Status: draft
2016-11-12

Saturday 11/12/16

Yesterday and last night, I mostly worked on website features, including making the BrowserFS file system “InMemory” for added security (easy!), adding a cookie notification script (easy!) and setting up the blog categories, folders, index pages, formatting, etc. I also did some reading on Pay What You Want (PWYW) pricing systems, esp. the Fair Pay What You Want (FairPay) approach, which looks especially interesting for my purposes.

Today, I’ll start generating blog entries (like this one), and work on the “Team Negotiation” model, especially task planning and execution as an affordance.


I ended up searching for and reading a bunch of articles and book chapters at the intersection of organizations, semiotics (Peircian), and affordances. Thankfully, I found a bunch, including:

cite:binnekamp_open_2006

cite:bouissac_bounded_2003

cite:bouissac_criteria_2003

cite:buckland_information_1991

cite:donnellon_communication_1986-1

cite:gazendam_models_2003

cite:goldberg_constructions:_2003

cite:gudwin_semionics:_2004

cite:helmhout_social_2004

cite:helmhout_social_2006

cite:heusden_reconsidering_2002

cite:liu_information_2001

cite:marengo_division_2005

cite:norman_cognitive_1986

cite:perlovsky_abstract_2011

cite:pike_supporting_2007

cite:souza_semiotic_2005

cite:steels_emergent_2002

cite:steels_grounding_2002

cite:steels_modeling_2011

cite:steels_origins_1998

I also reread some of Holland’s Signals and Boundaries (2012) chapters 10 and 11.

I also spent a lot of time in the evening writing and testing a regular expression induction program in WebPPL. It sort of worked, but there are problems in the grammar because it too easily leads to productions that don’t stop. This is due to uncertainty in probability of recursion in some of the production rules. Basically, probability of any recursion can’t approach 1.0, otherwise infinite loop! I’m not sure how to solve this, however. I am inclined to use a very different approach that does not involve generating full regular expressions and then conditioning on their length and pattern match true/false.