todays bird
Andulka
"I'm Dorothy Gale from Kansas"
Stranger Things
NASA
Jules of Nature
tumblr dot com
祝日 / Permanent Vacation
No title available
cherry valley forever
RMH
Janaina Medeiros
@theartofmadeline
No title available
wallacepolsom
oozey mess
pixel skylines
Show & Tell
he wasn't even looking at me and he found me
dirt enthusiast
seen from Russia
seen from United States
seen from Netherlands
seen from Türkiye
seen from Sweden
seen from Russia
seen from Netherlands
seen from Italy
seen from Norway
seen from United States
seen from Singapore
seen from Canada
seen from New Zealand
seen from Indonesia
seen from Australia
seen from United States
seen from Türkiye
seen from Netherlands
seen from Indonesia
seen from Sweden
@thefunctionalclub
Lispy Love
My friend Eric's got a Introduction to Clojure video project on Kickstarter at the moment.
He's nice, and he's the editor of the Clojure Gazette, so if your Sunday is going swingingly and you're feeling in a good and generous mood...
http://www.kickstarter.com/projects/376627045/lispcast-introduction-to-clojure-videos
Last Year
In all the excitement of the holidays and New Years, I completely forgot to post an update on the Fun Club meetup we had at the end of 2012. It was quite a good one too, as we joined forced with the Berlin Haskell User Group (Berlin Hug).
What happened?
It was a while ago now. But I think I remember. Florian Hartwig kicked things off by talking to us about Haskell optimisation and performance. Laziness is a beautiful but tricky business.
Then we moved on to talking about functional data structures, with a marathon series of 3 talks.
First came Alesha Summers who explained how Santa Clause could make sure to manufacture the right sorts of toys for Christmas with the magic of regression trees. Regression trees, are a type of tree often used in data mining to categorise data. Here are Alesha's slides from the evening.
Then came Frederic Kettelhoit, who did a good job of explaining what a functional data structure is, and talked us through the particular example of a finger tree. Here are Frederic's slides.
Last, but not least, Joel Wright, told us all about zippers, and also managed to nicely tie the evening together by explaining how he'd benchmarked and tweaked his zipper implementation to make it performant too. His code is here.
Link Love
Andrea Di Persio has set up a handy collaborative list of links relevant to the evening. There's some interesting stuff in there! http://urli.st/iWP
Apologies and Thanks
Many apologies for not putting this up until now. Better late than never I guess.
And many thanks to everyone who came along, and also to Berlin HUGs for sharing with us, it was really nice to do it together :-)
First for 2013
Our first Fun Club meetup of the year is scheduled! Following on from our Polyglot Puzzles meetup, Lars Peterson is organising another polyglot programming evening, only this time it'll be a bit more tough - oh yes, those fizzbuzz days are over - but don't worry, because we're giving you time to prepare in advance.
What's the plan?
Lars has devised a programming task, and your challenge, if you choose to accept it, is to program a solution, and then tell us all about it. You can pick your language.
If you'd like to contribute then please email [email protected]. If you'd rather not contribute, but would like to watch, then please sign up here.
When and where?
This Thurs (Feb 28th), starting 19:30
Generously hosted by SoundCloud @ Rosenthaler Strasse 13, 10119 Berlin
There will be beer, and pizza.
Our very own Berlin Geekette Alesha is pursuing a career in Business Analytics/Business Intelligence and looking for a new challenge. Her love for data is inspiring, courageous, and contagious all in one. And so, in her own words, we’d like the world to meet Alesha Summers:
...
W000t! Alesha's been coming to the Fun Club since it's started, and it's so lovely to see her featured in an interview :-D
FizzBuzz
Last Thursday we had a Fun Club meetup, and the theme was Polyglot Puzzles. This involved sitting around on laptops, solving easy problems in the language of our choice, pasting solutions to piratepad, and then telling each other about them.
