Re: Nigel's gratuitous troll

[Nigel Orr <nigel.orr@xxxxxxx>]

At 17:25 01/09/99 +0100, you wrote:
>How well do DECT telephones work in practice? If everyone in the street
>has one do they run out of available channels, or do they implement
code
>division as you mention?

They don't do code division- that might happen for the next generation of
mobile phones (I did some research on that in 1994).  I can't recall what
they do use- probably pick a quiet frequency and use that, and switch if
there is interference.

>Presumably if you transmit your signal digitally, using a DECT like
>protocol, it will be superior to an analogue signal over a basic cable.

No reason why it should be.  If anything it should be worse.  Instead of
just a cable from the handset to the telephone base, you have analogue
processing, maybe data compression, and a radio link.  All those are
unlikely to leave the signal unchanged.  A 'full quality' phone link only
has about a 4kHz bandwidth, or 64 kbits/s for a digital link- I don't know
if DECT uses 64kbit/s or not, but I would suspect not.

>Assuming that the average person doesn't use balanced audio cables 'cos
>they're not that bothered about quality/the distances are too small,
and
>assuming that the digital transmission has sufficient bandwith, what is
>the problem?

The problem is assuming sufficient bandwidth- the more people that use it,
the less sufficient it will be.  It isn't infinite, just 'enough for now'.
It's just like the quote from the MP3 Anywhere web page that I mentioned,
which said that because there weren't many people using the band, it
suffered less interference.  That's a _very_ short time view.

>Cabled solutions seem fundamentally flawed to me from a practical
level:
>whilst people are prepared to accept a second lead (second to power)
>going into the back of the TV, few would be happy with the idea of
>plugging two leads into the back of a kettle:

Again, my needs are different from some others on the list- if I need to
control my kettle, it will at most need the power switched on and off.  At
present, I use my Tefal teamaker, so the kettle just gets boiled for
washing dishes, when I am in the kitchen anyway.  I'd certainly be happier
with a wired link to my kettle, if only for reliability, than a radio
antenna which accidentally switched it on while empty just because it
happened to get confused by my neighbour's new garage door opener...

>control/feedback. This is why I like USB because of the convenience
>factor.

It has 2 links- power and data- they just happen to be on one cable- no
harm in that.  Your idea of a modified mains plug is probably the best long
term answer, although a localised mains transmission system to send data
only down the appliance cable could be made reliable enough.  Then the
cat-5 would just have to come into the wall socket, which would take the
data and add it to the local, suppressed, section of the power line.

>purely for the convenience of 'installing' an automatable light simply

Wireless does make installing easier.  It also needs power, and may need
setting up, and can be ruined at the whim of the RF planners.  When we
moved in, we found a burglar alarm system left by the previous elderly
resident.  She was delighted with it, and spoke highly of the installing
company.  All of the batteries, in all of the sensors, were flat.  The
sensors were installed at waist height, and were easily confused.  The
radio link was (IIRC) at 55MHz, no longer allocated for this purpose.  And
the cables that did have to be fitted (to the siren and mains supply) were
run on the surface with no attempt to use available spaces.  I hope this is
unusual, but I expect it isn't.  I contrast that with my parents house,
where a wired alarm was fitted recently.  There are _no_ visible cables.
The wiring took experienced installers one day.  The house is not small.

>for domestic home automation, you are never going to sell a system to
>ordinary people that needs a re-wire of their house and everything
>plugged in twice.

The rewire shouldn't be a problem- existing competent security installers
should have a bright future if they catch on to that.  Plugging everything
in twice is more of an issue, where items have to be controlled.  It's not
a problem for me, as I'm struggling to think of anything which would need a
data connection which I ever unplug.  The only plugs which get moved in our
house are the PC (very rarely), power tools (which don't need to be
automated), my inspection lamp, the lawnmower (won't be automating that for
a while), my wife's hairdryer (same again) etc etc.  I might have missed
something, but I don't think having to plug and unplug 2 sockets once in a
blue moon is a problem.  Others might have different scenarios, and I would
be interested to hear them.

>while. And if you don't sell the system to ordinary people, then
volumes
>will always be low, and we'll moan about high prices in the UK.

True

>why a standard for DECT like transmissions of audio and control
>information isn't possible within a domestic environment. If the
>transmissions via radio are digital and have proper collision detection
>etc then I don't see what the problem is.

Phones are less used than HA.  You want the phone to be in use for, say, 1
hour per day.  From that, and the expected takeup, you can generate
statistics for how many channels are needed at a given power to reduce the
likelihood of problems to, say, 0.1%.  However, you want your security
system to be 100% reliable, 24 hours per day.  You want the 'switch off'
signal to be received by your water boiler without fail.  So you add
collision detection, and now every node needs a transmitter and a receiver,
and you need twice the wireless traffic... each to their own.

>I accept that cabled protocols will always be better in principal, but
>then I also accept that Betamax was better in principal...

But sadly marketing won, so we're stuck with the poorer system- I hope we
won't let that happen to HA.

Nigel

------------------------------------------------------------------------
MyPoints-Free Rewards When You're Online.
Start with up to 150 Points for joining!
http://clickhere.egroups.com/click/855


eGroups.com home: http://www.egroups.com/group/ukha_d
http://www.egroups.com - Simplifying
group communications