Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [RE] What's wrong using FireWire as the home backbone?



David,

When you ask the question, "Wouldn't you prefer to be on top?" you seem
to be equating people (or possibly companies) with standards. That is
illogical. A person (or company) capable of building an 802.3 network
should also be capable of building a 1394 network. Using standards as a
means of "winning the market" (for one person or company) is a very
lousy way of making standards - and is generally not successful. The
best way to make standards is to define the best way of solving a
problem and then try to encourage the most people to follow that way.

I did not see a specific comment that "what's wrong with Ethernet" for
the home network is that "it's not FireWire." That was my interpretation
based on many of the comments that I read. Perhaps it's too hyperbolic -
that's my style :-)

Given that:

Many more home electronic devices have FireWire than Ethernet (cameras,
video cameras, etc.), that FireWire solves the synchronous networking
problem and that bridging an asynchronous protocol across a synchronous
network is many times easier than bridging a synchronous protocol across
an asynchronous network it would seem that the correct choice for home
networking would be IEEE 1394. There may be scope for developing a new
physical layer interface (or PHY, if you prefer) for FireWire that could
use the CAT-5 cabling.

Given that home networks are in their infancy, I do not see any
(inherent) problem with commercial viability of standardizing home
networks based on either 802.3 or 1394. Given the rapid rise in
popularity of wireless LAN products it may become common to see most
data devices, low bandwidth devices and low QoE devices connected with
wireless, while high bandwidth and high quality devices requiring wired
connections will use FireWire. Some (geeks) might use wired Ethernet to
recreate their office environments in their home (although more and more
office environments are incorporating wireless) - then again, some will
use IR bridges or ATM...

I am neither excepting nor accepting anyone else's idea. As far as I can
see there is a problem, it already has a solution and further discussion
seems moot. Of course, I may be missing something very important - so
once again, I'll ask:

What is missing from FireWire that prevents it from solving the problems
of wired home networking?

Hugh.



David V James wrote:

>Hugh,
>
>I'm afraid that your interpretation, when reading between
>the lines, is not the same as intended. Specific comments
>follow.
>
>
>
>>>So far there has been some discussion on the subject of "what's wrong
>>>with Ethernet" for the home network and, as far as I can see it, the
>>>main response seems to be "it's not FireWire."
>>>
>>>
>
>1) The limitations of Ethernet have been (I believe) reasonably well
>   stated, in terms of no timer synchronization, admission control,
>   and fixed-rate deliveries.
>
>2) Those experience with consumer devices and real time transfers
>   tend to come from the 1394/Firewire domain, since that group
>   has worked on the problem for some years.
>
>   I see no problem with excepting others' ideas, as the NIH
>   alternative is very costly in terms of time and results.
>
>3) The main argument for 1394-in-the-home is that Ethernet does
>   not support synchronous services. I personally prefer to solve
>   that problem, on Ethernet, rather than encourage two solutions
>   in the home, confusing the customer, and limiting the volume.
>
>4) I have seen no one say that 802.3 is defficient because
>   "it's not FireWire". Can you point out that quote?
>
>
>
>
>>>So, could someone with expertise in this field summarize what
>>>the deficiencies are with IEEE 1394 that require it to be
>>>replaced with an enhanced IEEE 802.3?
>>>
>>>
>
>To me, splitting the marketplace and excluding 802.3 from rapid
>expansion of the CE marketplace are the main disadvantages of
>wiring the home with 1394.
>
>
>
>
>>>It strikes me that the problem of bridging:
>>>
>>><-- IEEE 1394 -->...<-- IEEE 802.3 -->...<-- IEEE 1394 -->
>>>
>>>would be best served by:
>>>
>>><-- IEEE 1394 -->...<-- IEEE 1394 -->...<-- IEEE 1394 -->
>>>
>>>
>
>That's an approach, and it has its advocates. Particularly among
>those that believe 802.3 is incapable of solving the synchronous
>transfer problems. That's the primary reason that the 1394 bridge
>standard was developed: for 1394 networks within the home.
>
>But, if 1394 were to be the household connection of choice,
>then one would have to deal with:
>
>    within the room  | within the residence | within the room
>
>                       ..<-- IEEE 1394 -->..
>   <-- IEEE 802.3 -->..                     ..<-- IEEE 802.3 -->
>
>While that is a viable solution, and work has been done on
>sending IP over 1394, as well as 1394 over CAT-5, I'm surprized
>to see an 802.3 member apparently advocate such a position.
>
>I also doubt the commerical viability of replacing home Ethernet
>networks with home 1394 networks, or placing the two networks
>in parallel.
>
>Wouldn't you prefer to be on top? Given the choice, I would(:>).
>
>
>
>