[Offtopic] Line fitting [was Re: Numpy outlier removal]

Steven D'Aprano steve+comp.lang.python at pearwood.info
Wed Jan 9 08:50:45 CET 2013

On Wed, 09 Jan 2013 07:14:51 +1100, Chris Angelico wrote:

> Three types of lies.

Oh, surely more than that.

White lies. 

Regular or garden variety lies.

Malicious lies.

Accidental or innocent lies.

FUD -- "fear, uncertainty, doubt".


Lying by omission.

Exaggeration and understatement.




Deceit by emphasis.

And manufactured doubt.

E.g. the decades long campaign by the tobacco companies to deny that 
tobacco products caused cancer, when their own scientists were telling 
them that they did. Having learnt how valuable falsehoods are, those same 
manufacturers of doubt went on to sell their services to those who wanted 
to deny that CFCs destroyed ozone, and that CO2 causes warming. 

The old saw about "lies, damned lies and statistics" reminds me very much 
of a quote from Homer Simpson:

"Pfff, facts, you can prove anything that's even remotely true with 


More information about the Python-list mailing list