Fix some documentation examples involving the repr of a float.
diff --git a/Doc/faq/design.rst b/Doc/faq/design.rst
index 94afaff..02e3fad 100644
--- a/Doc/faq/design.rst
+++ b/Doc/faq/design.rst
@@ -75,9 +75,9 @@
function prints fewer digits and this often results in the more sensible number
that was probably intended::
- >>> 0.2
- 0.20000000000000001
- >>> print 0.2
+ >>> 1.1 - 0.9
+ 0.20000000000000007
+ >>> print 1.1 - 0.9
0.2
One of the consequences of this is that it is error-prone to compare the result