On 11/21/06, Colm MacCarthaigh <colm at stdlib.net> wrote:
>> The problem is that you are abusing statistics.
He didn't offer statistics.
Although it's possible
> to make general statements about the level of accuracy of statistically
> aggregated data it is *never* appropriate to make inferences from that
> to any one particular datapoint.
>> So, although I can say that 95% of ILUG subscribers are male, that
> doesn't mean "particular ILUG subscriber is male" is a "good guess",
> hell it's not even appropriate to say that "there's a 95% probability
> that this particular ILUG subscriber is male". Those statements are
> meaningless. The person either is or isn't male, there's no intermediate
> state.
Both statements have meaning. If you choose to pretend they've no meaning,
that's a completely separate problem.
GeoLocation is wrong in many cases, mostly for users of multinational
companies corporate networks which don't have local proxies. But, that's a
pretty small proportion of users. Certainly, it's financially worthwhile for
advertisers to use GeoCoding. And that's a real quality metric.
In a test of some commercial systems by the US govt., between 80 and 90% of
IPs were accurate to within 300 meters (for 2000 random IPs across the US).
That's pretty accurate.
John
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
