Damian Conway has a well-deserved reputation as the mad scientist of Perl. He was in full mastery of his powers of madness, science, Perl, programming, and even rhetoric on Tuesday night. As usual, he closed the opening keynotes with a trek through difficult subjects (and a fair amount of Photoshop-enabled manipulation).
As Damian recounted, two of his previous talks drew richly from higher physics. 2000 saw the release of his Quantum::Superpositions module, which simulates quantum computing and the Copenhagen interpretation of quantum mechanics to produce superpositions -- a scalar which is simultaneously all logically possible values until observed, at which point it collapses into a single value in constant time. (This is a core feature of Perl 6.)
In 2002, he explored the world of general relativity, demonstrating how to travel backwards in time with Lorentz contractions. Though Aron Wall, son of eternal champion Larry Wall (as Damian explained) questioned the physics, Damian demonstrated how to get correct answers not just in constant time, but in zero time. He called this module
Then, for a long time, nothing happened.
Last night, he diverted through the work of nanotechnology pioneer Eric Drexler to discuss carbon nanotube computing, incidentally porting the technology to Perl 5 in the form of
Rod::Logic (not a source filter). Unfortunately, he declared the result a partial failure, as, in his words, "readability [is] suboptimal". According to the ancient Romans, failure is a requirement of progress, so he returned to the world of physics.
Damian chose the work of Dirac who unified Quantum Physics and General Relativity by predicting electron spin states, incidentally predicting the existence of an anti-electron, or positron. It's a short leap through the work of Carl Anderson (who claimed to have photographed positron motion in 1932) to end up at "the world's least hip beatnik", Richard Feynman, whose Feynman diagrams demonstrate the possibility that a positron is merely an electron sent backwards through time.
This is clearly a technology worth porting to Perl 5, and so Damian introduced
Positronic::Variables, which allows you to declare a new variable type, the pv, which sends its value back in time from the end of its scope in your program to the point of declaration, giving you the answer to your question before calculating it.
You still have to calculate it, however.
There was only one remaining problem, where your Feynman diagram may not actually coalesce on a single stable state. (Visualization helps.) Then again, all physics builds on previous versions -- and recent developments have shown that the Copenhagen interpretation and the Many Worlds hypothesis are isomorphic, if you look at them from the right angle... so Damian added superpositions to the positronic variables, allowing them either to travel between multiple parallel universes (and back in time) or to superpose.
If you're confused, imagine watching this... and then realizing that the punchline is that this technology also works with the rod logic module Damian wrote earlier, proving that Perl 5 can solve problems at the molecular level with the spooky magic of quantum mechanics in what may be multiple parallel universes even before you start.
Just don't forget to do the calculation, however.