Hi Daniel, I agree with you, one way to split it would be hosts and servers. I have suggest this, but the WG wanted a very simple definition, with no new terms. Since host / server is not defined in earlier documents, there is a terminology problems with earlier documents, So the WG decided not to use such terms.
John > -----Original Message----- > From: ext Soohong Daniel Park [mailto:[EMAIL PROTECTED]] > Sent: 18 December, 2002 01:51 > To: Loughney John (NRC/Helsinki) > Cc: [EMAIL PROTECTED] > Subject: RE: draft-ietf-ipv6-node-requirements-01.txt > > > I am interested in home network of course using of IPv6. > When I thought a definition of IPv6 node, I separated two kinds of > nodes, > one is "active IPv6 node" which need to communicate another node and > bidirectional connection > the other is "passive IPv6 node" which is just controlled not try to > access another node and unidirectional connection > As you said below, thermometer is only passive node, > therefore it is not > need DNS resolving. > IPv6 node is very very widely, then almost possibility must be > considered not only PC or laptop... > e.g. > active IPv6 node : almost pc, laptop, PDA, hand-held devices, > passive IPv6 node : thermometer, lighting, washer, refrigerator, some > kinds of home appliances > > Daniel > > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, December 17, 2002 10:09 PM > To: [EMAIL PROTECTED] > Subject: RE: draft-ietf-ipv6-node-requirements-01.txt > > > Hi Daniel, > > > 4.5.5 "....Stateful Address Autoconfiguration MAY be > supported...." I > > am confusing "MAY" When above node is fail to obtain > stateless address > > > autoconfiguration, if node didn't have Stateful AA function, it is > > impossible to make each own address. so I think "Should" is better. > > In what cases will stateless address autoconfiguration fail? If it > fails, there is always the possibility of setting addresses by user > interaction (typing in the addresses). Stateful is dependent upon > DHCPv6, and users cannot depend that it will be implemented > everywhere, > especially > since it is not a standard yet. > > > 4.6.1 > > This section is more related mobileip, not general IPv6 node > > requirement. > > MobileIP is actually considered as part of IPv6, but I get your point. > This I could move into the MIP section. > > > > 5.2 "... Not all nodes will need to resolve addresses" > > If yes, how can we obtain wanted address ? , basically we > > don't memorize IPv6 address. > > Some nodes will be servers only and only accept incoming requestions. > Think of a thermometer with an IPv6 stack. It will never need DNS. > > > in addition, as 1.2, Desctiption of an IPv6 node is so > ambiguous. In > > this draft, almost requirement is for general node which is > pc, laptop > > > and that kind of thing. But implemented IPv6 is not only > general node > > but also various devices (home appliance, tiny device and so on) > > then you should write clear definition of IPv6 node. > > Since this requirement will be referred to implement widely, > > all section must be clear. > > It is ambiguous for a reason, as an IPv6 node is anything with an IPv6 > stack - cell phone, laptop, server, IPv6 thermometer (yes, > someone makes > one already) or any embedded device. If you have a better suggestion > for > a definition, I would be happy to review it. > > John > > > > > Daniel > > > > > > ===================================== > > Soohong Daniel Park > > Junior Researcher > > Mobile Platform Group > > Digital Media R&D Center > > Samsung Electronics Co.,LTD > > > > TEL:+82-31-200-3728 > > FAX:+82-31-200-3147 > > H.P:+82-11-9950-4655 > > mailto:[EMAIL PROTECTED] > > ===================================== > > > > > > -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
