add an introduction explaining what sparklines are, why you want them, a...

README.md

@@ -1,32 +1,44 @@
 # spark
-### sparklines for your shell
+## Sparklines
+
+[Sparklines](http://en.wikipedia.org/wiki/Sparkline) are small in-line infographics
+which sum up a data-series ([examples](http://commons.wikimedia.org/wiki/Category:Sparkline)).
+They don't take up as much space as a full-blown graphic and so you can use them all over.
+But classic [Edward Tufte](http://en.wikipedia.org/wiki/Edward_Tufte)-style
+sparklines are hard to make - you either need to be an artist or know a library or
+complicated tool.
+
+Why not have simpler [Twitter](http://kottke.org/11/05/twitter-sparklines)-style sparklines,
+drawn with some handy-dandy Unicode characters?
+
+## Sparklines for your shell
 
 See? Here's a graph of your productivity gains after using spark: ▁▂▃▅▇
 
-## install
+### install
 
 spark is a [shell script][bin], so drop it somewhere and make sure it's added
 to your `$PATH`. It's helpful if you have a super-neat collection of dotfiles,
 [like mine][dotfiles].
 
-## usage
+### usage
 
 Just run `spark` and pass it a list of numbers (comma-delimited, spaces,
 whatever you'd like). It's designed to be used in conjunction with other
 scripts that can output in that format.
 
-    spark 0 30 55 80 33 150
+    > spark 0 30 55 80 33 150
     ▁▂▃▅▂▇
 
 Invoke help with `spark -h`.
 
-## cooler usage
+### cooler usage
 
 There's a lot of stuff you can do.
 
 Number of commits to the github/github Git repository, by author:
 
-```bash
+```sh
 › git shortlog -s |
       cut -f1 |
       tr "\n" ',' |
@@ -37,7 +49,7 @@ Number of commits to the github/github Git repository, by author:
 
 Magnitude of earthquakes over 1.0 in the last 24 hours:
 
-```bash
+```sh
 › curl http://earthquake.usgs.gov/earthquakes/catalogs/eqs1day-M1.txt --silent | 
   sed '1d' |
   cut -d, -f9 |
@@ -49,7 +61,7 @@ Magnitude of earthquakes over 1.0 in the last 24 hours:
 
 Code visualization. The number of characters of `spark` itself, by line, ignoring empty lines:
 
-```bash
+```sh
 › awk '{ print length($0) }' spark |
   grep -Ev 0 |
   tr "\n" ',' |
@@ -64,12 +76,12 @@ ruby-1.8.7-p334 in spark/ on master with history: ▂▅▇▂
 ›
 ```
 
-## wicked cool usage
+### wicked cool usage
 
 Sounds like a wiki is a great place to collect all of your 
 [wicked cool usage][wiki] for spark.
 
-## todo
+### todo
 
 so hint hint hint if you're looking for something to hack on.