#+title: Why Wisp?
#+date: 2014
#+OPTIONS: H:1 num:nil toc:nil \n:nil @:t ::t |:t ^:t f:t LaTeX:t
#+BIND: org-export-html-style-include-default nil
#+BIND: org-export-html-style-include-scripts t
#+BIND: org-export-html-auto-preamble nil
#+BIND: org-export-html-auto-postamble nil
#+BIND: org-export-html-style "<!-- configuration parameters --> <meta name='defaultView' content='slideshow' /> <meta name='controlVis' content='hidden' /> <!-- style sheet links --> <link rel='stylesheet' href='ui/default/slides.css' type='text/css' media='projection' id='slideProj' /> <link rel='stylesheet' href='ui/default/outline.css' type='text/css' media='screen' id='outlineStyle' /> <link rel='stylesheet' href='ui/default/print.css' type='text/css' media='print' id='slidePrint' /> <link rel='stylesheet' href='ui/default/opera.css' type='text/css' media='projection' id='operaFix' />"
#+BIND: org-export-html-style-extra "<!-- S5 JS --> <script src='ui/jquery.js' type='text/javascript'></script> <script src='ui/org-slides.js' type='text/javascript'></script> <script src='ui/default/slides.js' type='text/javascript'></script> <script src='ui/default/style.js' type='text/javascript'></script>"
#+BIND: org-export-html-preamble org-s5-html-preamble-function
#+BIND: org-export-html-postamble org-s5-html-postamble-function
# Export as s5 presentation. See http://orgmode.org/worg/org-tutorials/non-beamer-presentations.html
#+BEGIN_SRC elisp :exports none
(require 'ox-s5)
#+END_SRC
* Elegance
#+html: <br />
** =.,":'_#?!;=
/The most common non-letter, non-math characters in prose/
** =()=
/The most common paired characters/
(From letter distributions in newspapers)
* On Words
#+BEGIN_SRC elisp :exports results :results results raw
(defun shuffle-wordcontent (string)
(interactive)
(let ((strings (split-string string)))
(substring
(apply 'concat
(loop for s in strings
collect (if (< (length s) 4)
(concat s " ")
(concat (substring s 0 1)
(apply 'concat
(mapcar 'string
(shuffle-vector
(string-to-vector
(substring s 1 -1)))))
(substring s -1)
" "))))
0 -1)))
(concat "*" (setq why-wisp-text-scheme-has (shuffle-wordcontent "Scheme follows a minimalist design philosophy")) "*")
#+END_SRC
#+RESULTS:
*Shcmee fowlols a malisimint dsegin plhpoihosy*
- /Defocus a bit/
- /Can you see the meaning?/
* Scheme for Newcomers
#+BEGIN_SRC elisp :exports results :results results raw
(concat "*€" (combine-and-quote-strings (split-string why-wisp-text-scheme-has) " €") (apply 'concat (loop for word in (split-string why-wisp-text-scheme-has) collect "¥") ) "*")
#+END_SRC
#+RESULTS:
*€Shcmee €fowlols €a €malisimint €dsegin €plhpoihosy¥¥¥¥¥¥*
- /Try it again/
- /Do you see how much harder it got?/
- /€ and ¥ escape your existing filters/
* On Lisp
1. Lisp syntax uses the *most common* non-letter, non-math characters.
2. The *first and last characters* are important for text-recognition.
Let’s call these elegance 1 and elegance 2
Any fix for elegance 2 should preserve elegance 1
Elegance 0: generality and homoiconicity: *code is data*
* Summary: Current Lisp Syntax
- √ *Code is data*.
- √ Uses the *most common* characters.
- *×* The *first and last characters* are always *the same*.
* On Wisp
#+html: <br />
#+BEGIN_SRC scheme
define : hello
display "Hello Schemers!\n"
#+END_SRC
becomes
#+BEGIN_SRC scheme
(define (hello)
(display "Hello Schemers!\n"))
#+END_SRC
* Why not SRFI-49 (Indentation-sensitive syntax)?
** Scheme
#+BEGIN_SRC scheme
(+ 5
(* 4 3)
2 1 0)
#+END_SRC
** SRFI-49
#+BEGIN_SRC scheme
+ 5
* 4 3
2
1
0
#+END_SRC
- Cannot continue the argument list
* Continuing the argument list in wisp
** Wisp
#+BEGIN_SRC scheme
+ 5
* 4 3
. 2 1 0
#+END_SRC
- Complete representation of arbitrary structures
- Generalize =(. x) ⇒ x= to =(. x ...) ⇒ x ...=
- Use =. . x= to get a real =. x= after processing
* Why not SRFI-110 (Sweet-expressions (t-expressions))
** SRFI-110
#+BEGIN_SRC scheme
myfunction
x: \\ original-x
y: \\ calculate-y original-y
#+END_SRC
#+BEGIN_SRC scheme
a b $ c d e $ f g
#+END_SRC
#+BEGIN_SRC scheme
let <* x getx() \\ y gety() *>
! {{x * x} + {y * y}}
#+END_SRC
/This breaks elegance 1/
/The problems of SRFI-49 are preserved, but their impact reduced./
* Summary: Why wisp?
** Wisp
/minimal indentation-based syntax/
** Wisp vs. SRFI-49: continue the argument list.
** Wisp vs. SRFI-110:
*** use common characters (elegance 1).
*** KISS.
* Wisp syntax 1/4: function calls
** Indentation
#+BEGIN_SRC scheme
display
+ 3 4 5
newline
#+END_SRC
becomes
#+BEGIN_SRC scheme
(display
(+ 3 4 5))
(newline)
#+END_SRC
* Wisp syntax 2/4: Continue Arguments
** The dot
#+BEGIN_SRC scheme
+ 5
* 4 3
. 2 1
#+END_SRC
becomes
#+BEGIN_SRC scheme
(+ 5
(* 4 3)
2 1)
#+END_SRC
* Wisp syntax 3/4: Double Parens
** The colon
#+BEGIN_SRC scheme
let
: x 1
y 2
z 3
body
#+END_SRC
becomes
#+BEGIN_SRC scheme
(let
((x 1)
(y 2)
(z 3))
(body))
#+END_SRC
* Wisp syntax 4/4: Resilient Indentation
** The underscore (optional)
#+BEGIN_SRC scheme
let
_ : x 1
__ y 2
__ z 3
_ body
#+END_SRC
becomes
#+BEGIN_SRC scheme
(let
((x 1)
(y 2)
(z 3))
(body))
#+END_SRC
* Summary: Wisp syntax justification
#+html: <small>
http://draketo.de/light/english/wisp-lisp-indentation-preprocessor#sec-4
#+html: </small>
/Required for the goal of wisp: indentation-based lisp with a simple preprocessor/
- =.= to continue the argument list
- =:= for double parens
- =_= to survive HTML
* Wisp mission
#+html: <br /><blockquote>
“I love the syntax of Python, but crave the simplicity and power of Lisp.”
#+html: </blockquote>
See the site for info how to test wisp:
#+html: <small>
http://draketo.de/light/english/wisp-lisp-indentation-preprocessor
#+html: </small>
# Local Variables:
# org-confirm-babel-evaluate: nil
# org-export-allow-bind-keywords: t
# End: