λ Tony's Blog λ

APLAS ‘06

Posted on November 9, 2006

After surviving day 1 of the Fourth ASIAN Symposium on Programming Languages and Systems, I have to say it is refreshing to be reminded of how much academic work is being put into providing solutions to the diverse set of software problems that exist today. I use the term “surviving” to annotate my lack of coherence as a result of attending the U2 Vertigo concert in Brisbane the night before and having received a total of 90 minutes sleep (0230-0400) before heading to the airport bound for Sydney on a 0530 (GMT +1000) flight. Since my spatial ability is generally proportional to the quality of my sleep, I have had to memorise much of the latter part of day 1 of APLAS ’06 - ready for processing at a later date - batch processing if you will :)

There were some excellent presentations - all of extremely high quality - and more than expected in my specific areas of interest. There seemed to also be a pre-requisite understanding of the Haskell programming language and much of the theory behind it, which was also great to see, since it allowed some presentations to go into great detail about various topics. The invited presentation scheduled for 0930 (GMT +11) was given by Professor Peter Stuckey titled, Type Processing by Constraint Reasoning. It is in this talk that I learned of a project titled, Chameleon, which goes about providing a rich set of information for type errors. Professor Stuckey also seemed to highlight the common misunderstanding that comes with a type error as being interpreted as “something is wrong” with the correction “at least one of two things went wrong, since you have at least one contradiction that cannot be resolved” - at least, this is my translation of his emphasis. I think this distinction is extremely important and I applaud anyone who sets about making it. I won’t continue paraphrasing what appears to be a massive amount of work in understanding and then implementing an artifact of Type Theory, so if you have a chance, I encourage anyone to take a look at the paper.

It is also during this presentation that I learned of Helium for teaching Haskell to beginners. I have yet to take a detailed peek at Helium, but since I have an interest in teaching at the tertiary level as well as “weaning people of the mainstream guff”, I have hopes that Helium can be an extremely useful tool in achieving my objectives.