Statistics aren't everything. [snip]
The 1.5% says to me [snip]
Did your first point just invalidate your second?
I think khasim brings up a good point. If your argument hinges on quoting a statistic and inferring all sorts of things from that, it seems fair to invalid it with another statistic.
will return true if someObject is defined, regardless of whether someObject = true/false, so you have to compare it to a value.
I did something similar with javascript a few years ago. The javascript is used to codify the RFC BNF, which then generates the regex.
http://www.digitalxen.net/files/emailValidation.js
The other one was a regex for validating phone numbers (at least in the US). It was based on standards from NANPA and ATIX. It worked great until a phone company "accidently" started issuing numbers using one of the exchange codes set aside for testing.
My sister opened a computer store in Hawaii. She sells C shells down by the seashore.