Browse files

WIP: strangeloop blog post

  • Loading branch information...
twopoint718 committed Sep 25, 2011
1 parent 313c81d commit 63d3bb0bbabb2a0e98836ad6bfcc097a1238a2e6
Showing with 134 additions and 0 deletions.
  1. +134 −0 posts.yaml
@@ -1,3 +1,137 @@
+- date: 2011-09-24
+ title: "strangeloop 2011 notes and flying"
+ postbody: >
+ I got back from [Strangeloop 2011](
+ just this week and wanted to cover some of the interesting points
+ from this *really* fascinating conference (it is on my *must go*
+ list from now on)!
+ It was incredibly difficult to get to all the talks that I wanted
+ to see because the conference was "seven talks wide" at most
+ points. A common theme emerged where, as I finished up a talk in
+ one room, I would see the stream of tweets start rolling in about
+ some incredible talk that I had just missed; I can't wait for
+ those videos.
+ Here's my recap of the stuff that I went to:
+ **Sunday (workshop day)**
+ * *Haskell: Functional Programming, Solid Code, Big Data* with
+ Bryan O'Sullivan - this was a really nice intro to Haskell for
+ someone that hadn't ever seen it before. I've worked through
+ about half of the "Real World Haskell" book so a lot of this
+ was not new. But it was great to see one of the authors
+ explain some points himself. There was also some interesting
+ comments from [Gerald
+ Sussman]( about
+ how haskell is the "most advanced of the obsolete languages"
+ (more on that later).
+ **Monday (first day of conference)**
+ * *Category Theory, Monads, and Duality in (Big) Data* with Erik
+ Meijer - This was a really cool opening keynote where Erik
+ Meijer launched the new term *CoSQL* instead of *NoSQL* by
+ showing how the two concepts are
+ [duals]( of
+ one another (in the mathematical, category theory sense). This
+ proved to be something of an overarching theme of the
+ conference, things being different but mirrored versions of
+ the same thing. see: [A co-Relational Model of Data for Large
+ Shared Data
+ Banks](
+ * *[I skipped this timeslot because I was on the [hallway
+ track](
+ listening to Erik Meijer talk about static typing with some
+ scala folks; very interesting!]*
+ * *An Introduction to Doctor Who (and Neo4j)* with Ian Robinson -
+ I have to admit, I got sucked in because I'm a huge Doctor Who
+ fan, but I *had* heard of graph databases before and Neo4j
+ looked to be a really interesting one. In particular, I wanted
+ to see if this could be used from Clojure (yes:
+ [borneo]( and
+ [clojure-neo4j]( The
+ talk concerend building a very complicated network of the
+ relationships between several Doctor Who props (Daleks!) over
+ time. It was pretty easy to see how these mapped nicely to
+ nodes with arcs between them.
+ * *Skynet: A Scalable, Distributed Service Mesh in Go* with Brian
+ Ketelsen - this was a cool talk about a lightweight framework
+ written in [go]( for writing distributed
+ applications that are highly resilient. It uses
+ [Doozer]( for data storage
+ (though it didn't in this talk).
+ * *Parser Combinators: How to Parse (nearly) Anything* with Nate
+ Young - This talk gave examples of writing parser combinators
+ (where a *parser* here means a function that can consume a
+ little input, and then returns another function that consumes
+ input after it). The idea is to chain these parsers together
+ with *combinators* (higher-order functions which take parsers
+ and operate on them, like "oneOrMore" etc.). This talk
+ reminded me of Bryan O'Sullivan's funny phrase about how
+ haskell's ">>=" operator (read "bind") is written in "moon
+ language".
+ * *Getting Truth Out of the DOM* with Yehuda Katz - This was a
+ talk about the [SproutCore](
+ framework. Katz had a lot of insight about how to keep the
+ browser interaction abstract and event-based rather than
+ mucking about (and then being mired) in the DOM.
+ * *We Don't Really Know How to Compute!* with Gerald Sussman -
+ This was a mind-blowing keynote. In fact, I had to develop a
+ new unit of measure, the *Eureka*, which denotes having one's
+ mind blown once per minute. I think that in the 50-some
+ minute talk that Sussman gave, I may have had more than about
+ 50 mind-blowing thoughts. At one point Sussman asked how much
+ time he had left and someone from the audience yelled out "who
+ cares?", which was pretty much the feeling in the room.
+ Sussman started out the talk with a picture of a [Kanizsa
+ Triangle]( and
+ mentioned that the brain can infer that there is a hidden
+ triangle in just about 100 ms which is a few tens of "cycles"
+ for the brain. With a computer, we don't know how to even
+ begin to solve this recognition problem in that few of cycles;
+ *we don't really know how to compute*. Sussman's idea (which I
+ can't do justice to here), was that computing as we know it
+ has to and will change in the near future. Computing will
+ become massively distributed ("ambient", but this term is from
+ a later talk) and in disparate nodes that must collaborate to
+ arrive at answers.
+ His
+ [example](,
+ a *Propagator* was a program that can integrate more annd more
+ data while keeping track of the provenance of that data. Or
+ another way an "independent stateless machine connecting
+ stateful cells". Amazing!
+ **Tuesday (second day of conference)**
+ * *Embedding Ruby and RubyGems Over RedBridge* with Yoko Harada -
+ * *Event Driven Programming in Clojure* with Zach Tellman -
+ * *Teaching Code Literacy* with Sarah Allen -
+ * *Post-PC Computing is not a Vision* with Allen Wirfs-Brock -
+ * *Simple Made Easy* with Rich Hickey -
+ I also decided to take a flying lesson today...
- date: 2011-08-16
title: Madison Ruby Conference
postbody: >

0 comments on commit 63d3bb0

Please sign in to comment.