Hi, As I know, the very first reference and implementation made public for Squeak was mine, based on private developments for other smalltalk platform and implemented in public media as minimal changes to Squeak VM (it was under a name like non-compiledMethods). The main idea was based in the fact that methods are objects (as required by reflection features and underlying machinery), and if we use an object of any class to implement behavior it is legal and the responsibility to resolve the understanding of the message (as part of the contract assumed when message was received) if for that object... That kind of reasoning was discused many times in Smalltalking list (www.smalltalking.net), talks about Object Ambiences and before, in regional meetings (in Latin America) related with Object Technology, looong time ago (+15years) In a decade++ we have reflected in other topics of interest based on the concept that smalltalk change (contents) without loosing identity; so we are still naming Smalltalk to media that reveal us evolution, preserving it's identity (... name). As the ideas and implementation was made public without any copyright conditions there was people that has taken the idea and implementation as sources for inspiration without mention the origins. Hope this bits of history are useful to draw one path of the set of histories related with this "technique", Ale.
----- Original Message ----- From: "Yoshiki Ohshima" yoshiki@vpri.org To: "The general-purpose Squeak developers list" squeak-dev@lists.squeakfoundation.org Sent: Wednesday, October 12, 2011 7:25 PM Subject: [squeak-dev] History of ObjectsAsMethods?
Hello,
I used ObjectsAsMethods feature for something by myself first time yesterday. It is used as a per-method basis dNU: override but much cleaner.
I chatted with my colleagues about the origin of its idea and implementation. Does this trace back to an implementation Andreas did 10 years ago, or is this something else came from Europe?
-- Yoshiki