Re: parallel processsing

From: Emlyn (emlyn@one.net.au)
Date: Sun Dec 17 2000 - 02:39:19 MST


Anders wrote:
> "Emlyn" <emlyn@one.net.au> writes:
>
> > Anders wrote:
> > > Another tidbit I noted: the number of basepairs that can be sequenced
> > > or synthetized together into strands per day and person appears to
> > > increase superexponentially right now. I saw some plots comparing it
> > > to Moore's law, and it was much more dramatic. After listening to
> > > Craig Venter, I have the feeling that we should look for a biotech
> > > singularity rather than a nanotech one :-)
> > >
> >
> > Isn't that the same thing? I thought the vision was that nanotech
> > would come out of biotech, particularly the protein folding stuff,
> > as general nanotech would bootstrap from protein based work?
>
> Maybe, I think it is likely. But we should remember that there are
> many ways technology can develop beyond the ones we consider
> 'obvious'. Maybe biotechnology leads to a very different society with
> goals that instead create some other kind of singularity than the one
> we are usually talking about? I don't know what it could be or if it
> is likely, but we should avoid becoming technological determinists.
>

I guess it's easier to cling to what you can see, rather than what you
can't, no matter the potential (probable) error.

I asked if it was the same thing, because I thought that protein folding
work in particular was driven by a desire to be able to determine what
protein, or maybe I should say nano machine, would be created by a
particular gene sequence. This in turn so that we can know what effect
changes to the genome will have.

Isn't this so that we can work out what we want to construct, genetically,
and then go ahead and do it? I guess that does have other applications than
building non-biological nanotech. What are the primary motivators for this
line of research, then?

Another question... Blue gene, etc, are being built to enable us to get to
the point where we can code up a gene sequence, then find out what it will
produce. This may be a dumb question, but don't we want to go the other way?
Don't we really want to decide what shaped proteins we need, and then run a
reverse simulation to come up with a sequence which would produce this?

Emlyn



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 09:50:37 MDT