It was pretty fun to see everyone's code popping up on the piratepad - you can see what we did here. Although Haskell was a clear favourite choice of language, we also had solutions in Erlang, Scala, Clojure, Common Lisp, Ruby, Python, D, Java, Mozilla's Rust, CoffeeScript, SmallTalk, C++, and J.
Also we had solutions is a functional language that's new to me, called Koka. Koka is a language that's been developed by Microsoft Research. It's a function-orientated language with effect inference, and it certainly seems to take an interesting approach to the whole awkward squad thing. You can read more about it here.
In the end we only had time for 3 problems, and I think the general consensus was that they were a bit too easy, and people would prefer harder problems, and a strict time-limit, so we can fit in more.
We had:
1) fizzbuzz
2) count backwards from 200 to 1, printing numbers divisible by 3
3) check if a string is a palindrome
The format of the evening was a bit of an experiment, but I think with a few tweaks we will be repeating it again sometime. So thank you everyone who came, it was delightful to get to share all your puzzle solutions. And also thank you again to SoundCloud for hosting us, and providing us with drinks.
Now, to get excited about this Thursday, because it's the next Berlin Haskell User Group meetup, and there's going to be sessions on lenses and performance optimisations.
Believe it or not, they're doing fizz buzz.
Super Haskell eXchange!
Had so much fun at the fist ever Haskell eXchange. Hosted by Skills Matter in a quite gorgeous crypt (not as creepy as it sounds).
So let me tell you all about it. Well first off, there was breakfast and a welcome from Neil Mitchell, but unfortunately I completely missed that bit, because I failed to get out of bed.
Front-End Language Features, Simon Peyton Jones
Next up came Simon Peyton Jones, utterly warm, charming and energetic. During the course of just one talk he somehow managed to convey what Haskell was about, give an idea of how the language developed, and also what's being worked on for the future, all whilst keeping everyone laughing.
For me the loveliest part of this was to find out about the things for the future, because they seem genuinely useful, and because it's nice that Haskell, even at it's relatively grand old age, is continuing to be a play ground for ideas about how to make this thing called programming even better.
[video]
High Performance Concurrency, Simon Marlow
Simon Marlow told us all about his async package. Not only was this a useful a practical guide to a package that would probably come in handy for all of us (it already has for me), but also a really great example of how a really well thought through libraries can help developers, and really give a language power.
[video]
Making EDSLS Fly, Lennart Augustsson
Okay, I admit, I didn't pay good enough attention to this one, as I was playing on my phone. Lennart Augustsson showed us an example of a DSL language implemented in Haskell, that is compiled to code that runs on a VM (specifically the LLVM), so that it can run super fast. I recommend watching it for yourself.
[video]
Scalable Web Applications with Yesod, Blake Rain
In this talk Blake Rain took us through a fairly comprehensive whirlwind tour of the Yesod web framework. If you're thinking you'd like to build a website in Haskell I'd definitely recommend watching it.
I couldn't help but think during the talk however, that maybe Yesod is trying to do too much including; webserver, templating for css, html, and javascript, stuff for chatting to sql, composable server-side widgets. It feels like a tighter focus might be more natural.
Also Yesod achieves all this with an awful lot of template Haskell magic, so that you can write some html, in a similar way to the way you write normal html, only the amazing Haskell compiler is checking through everything you do. Now I love the idea of getting the Haskell compiler to help me with this stuff, but the template Haskell stuff is a bit scary, because it's quite confusing, and I guess one of the things I love about Haskell is its transparent and concrete nature. Apart from this I worry about what would happen if your website miraculously makes some cash, and you can hire a front-end dev, they're not going to be too keen to work with all this stuff - it might be technically scalable but maybe not so organisationally scalable.
Having criticised Yesod's approach, however, I think that they're up to some really interesting stuff, and Blake did a fantastic job of covering an awful lot of content.
