Re: 1teraflops cell processor possible?

Nick Maclaren wrote:
> Jan C. =?iso-8859-1?Q?Vorbr=FCggen?=  <[EMAIL PROTECTED]> wrote:
> > It was proposed, quite earnestly, early in the public history of AIDS in
> > Sweden. While there was the obvious outcry arguing based on human rights,
> > I'm convinced the proposal died for quite another reason: There simply is
> > no way to achieve statistically valid results at current non-African
> > levels of prevalence - the usual hypothesis-testing dilemma of balancing
> > false-positives vs false-negatives. Currently, an "AIDS test" actually is
> > two tests in sequence, with carefully balanced error rates. Of course, you
> > can tolerate substantial false-positive rates for things such as blood
> > and blood products, but they are another issue.
>
> The UK has the means, but it requires a declaration of a State of
> Emergency.  If the current anti-Terrorism law goes through, that
> obstacle will be removed.

Norway also have the legal means, in the form of the 'Smittebćrer' (lit: 
Contagion Carrier) act instituted when tuberculosis was a big problem 
(or should I say, the last time TB couldn't be treated reliably with 
antibiotics?).

However, when the first wisper about using it against HIV/AIDS was 
heard, there was such an outcry that it was politically DOA.

Terje

--
- <[EMAIL PROTECTED]>
"almost all programming can be viewed as an exercise in caching"