[FoRK] Rules of thumb

Stephen Williams sdw at lig.net
Mon Apr 6 10:23:33 PDT 2009


That's not quite what happened, at least with respect with my "howls": I 
agreed that actual is often likely to be a multiple of planned in 
general.  What I "howled" about is statements that "PersonA's plan is to 
spend X" when the stated plan is to spend X/3.  The more-true statement 
would have been "PersonA states their plan is Y, however I think that A) 
they are lying and are really going to put in place Y*3 or B) to do what 
I think they mean by goal Z, they would have to spend Y*3, or C) 
standard inflation of budget items will make the total Y*3."  All kinds 
of misleading statements were coming out of the right; misleading 
statements shouldn't be tolerated from anyone by anyone who is willing 
to think. 

And this was with respect to health care, not financial industry 
bailout.  The former seems to be on track, even if the spin is "this is 
a start", while the latter was always considered to be an unknown 
quantity by everyone involved.  I think we all groaned when the Bush 
administration (if you remember) committed to the bailout.  And I just 
read that Warren Buffet was the biggest beneficiary, not to mention 
mega-bonus recipients at a functionally bankrupt company.

sdw

Jeff Bone wrote:
>
> Hey, you blokes remember last fall when I tossed out as a good, 
> approximate, non-Fermi rule-of-thumb the idea that figuring out how 
> much something is *actually* going to cost our fine federal government 
> is quite (unreasonably!) often just a matter of doubling their early 
> estimate?  (Or at least that this formed a pretty good lower bound 
> plus or minus a little, particularly when talking about 
> government-provided estimates about particularly very large projects, 
> where their slop factor can generally be historically ascertained...)
>
> Remember the howls of protestation some of you raised?
>
>   http://www.reuters.com/article/topNews/idUSTRE53323420090404?rpc=64
>
>
>
> jb



More information about the FoRK mailing list