Re: [homegate] A new proposed charter (and name) for HOMEGATE

[Ole Troan <[e-mail]>]

Barbara,

>> 802.15.4 is not bridgeable to other common home network technologies,
>> so a router will be required to connect an 802.15.4 subnet into a
>> home/premises/office network.
>> 
>> - Ralph
> 
> Please forgive my ignorance -- but I thought 802.15.4 was a PHY + MAC
> definition, and that it had absolutely no L3 in that standard. ZigBee is
> a specific technology that uses 802.15.4 PHY + MAC and defines L3 to run
> over it. BTW, ZigBee is not the only group to define L3 over 802.15.4.
> They're just the most successful/well-known, to date.
> 
> As such, I'm completely confused why 802.15.4 is relevant -- of course
> it isn't bridgeable (at an IP layer -- I assume you aren't referring to
> IETF tackling bridgeability at L2), it's just PHY and MAC! But, if
> you're saying that ZigBee isn't bridgeable, then your statement makes
> sense. But if that's what you mean, then you should say so. And if there
> are other technologies over 802.15.4 that you want to create solutions
> for, then those should be called out, too, because they may have
> different needs than what ZigBee has.

there are / will be home networks with data link layers which cannot be bridged together. so we need to connect these links at layer 3.
that text is there to give one reason to consider a "routed home", and from that perspective 802.15.4 is just one example.
others could have been Frame Relay, ATM, Cisco HDLC, FDDI, PPP, LocalTalk, Token ring..., but you can judge the relevance of those yourself. ;-)

cheers,
Ole


> 
> IMO, a router that connects to a ZigBee subnet is more of a "gateway"
> device. That is, if it takes something special (at L3 and above) to go
> between a "standard" home network, and a ZigBee network, then it's a
> gateway. [BTW, ZigBee Alliance has been working on such gateway
> definition, according to their website.] And ZigBee needs to be
> primarily responsible for figuring out how this works. IETF homenet
> would be well within its charter to advise ZigBee to help figure this
> out. But that falls clearly within the "work with other orgs/fora" part
> of the charter, and I don't see why it needs to be noted as a specific
> PHY.
> 
> Will this group also be defining a gateway function for Z-Wave? Or is
> there a decision that this group will support ZigBee and not Z-Wave? Or
> do we really want to go there? [I know I don't.]
> Barbara
> _______________________________________________
> homegate mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/homegate