More about threading

Wednesday, August 22nd, 2007

It is always interesting to get alternative views on the state of the world. For example on the need of developing multithreaded applications. I wrote last about it January 6th, 2007 in entry The Core Threading software problem.

Now I found (via) a video presentation at Microsoft Research by Edward A. Lee which presents his view on the core threading software problem. His statement is that the current threading programming mechanisms essentially are flawed. Programmers simply do not understand what is going on, leading to thread bugs, e.g. race conditions, deadlocks etc.

I suggest you watch the video and you should arrive at a understanding of the conclusion (my text in italics):

• The current foundation
  • Transformation of bits is the foundation of computation (programming).
  • Threads are the foundation of concurrent programming practice.
• The foundations are flawed
  • Threads discard the most essential features of computation.
  • Threads are incomprehensible to humans.
  • They appeal because they make (almost) no changes to (programming) syntax.
• Hence - concurrent computation needs a new foundation
  • Actor oriented component models.
  • Coordination languages with actor semantics.
  • Visual syntaxes to seduce users (programmers).
  • These have a change of acceptance!

Links:

• The Ptolemy Project (and for curiosity Ptolemy at wikipedia)
• Edward A. Lee's website