Re: Controllerless distributed automation

["David Paterson (Pixi)" <david.paterson@xxxxxxx>]

On Wed, 14 Feb 2001 09:31:02 -0000, you wrote:

>> This method has been discussed before on this list David.  I think
Dr Jo=
hn
>> was developing a system similar to this too.
>
>Yes I am. But not entirely for the reasons of reliability, its a hobby
and=
I
>enjoy doing it. I would have to agree with a comment Paul mentioned, I
hav=
e
>a HomeVision and it has never failed, Its easy to use, and if it did
get
>fried I could buy another, I also have a RF remote system which has
been
>working for 5 years without any problem also Micro based.

My reasons are very similar really, and I don't expect my own designs to
actually outperform systems like HomeVision in the reliability stakes.
I had this inspiration, probably derived partly from research into
dynamic serverless networking I was doing, and it might be an
interesting project putting even a basic system together.  Anyway, it'll
keep me off the streets :-)

>I live out in the sticks without mains services so I have two pumps for
my
>water system just in case one fails, I could have two HomeVision's for
the
>same reason, I am sure with a bit of hardware, and a good bit of code
you
>could have one monitor the other and switch over in the event of a
failure=
,
>but I have to say my HomeVision is much more reliable than my water
system=
!

I've just moved waaay out into the wilds too, but luckily I've got mains
water.  The peace and quiet is brilliant, and dark skies for my other
main interest - astronomy.  Looks like being a good night for star-
gazing tonight too

>There are commercial systems which are autonomous but they tend to be
very
>expensive.
>
>The problem with developing your own system is technology keeps
overtaking
>you, I have several times re designed because new parts meant I could
do
>more.

LOL - I expect this project will go the same way, but it should be a lot
of fun doing it.  I've been in computers since before most of the people
I work with were born, so I'm used to the insane curve of Moore's law.
I expect before I finish I'll be putting 1 GHz processors with 128 Mb of
RAM into light switches.

<OBreminiscence>
Actually, that reminds me of a lecture by Danny Hillis I was at last
year.  He's one of the pioneers of computing, and told the story of when
he was consulting at RAND just after the first microprocessors were
produced.  He told them there would be a huge market for them, probably
way more than the entire population of the planet.

They scoffed and said "That's crazy, you'd have to have a processor in
every door knob to sell that many"

A few years later, he was checking into a hotel, and he realised there
_was_ a processor in every door knob!
</OBreminiscence>

So you never know, distributed intelligence.... smart light switches....

DP

--=20
David_Paterson =3D david.paterson@xxxxxxx =A6 david.paterson@vissci.=
com;

R & D programmer        There are three kinds of people in the world -
Visual Sciences Ltd.      those who can count, and those who can't...