RE: [EFM] EFM Requirements
I partially agree with you both. As always its a grey world. I can clearly see both requirements.
In the case of TV generally they're pretty stationary devices, so moving the set-top wouldn't happen. My understanding of a cable modem is it is coax fed, so this is not intuitive for users to move anyway.
I also agree that home networks will likely become an increasing norm, more though I see the requirement being that self install EFM over copper will likely be provided over the existing pair and therefore into the existing telephone sockets - now this is something I do know about :-) Home wiring varies from country to country , but the most difficult for us as far as I can see is what is implemented in the UK. The line comes into the site to a master point, and is subsequently star wired(bridgetaps) from there to the varying points around the house. Generally the master is in the reception/hall of the building. Here comes the challenge - typically dsl doesn't like bridgetaps, so you have to filter at the master and connect the modem in, but who the hell wants their modem in the hall? so engineers have to visit to ensure the wiring is sorted correctly. So if I understand you correctly the ability to ignore the master, just plug in to the line next to my pc (ive probably been!
using modem) and work would be excellent. I can see this as truly enabling self install for carriers and removing a lot of set-up costs from the end users perspective.
R's
Andy
-----Original Message-----
From: Fletcher E Kittredge [mailto:fkittred@xxxxxxx]
Sent: Wednesday, August 15, 2001 5:43 PM
To: Sherman Ackley
Cc: 'Christophe.DEL-TOSO@st.com'; stds-802-3-efm@ieee.org
Subject: Re: [EFM] EFM Requirements
On Tue, 14 Aug 2001 11:58:09 -0400 Sherman Ackley wrote:
> The service should operate on a point to point basis with the modem in the
> house/business. There, the user will install a firewall to separate their
> home network from the Internet access service. The consumer can be expected
> to move the modem from room to room and they will expect it to work from any
> jack. They do not understand being restricted to just one jack. What this
> means is that the Internet access service and the home networking must be
> divided by spectrum on the same cable pair. This should be possible through
> the use of intelligent filters such as those used today to separate voice
> from DSL.
Sherman;
First I would like to say that I could not agree more with the
vast majority of your statements and your excellent analysis. "Me
too!" gets boring so I have not bothered to actively signal assent.
The only statement I don't agree with is the scenerio listed above.
Our experience does not match the scenerio you outline.
We sell high speed Internet access via both ADSL and DOCIS
modems. We have found that customers are quite willing to do
self-installs of their own splitters and that they do not seem to want
to move "room-to-room" in the manner you describe. Rather, they are
quite happy having in-house networks that are seperate from a phone
line.
Right now, I am sending from a laptop on my kitchen table in
rural Maine. The laptop, like all of our five computers, has a
Lucent wireless ethernet card ($60). We have an wireless node
connected via Ethernet to an integrated Ethernet hub/router/modem
which acts as firewall. Since we are about 8 miles from the CO, we
use ISDN rather than DSL.
The last thing I want to do with my laptop is move around the
house connecting it to *any* wires. As it is now, I can move around
to any room at the drop on a whim. I also can't imagine wanting to
move the TVs around.
Talking to others, I know that the way I use the service is
common. I have never heard of anyone saying they use either IP over
DSL/DOCSIS in the scenerio you describe.
So while I agree with you that the scenerio you describe is
possible, I do not agree that it is the only one, the most common one
or even widely users. I do not believe that the design should be
constrained to meet this scenerio unless satisfying this contraint
causes no burden.
regards,
fletcher