# HG changeset patch # User Steve Losh # Date 1467655436 0 # Node ID 8641da7e906206c7b34d3d865ac961eedf32464f # Parent a9e6ab6bc33302f29fe2773a349d53d4cf57e98a Update diff -r a9e6ab6bc333 -r 8641da7e9062 README.markdown --- a/README.markdown Mon Jul 04 12:04:16 2016 +0000 +++ b/README.markdown Mon Jul 04 18:03:56 2016 +0000 @@ -459,3 +459,19 @@ familiar. It also matches GDL, so I won't have to transform things when parsing a GDL game, which is nice. It'll also let me write a macro to bind results to variables, which is a nicer UI than having to pull them out of a data structure. + +### 2016-07-04 + +* Watched the "Julia: To Lisp or Not To Lisp?" talk from [ELS 2016][]. I wish + he had gone further in depth on how Julia does macros. He showed a simple + `timeit` macro but I would have liked to know how to introspect the + expressions a macro gets and work with them. + + One of the things that makes Lisp macros so nice is that the expressions + macros get (and the tools you use to work on them) are the same kinds of + things you use everywhere *else* in the language (lists, symbols, etc). This + makes it really frictionless to switch back and forth. If your macros are + getting some special `AST_Node` type that you need to learn to work with it + feels a lot more like work. +* Poked around in `#lisp` and `#sbcl` to clear up some performance-related + questions I had about Lisp arrays. I have exciting new ideas for [Bones][].