Re: Dedicated Z-wave sites?

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

"Dean Roddey" <droddey@xxxxxxxxxxxxxxxx> wrote in message
news:u_Dfh.27540$qO4.11228@xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> I think that the thing limiting Z-Wave is their choice for a very low
speed
> network. It's going to be hard to achieve ubiquity when your system cannot
> ever be used to even transmit media metadata, much less media data. No
> powerline technology really could even be used for either of those things
> either I don't think. Zigbee seems to be the only one that could really
have
> a chance at being a fairly ubiquitious wireless control technology that
> could move up out of the trivial amounts of data involved in turning some
> lights on and off.  It could not transmit media data, but it could
transmit
> metadata and some other small stuff.

I think Zigbee's got the better chance for the same reason VHS beat Beta.
More capacity.  Beta tapes came in some god-awful fractional format and VHS
tape came in 120, 160 and 180's and it was easy to know what that meant in
recording times.  Two hours at SP, six hours at EP.  The slightly longer VHS
blanks could always record one more show the Beta counterparts.  When the
Video Rental Wars began, customer preference for format didn't matter as
much as not having to stock duplicate inventory and the Beta faded away.  It
will be interesting to see who wins the HiDef DVD wars.

> But, in the end, media is going to be the single biggest driver of
> acceptance of home automation technologies. Actually, it may already be.

It's well on the way if it's not there yet.  I keep pulling RG-59, power
cables, CAT5 and doing all sorts of things that will eventually be obsolete
just to pipe TV and audio images from many sources, including CATV and CCTV.
The future will be lots of flea-powered devices running off ethernet cables,
hooking together by stacks of hubs and switches.  A smart wall switch can
have all the sensors I mentioned elsewhere, and probably a lot more, if it's
designed at the chip level and components share power supplies.  Everything
a sensor and switch does can be translated into digital data.  Every sound
and video signal you want to send can be digitized as well.  The interesting
issue to me is how much DRM and things like HDMI connectors will screw up
people's ability to watch and listen to what they want anywhere in the
house.  I think it's going to be a lot worse than many people realize.

> Therefore the system will have to use a backbone that supports media data,
> which pretty much means the ethernet network in any practical sense at
this
> point. There are some other technical possibilities, but no one seems to
be
> really in a position to push them (such as Firewire.)

Ethernet's already there because network cards cost $5 and 8 port hubs $10.
Once a network topology becomes established at a price point like that,
almost nothing else can compete.  It's partly because a lot of cable jockeys
know how to string it.  It's partly because there's a testable standard that
has to be met and pros have the tools to do it.  You may have two or more
networks in the house.  Lots of techies do.  Red cable for HA, blue cable
for PC's.  A bridge of some sorts if, for some reason, you want them talking
to each other.  Just because they use the same topology doesn't mean PCs and
HA gear HAS to live on the same wires.

> If you've already got an ethernet based automation/media backbone, it
seems
> to me that anyone who can build on that backbone, providing just the small
> amount of extra stuff required, would have the best chance. For instance,
> something like Zigbee but very localized, so that you can hang several
> transmitter/receiver devices off of the ethernet network around the home
to
> get very good coverage and good speed. So you can provide retrofit
friendly
> support for lighting and a few other things that would benefit from being
> wirelessly controlled, but not have the data speed downside and coverage
> problems and limits on modules that come with existing wireless/powerline
> technologies.

HomePlug is quite fast enough.  I'm not sure how it would handle HA thrown
into the mix but it's clear to me the powerline's not dead yet.  Homeplug
would always worry me because I'd be subject to anything the power company
might inject into the lines upstream.

> Wireless IP would be optimum in that scenario, but it doesn't seem like
> anyone has managed to get close to providing the small, cheap wireless IP
> package that would be required.

As my wife has often remarked - "why does this wireless gear have more wires
coming into than the wired stuff?"  A single CAT6 cable probably has less
negative SAF than a device with a wall wart that's always sucking power,
that's probably blocking two outlets on a power strip and has to have some
piece out in the open with an antenna showing.

> The downside of course is that IP networks for the home have to get a lot
> smarter and self managing. As the vendor of a networked automation
product,
> we have a fair amount of problems that are not related to our product but
to
> the network itself. It's too easy to get two machines on the same address,
> or to mess up network settings or DHCP settings, or firewall settings and
so
> forth.

Which is why Red Cable for HA, Blue for PC.  Handling communications through
a bridge would seem to be one way to limit the potential insanity of
plugging a smart TV or other ethernet-smart device into your home data
network.

--
Bobby G.