worthless for error prevention, the flexibility of python gains you more than their type systems do.

In this sense I suspect the OP would say the same about any ML-like static type system, but not many other static type systems.

cies• 13 years ago

in though statically typed meant C++ or Java, and probably it does.. i never felt compelled to the way they do typing, so i gravitated towards Ruby and Python.

but to me Haskell is a different league.. Haskell-typing actually helps /me/, where i feel C++/Java-typing merely help the compiler.

minikomi2• 13 years ago

I would love to see a "from nothing to web app" style piece of writing on this. It sounds intriguing. Blog, paid videocast, book, set of tutorials.. If you're interested in doing one, perhaps as a noob I could provide some "that's hard to grasp" style feedback and collaboration (mail in profile hinthint).

melling• 13 years ago

Kickstarter project? Chris Coyier did a great set for his site redesign, for example.

http://www.kickstarter.com/projects/150422311/screencasting-...

s_kilk• 13 years ago

I've been toying with the idea of writing a Yesod equivalent for the excellent Rails Tutorial by Michael Hartl. I need to get a bit more comfortable with Yesod first however, so it may take a while. I do think that is what the Yesod world is in most need of right now, a "get stuck in" tutorial that takes the reader from zero to a substantial web app.

Ixiaus3• 13 years ago

Great idea; if I find the time I'll think about putting something together. Good to know there is interest in this.

FPguy6• 13 years ago

Thanks. In fact we used Haskell + Yesod + Fay to build the School of Haskell itself.

bpolania• 13 years ago

I learned Haskell in my first algorithm course in the University, mine was the last class ever to do it in that university, the following semester they switched completely to Pascal and later to Java.

These are good news, I just can't wait to see that IDE.

shurane• 13 years ago

Isn't switching from Haskell to Pascal or Java a huge switch? That's a change in thinking style, even!

maximveksler• 13 years ago

HN never fails, just as I was browsing the web after reading pg assay about programming languages and the story of how he built what is now yahoo shop using lisp and deciding that instead of lisp I will learn Haskell this comes along. Looks very good and hope that it won’t be /too/ easy start without any challenges...

jhuni2• 13 years ago

You should still learn Lisp unless you are certain that Haskell is already in the perfect form for what you want to do. Lisp gives you unprecedented freedom to transform the language into the form that suits you best.

tikhonj4• 13 years ago

That's not really unlike Haskell: in Haskell, you just have a different set of tools for transforming the language.

I just came off a largish project in Racket, and do appreciate that it's very flexible. However, I've found Haskell to be roughly as flexible in practice. Haskell has surprisingly flexible syntax and semantics, and I've found it very easy to have parts of my program with radically different semantics from normal Haskell.

You can very easily layer on things like non-determinism, logic programming, error-handling, asynchronous programming or even continuations onto Haskell code. And since it's lazy by default, you do not have to do anything special to add new control structures--you don't even need to use the macro system. (Which, of course, Racket does much better--template Haskell works, but it's a bit of a pain.)

I've also spent some time embedding DSLs in both Racket and Haskell, and have found that they are about roughly equally convenient. Racket is more flexible syntactically--it's very easy to do things like inspect variable names and the like. Haskell, on the other hand, has some very lightweight tools (laziness, do notation and a very spartan syntax) that make it very easy to change its semantics. The type system--especially things like GADTs--also makes life much easier for DSLs.

eru7• 13 years ago

Lisp (and especially Racket) is fun. But Haskell also allows preprocessing your code.

yogsototh• 13 years ago

I don't think it will be too easy. For a taste, you could take a look at my "Haskell Fast & Hard" tutorial. If you signed for the beta you can get try it here:

https://haskell.fpcomplete.com/user/yogsototh/haskell-fast-h...

If you don't already have a beta account. You can try the original non interactive version here:

http://yannesposito.com/Scratch/en/blog/Haskell-the-Hard-Way...

I had the chance to be an alpha user of School of Haskell and I copied the tutorial to it. Clearly the interactivity provide a clear benefit.

Concerning the difficulty of my tutorial, it goes like this:

1 - very easy 2 - easy 3 - medium 4 - hard 5 - very hard 6 - annex is kind of brutal difficulty

And actually, there is at least five more level of difficulty if you want to "master" some advanced feature of Haskell.

But one thing that no tutorial is able to give, is the feeling of joy when writing Haskell. It is something you experience only when doing your firsts non trivial programs.

arocks• 13 years ago

This is a very promising approach. I was blown away when they actually executed a web application and the working web page was shown inside an IFRAME. This means that I could learn programming from any device, even a tablet, without needing to install anything.

FPguy6• 13 years ago

Thanks. One of our design points in fact is that you can use the School from a tablet. We're still a bit concerned about Safari performance on the iPad, but we're on the case!

pohl• 13 years ago

I suspect that you are not running into safari's performance limits so much. I think it's more likely that your client-side code is doing something that interferes with the browser's event loop, and is thereby preventing the browser from processing ordinary operations like doubletap-to-zoom, scrolling, pinch-to-zoom, etc. If you have a lot of JavaScript operations to perform, sometimes you need to defer some of them until the next time the event loop comes around — so that the browser can "breathe", as it were.

GWT, for example, has a DeferredCommand and IncrementalCommand facility for this situation. I believe they are just wrapping the setTimeout(func, 0) trick. (See link below.) If Fay does not have such a facility, then you may need to do some upstream work to teach it how.

http://b.javascript.info/tutorial/events-and-timing-depth#as...

But I guarantee that even an iPad 2 should have enough oomph to tackle your modest needs. It does just just fine on more intensive things like Google docs, spreadsheets, reader, and plus. And Facebook.

If, for example, I have not yet asked the monads tutorial to compile and run a snippet, then the browser's JS thread should be entirely quiescent and the event loop should be responsive to my zoom, pan, and rotation requests. That it isn't implicates your code (and/or the code Fay emits) — not the browser. I hope the creators of Fay considered this while architecting it. If they didn't, it's unlikely that they would have arrived at the correct solution by accident, because ordinarily compilers never need to worry about being "a good citizen" to a browser's event loop.

Best of luck. I envy that you get to work on this!

silverlake• 13 years ago

How is this better than working through a book? If I can't download and install Haskell, then learning the language is the least of my problems.

tikhonj4• 13 years ago

The important part is to engage newcomers. This is easiest to do if there is no barrier to playing with Haskell. They can always install Haskell once they're a bit more familiar with it and see why it's awesome.

If you're already planning to learn Haskell and have a book about it, this is useful because it lets the author of the tutorial integrate the material more closely with the language environment. This makes things like little exercises to illustrate a particular point easier to do.

misleading_name• 13 years ago

just got my log in

School of Haskell Goes Beta

Related Stories