[video]
Cloud Haskell, Duncan Coutts
Here Duncan Coutts told us all about the Cloud Haskell implementation that he's been working on. Cloud Haskell is a library that brings cool Erlang distributed stuff to Haskell. It looks like super fun to play with, now to think of an excuse!
[video]
Integrating Haskell using AMQP, Rob Harrop
Here Rob Harrop offered a practical solution to all those Haskell lovers who can see how Haskell could be so useful at work, but don't feel able to use it. He took us through an example where he carved off a piece of important functionality from a Ruby Sinatra app, and implemented it in a Haskell server. He then stuck the two together with the sticky-tape of message queues, specifically RabbitMQ. Also, for good measure he threw a node.js middle man into the mix, but that was to show us it was possible, rather than because it was needed.
For me the demonstration was very useful, as the technologies at my work match up well with the ones he was using, but apart from that it was quite joyous to see a representation of the varied and polyglot world that development is, and how different languages can all get along together.
[video]
Park Bench Discussion
Aided by a break for beer and pizza, the park bench was a fun occasion. I was struck by the openness of everyone involved, people genuinely encouraged others to get up on the bench too - this sort of generosity was especially lovely coming from people like Simon Peyton Jones. Topics were varied, including talk of whether if Haskell had it's time over again it would be lazy. Some discussions centered on how support for Haskell can transition more from it's traditional academic community, to a wider community of developers.
[video]
Then we all went to the pub.
It was a lovely day, and I think my favourite conference yet. The line-up was just amazing, sessions were useful and informative, and the feeling was friendly, and open. It feels like Haskell is all grown up and ready for the real world now!
Haskell eXchange 2012
Good heavens above, I'm getting all excited because I'm very luckily going to the London Haskell eXchange 2012, on October 10th.
What a line-up, 2 Simons! I'm a little worried I'll just get star-struck and forget to listen to anyone, and turn into an incoherent puddle on the floor.
Also, the next Fun Club meetup is scheduled for Thur 18th Oct. It's called "Polyglot Puzzles", and we're going to be all solving simple programming puzzles in the language of our own choice, and showing each other the results.
You can do the puzzles in whatever language you want, it doesn't have to be functional; it's just nice to learn by comparing.
Sign up here.
Functional at GeekGirlMeetup
Wondering why The Fun Club hasn't really done anything recently? Well, there's two reasons:
* I've been really busy organising a thing called GeekGirlMeetup, which is a one day female only tech conference.
* Markus is in Scotland, and discovering the joys of British teabags and the sorrows of British bread (hope you find a good bakery soon Markus).
But don't worry, at the GeekGirlMeetup we kept the functional flag flying; we had two functional talks. The first was from myself and was a basic introduction to functional programming. This was followed by a fantastic talk from Monika Moser all about Erlang. I'm totally inspired to head along to the next Erlounge meetup.
Right, so now that's all done with, time to cook up something fun for The Fun Club!
Talking about Fexprs
On Wednesday the Fun Club was talking about Fexprs. I don't think about of us had a clue what a Fexpr was before the evening, but thankfully Tim Felgentreff does, and so he told us all about it!
It turns out Fexpr were a sort of evolutionary dead-end, that had been in Lisp up until about 1956 (I think I remember that date correct), but although powerful and simple, but were a nightmare to optimise and ran like a dog, so had to be dropped.
The evening was generously hosted by Tim and Markus's employers finnlabs.
Thanks ever so much Tim for such a lovely talk.
Fexpr
I don't have a clue what an Fexpr is, except that's it's not an S-expression.
Wikipedia says:
In Lisp programming languages, a fexpr is a function whose operands are passed to it without being evaluated. When a fexpr is called, only the body of the fexpr is evaluated; no other evaluations take place except when explicitly initiated by the fexpr.
Hmm, I'm still a bit confused. Anyways, not to worry because Tim Felgentreff is going to explain all on Tuesday 14th Aug. Please join us!
I did a work swap with a designer and got The Fun Club an official logo, and here it is.
