BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Julia: A Fast Dynamic Language for Technical Computing - Stefan Ka
 rpinski\, MIT
DTSTART:20131202T130000Z
DTEND:20131202T140000Z
UID:TALK48724@talks.cam.ac.uk
CONTACT:Andrew Rice
DESCRIPTION:\nJulia is a high-level\, high-performance dynamic language ca
 refully designed to be excellent for technical computing. It has been gain
 ing traction as a an alternative to Matlab\, R and NumPy – especially in
  performance-critical areas such as machine learning\, "big statistics"\, 
 linear algebra and image analysis. One important feature that distinguishe
 s Julia from other systems – even general purpose languages like C or Sc
 heme – is that it doesn't special-case numerics at all. Basic types like
  integers and floating-point numbers are actually user-defined\, with layo
 ut and behavior defined in Julia – they just happen to be defined before
  your program starts. As a corollary\, your custom data types – be they 
 exotic kinds of numbers or standard data structures – will be just as fa
 st and efficient as Julia's builtins. We'll explore the language design ch
 oices and implementation techniques that enable this\, including pervasive
  dynamic multiple dispatch\, run-time code generation with aggressive spec
 ialization\, and dataflow-based type inference. There will be lots of live
  coding and demos: from creating custom numeric types to analyzing and vis
 ualizing data. The talk will wrap up with a discussion of how we're pushin
 g the envelope of open source by making it unprecedentedly easy for people
  to contribute to the language itself\, its packages\, and its ecosystem.\
 n\n
LOCATION:LT2\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR