Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 02 Feb 2003 17:09:45 -0800
From:      Terry Lambert <tlambert2@mindspring.com>
To:        Edward Brocklesby <ejb@lythe.org.uk>
Cc:        current@FreeBSD.ORG
Subject:   Re: rand() is broken
Message-ID:  <3E3DC159.CC749944@mindspring.com>
References:  <200302022039.PAA13951@warspite.cnchost.com> <200302022117.49248.ejb@lythe.org.uk> <3E3DB5C4.56A04F70@mindspring.com> <200302030042.06444.ejb@lythe.org.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
Edward Brocklesby wrote:
> On Monday 03 February 2003 12:20 am, Terry Lambert wrote:
> > Edward Brocklesby wrote:
> > > Where was it indicated that random() wouldn't change?
> >
> > Right there in the boot message, and again when you logged in,
> > where the system indicated to you that it was a BSD system;
> 
> Sorry, I can't quite work out what you mean by this.  Are you saying that it's
> assumed random()'s API won't changed because it's guaranteed by BSD?

Some behaviours are assumed to remain constant across BSD systems,
which give the system it's "BSD flavor".  For me, the random number
generators have always been one of these.

Personally, I'm concened about being able to repeat physics simulations
that use drand48(), the code for the pair production coming from out
of Berkeley, as well.  So long as the algorithm doesn't change, the
results obtained in 1982 are going to be repeatable at any point in
the future.

This is needed, if you intend to be able to build on published work
that used a transformation of the raw numbers out of the simulation,
and you need the same raw numbers, in order to do it.  The experiment,
in other words, needs to be repeatable.

Basically, for simulations results like this, where the results are
pair productions which are then kept or thrown away, based on
whether the particular pair production is "possible", given the
theorized physics, changing the random number generator is morally
equivalent to changing the data from accelerator experiments, after
the fact.

Lucent just fired a PhD, very loudly and publically, for that type
of number-fudging.

-- Terry

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3E3DC159.CC749944>