For my side of the swap I made a Wordpress theme, and there was PHP involved; I still feel a tiny bit dirty. Nah, just kidding, the designer Jian is a pleasure to work with, and I enjoyed myself.
Anyways, I think the logo's turned out pretty good!
Btw, our next meetup is on f-expressions, and is on Tuesday 14th August; come join us!
Learning Functional Programming with Scala
Through the London Scala group, just found out about a 7-week course on functional programming from Martin Odersky and Coursera. Most importantly it's free and online!
https://www.coursera.org/course/progfun
A way to look at monads. Btw, you can buy this necklace here.
Monad Self Help Group
Last night we had our second meetup of The Fun Club. We were generously hosted by SoundCloud; many thanks to Phil and Eliot from SoundCloud for that.
The evening was all about monads. There are thousands of blog posts and tutorials about monads out there, and yet somehow the concept still remains quite a difficult one to get to grips with.
The plan for this evening was for people who understand monads to give a short explanation on how they see them, so that together we could swap insights, and help those who are just learning about monads to get to grips with them. We called the evening a "Monad Self Help Group".
Now in the event, it turned out that it was just me and Markus who had prepared a monad explanation. But that was okay, because we half suspected this would happen, and prepared extra material just in case!
First up was me. I decided to explain monads using necklaces as an analogy. This sounds like a terrible idea, but I think it worked out okay in practise. Here are the slides. The idea was to think of monads as a programmable chain that you can use to string together pieces of code, like beads on a necklace chain. I then worked this analogy through with a concrete example of validating user input in Haskell using a home-made monad, that was very similar to the Maybe/Option monad.
Next up came Markus. He had prepared the code for a numeric expression parser in both Java and Scala. First off he went through the Java code. This was a little verbose, and kind of icky, because it had to make use of Exceptions to stop parsing. Then he took us through the same thing, but in Scala. He had defined an object called Step, which represented a step in the parsing, and could either be a Reject or an Accept. He also impemented a flatMap function (a.k.a. >>= or bind) such that Rejects stop parsing. Using this monad his Scala implementation was significantly more shorter and more readable than the Java implementation. The code is on GitHub https://github.com/machisuji/about-monads.
Now, because no-one else had prepared a talk I thought it might be nice to discuss how we came to understand monads, so which resources helped the most etc.
I personally found that going through functors and applicatives first, helped a lot, and also the book Learn You a Haskell (free for online reading). Others agreed that learning going through functors and applicatives is helpful, and that the Haskell Typeclassopedia and also the Scala version (can't find a link!) are especially good for this.
We also talked a bit about whether unix pipes, or perhaps even the unix 'tee' command were monads. (I think the conclusion was that they're sort of like instances of monads, but you can't program a pipe, so not really.)
Meanwhile, inspired by our talk of functors and applicatives, the marvellous Piyush from the Scala User Group coded up some Scala implementations of them for us, and after a quick beer and pizza break (thanks SoundCloud), opened up an interpreter and took us through them step-by-step. Lots of thanks to Piyush for doing that.
Then it was 10pm, and we felt sleepy and went home.
Thank you everyone who attended and made it such a delightful evening.
Next time, it looks like we're going to be looking at S-Expressions. I had no idea what these are, but Wikipedia says:
In computing, s-expressions, are a notation for nested list (tree-structured) data, invented for and popularized by the programming language Lisp.
Sounds like fun!
FP Days 2012, Call for Speakers
FP Days 2012 is currently calling for speakers. FP Days is a functional programming conference, held in Cambridge (UK), on 25th & 26th October.
They say
"We're looking for great speakers to share their practical experiences of Functional Programming. Accepted speakers pay no conference fees. You don't have to be a big name and we encourage and support first-time speakers willing to share their experiences of FP."
Deadline for speakers is Friday 6th July.
If you fancied going, volunteering to speak would be a great way to cut costs!
