Dear friends,
I would like to start a discussion on ZigBee protocol for LPDR.

What do you think about it? Do you think it will be the standard de-facto
in LPDR? Or it will be erased by another emerging technology?

Do you think it could be a good idea to specialize in ZigBee development?

Thanks to all

I wait for your thoughs!

Saro

How can you expect to get useful answers to these questions?

Anybody who is part of the ZB empire will obviously tell you that the
protocol is the uncontested wave of the future, and encourage you to
"specialize" in ZB to help build a critical mass of developers.

Anybody who plans to specialize in ZB development will probably
encourage you to look elsewhere, so that the consultant ZB developer
pool remains small and lucrative.

People who know nothing about ZB mostly have nothing particularly
useful to say to you.

FWIW, my opinion of ZB is roughly the same as my opinion of UPnP
<http://www.larwe.com/technical/commonsense.html>, with the caveat
that ZB involves adding stack license fees to products that are
currently HIGHLY price-sensitive. My current employer, given the
products we make, is theoretically an ideal candidate for ZB, but we
will not consider adding it until it is either free or customers
actively demand it. Chicken and egg, since nobody will actively demand
it until it has become an important standard, and it will not become
an important standard until it is designed in to a significant number
of devices.

"Lewin A.R.W. Edwards" <larwe@larwe.com> wrote in message
news:608b6569.0411301913.1f1c1c6b@posting.google.c om...
Lewin,

Great article! It roughly sums up my arguments against too complex networks
on simple devices. It reminds me of someone who is working on a protocol
based on XML over ethernet, to replace NMEA-0183 (marine instruments, ASCII
over RS-422). As many other solutions: completely overkill.
In the future, I'll direct anyone who talks about 'neat' protocol to this
page.

Regards,
Meindert

Meindert Sprang wrote:

I, too, liked Lewins article and seems to sum up my feelings quite well.
Adequate Simplicity is really wonderful and, gadget freak I am, I will not
succumb to sensless networking of all and sundry.

I can see why I would want to log my fridge/freezer temperatures, room
temperatures, outside temperatures and my home's energy consumption. I can
perceieve reasons to monitor what is watched and by whom on TV and computer
(by whatever route - as that keeps a protective eye on my son). However, I
want that information for just my own purposes, not to have it available
for anyone else to poke their noses into.

Seems to me that embedded systems have to build in a secure kernel that can
keep the systems they look after safe from un-authorised prying eyes and
the malicious activities (inteded or accidental) of the interfering
hackers. One way to keep embedded systems secure is not to connect them to
any widely accessible network.

--
************************************************** ******************
Paul E. Bennett ....................<email://peb@a...>
Forth based HIDECS Consultancy .....<http://www.amleth.demon.co.uk/>
Mob: +44 (0)7811-639972 .........NOW AVAILABLE:- HIDECS COURSE......
Tel: +44 (0)1235-811095 .... see http://www.feabhas.com for details.
Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
************************************************** ******************

On Wed, 01 Dec 2004 22:35:19 +0000, Paul E. Bennett wrote:

The discussion in the article about security seems a bit dated. Most
current home networking routers these days seem to have fairly effective
security with a combination of NAT and firewall features. The weakest
links these days is, IMHO, the PC (Microsoft browsers, in particular) and
the wireless links.

--

Phil

On 30 Nov, in article
<608b6569.0411301913.1f1c1c6b@posting.google.com >
larwe@larwe.com "Lewin A.R.W. Edwards" wrote:

Which confirmed my opinions of a lot of 'announcements' that are usually
hey look a student did this great project/research lets do some PR to
get more funds in.

My main response to a lot of such things is

"WHY?"

Classic examples is/was the fad to be web/internet connect appliances
(washing machines, toasters, even pens), which other than "because we can"
serve no useful purpose.

Unless your washing machine with its web interface is going to email you
at work to say

"I cannot do the next three loads because you need to buy
washing powder/liquid on the way home."

or
"I have a fault and I have called out your preferred repair
company, who will come on day month year, do you want to
book that day off"

Of course this only of any use if the washing machine has its own attendant
droids that

collect the washing (from wherever it is in the house)
sort the washing loads by type
load and start each wash load
remove finished loads and do extra drying as necessary
iron clothes that need ironing
return clothes to approriate storage places.

Imagine the chances of that screwing up, especially if the cat/dog decided
to sleep in one of the piles of washing waiting collection/loading/ironing!

Consider this also why does a networked printer need a default gateway set?

--
Paul Carpenter | paul@pcserviceselectronics.co.uk
<http://www.pcserviceselectronics.co.uk/> PC Services
<http://www.gnuh8.org.uk/> GNU H8 & mailing list info
<http://www.badweb.org.uk/> For those web sites you hate

Well, it WAS written in 2001 Actually I wrote it shortly after I
got back from that dizzying bullshit UPnP session at IBM's campus in
Texas. Shortly after that, our working group broke down.

This is no kind of security at all. By definition, web-accessible
services have to be able to penetrate firewalls. Any service that
tries to call home (e.g. a DSS receiver programmed to deliver watching
statistics to home base periodically) also gets straight through those
simple features.

On Wed, 01 Dec 2004 18:53:12 -0800, Lewin A.R.W. Edwards wrote:

In the context of the article, it is orders of magnitude better. Your
article seemed to be assuming that there was no barrier at all between the
home network and the internet, using the horror story of windows computer
sitting naked on a broadband network with (apparently) file sharing turned
on.

You are correct, of course, in that simple firewalls are not adequate for
securely interfacing a sophisticated home network (with lots of embedded
devices) to the world at large.

--

Phil

In article <20041201.2334.304676snz@pcserv.demon.co.uk>,
paul$@pcserv.demon.co.uk says...
with a built in TV, Complete with people staring vacantly at the TV from
stools in the kitchen.


Robert

Lewin A.R.W. Edwards wrote:

I'd add to that by saying the ZigBee community have been remarkably
successful at de-emphasising the underlying 802.15.4 wireless
protocol, since they can't make any money from it. You can create a
low-cost 802.15.4-compliant data transport that doesn't use their
code.

Jeremy Bentham
Iosoft Ltd.

Paul Carpenter wrote:

What about finding its syslog server (if on another net) ?

--
Wil

I think maybe I left out some information in that article, or assumed
some things about the reader.

One of the UPnP working groups was the home router WG. One of the
purposes of this WG was to establish mechanisms whereby home routers
could *automatically* announce themselves to external services and
establish port mappings so that external devices could *automatically*
(with NO USER INTERVENTION) connect to things inside the firewall. The
object was to enable users to remote-control their home appliances over
the Internet without manually setting up dozens of port mappings. A
secondary object was to make certain types of online games easier to
code (I don't know the details there, since I wasn't in the WG, but I
believe essentially this was something to make every router in the
world more XBox-friendly).

This is the type of insanity I was talking about in the article when I
said I was worried that users would buy "simple routers with these
kinds of features". And it's the kind of feature that I would INSIST
should ship factory-disabled and require explicit user intervention
before it can be enabled.

On Thursday, in article <41AF0C97.F1C4F751@nogo.wtms.nl>
wil@nogo.wtms.nl "Wil Taphoorn" wrote:

Most networked printers have not got the faintest idea about syslog, this
should be directly or indirectly on the sending machine (IP address print
job from) and in Windows environments they just use simple lpr protocols.

Setting a default gateway to 0.0.0.0 does not cause any problems.
It is a redundant setting because they have used a 'standard' TCP/IP
stack without understanding what the implications are. That is you could
potentially get a printer to send details to external systems, like the
classified document, internal payroll details, new product specs
or contracts.

--
Paul Carpenter | paul@pcserviceselectronics.co.uk
<http://www.pcserviceselectronics.co.uk/> PC Services
<http://www.gnuh8.org.uk/> GNU H8 & mailing list info
<http://www.badweb.org.uk/> For those web sites you hate

Hi Lewin,

That was indeed a nice and concise writing.

In addition, I think that the practical range of Zigbee devices needs to
be thoroughly tested. Wireless gear, even when maxing out the FCC limits
will not always reach where you want it to. For example, in our house
there are several fiber insulated walls. This fiber is aluminum backed
so that wireless usually quits working in the garage or in certain areas
of the deck. Personally I believe low bandwidth PLC is better for motes
and appliances in a household. And that is or would be the main market.

Regards, Joerg

http://www.analogconsultants.com

couldn't even go out to the bathroom?

reminds me of the announcement of a networked microwave oven with
facilities for online banking. the motivation seemed to be that the
microwave contained a computer, and online banking requires a computer.
and the one computer should be used for both.

Paul Carpenter wrote:

"Why?" is a very good question. One simple reason to want to connect
appliances is that many of them contain clocks or clock/calendars. In
the US, for example, public TV broadcasting stations transmit a
Time-Of-Day data signal (see EIA-608-B if you're curious) that many
modern TVs decode and use. It would be convenient if my oven, microwave
oven, coffee maker, etc. all had access to that, but it's not reasonable
to put a whole TV receiver inside a coffee maker (although some
manufacturer has probably done it). If, instead, these devices were
networked and could get the time from the TV or VCR, that would be quite
handy. My car also has a clock, so wireless might be handy there.

And, no, I don't sell or design any of that stuff.

Ed

Ed Beroset <beroset@mindspring.com> wrote:

Yes. It's so good it can serve as a torture device.

Ever observed a discussion between an curious 8-year-old and his dad?
You can accurately measure the frustration tolerance of parents by
counting the number of "why" question in a row they'll answer, before
that old fight-or-flee instinct kicks in.

--
Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.

The complexity level of decoding the TOD signal directly in the
appliance is orders of magnitude less than the complexity of
implementing a bidirectional horror like UPnP. You also avoid the need
to get both the TV set and the coffee maker type approved as
intentional radiators.

Look at the (low) cost of those radio-clocks that get the atomic time
signal off the air, by the way.

On Friday, in article
<7DXrd.2001$714.499@newsread2.news.atl.earthlink.n et>
beroset@mindspring.com "Ed Beroset" wrote:

There LOTS of ways to do that simple task, if the companies could be
bothered.

In most countries of the world there are Atomic clock radio signals as well
as:-

1/ TV stations using teletext encoded time signal that
displays on screen.
Already used by a lot of VCR manufacturers to set clock
and station ident for auto tuning.

2/ Digital Radio and TV signals in UK contain time signal as well.

3/ Digital RDS on analog RDS system has time within it.

4/ GPS is a pretty accurate time signal.

See above, if the manuafacturers were bothered they would use the existing
and mainly very cheap methods to do the same. Often the cost/benefit to
the *manufacturer* of increasing sales is non-existant.

Putting bloated communications in to wire up the house/office/etc.. is
way too costly to acheive any benefit.

Distributed Time Of Day is at mininum a single pair transmitting 'n' pulses
a second, to a short encoded data burst, with local oscillator to cover
for problems.

For MANY years schools and public buildings have often used synchronised
clocks on a common drive circuit to synchronise all the clocks in a building!
I even remember them from my school days, mumble, mumble, mumble years ago.

A lot of the railways in the UK used synchronised clocks on a station, with
local backup, referenced to MSF (Rugby Atomic Clock). Not that MSF is known
for 100% uptime! That transmitter has a wonderful history, and some of the
stories I have heard over the years about Rugby transmitter during WWII
were interesting.

Thankfully.

--
Paul Carpenter | paul@pcserviceselectronics.co.uk
<http://www.pcserviceselectronics.co.uk/> PC Services
<http://www.gnuh8.org.uk/> GNU H8 & mailing list info
<http://www.badweb.org.uk/> For those web sites you hate

Ed Beroset <beroset@mindspring.com> writes:
And this is a good example of overkill. Unless you live in an
area with very unreliable electrical service, you rarely have
to set clocks. Some of us don't even bother to reset all of
them when the U.S. switches to daylight wasting time.