Hello Lassi,
[EMAIL PROTECTED] wrote: > Suppose the mobile end subscribes to local services using the CoA in stead of > the Home Address. The local service doesn't have to know about the home address > at all. While having your beer in the pub (for privacy reasons you actually > don't even *want* to use the HA route!) you can check the latest results of > Swamp Soccer World Championships (http://www.swampsoccer.net/) using the > optimal route. Multiplier 1. You are describing the well-known case where Mobile IP is not used at all. It does not necessarily relate to discussion about Mobile IP specifications. > But of course you'll lose the connection, if you happen to do a handover > out of the pub WLAN, so better sit down and order more beer. Or be prepared > to click 'reload' at the new address. I call this URP -- the "User-controlled Retransmission Protocol". The name is quite appropriate. > Now we're getting to percentages of traffic. How many services need to follow > you while moving? How many are used while being static for the duration of the > service session? Most of us don't browse the web while walking or driving! It is possible that a majority of the future Internet will be embedded devices. It is possible that voice applications will continue while people are moving. It is possible that video displays will be of interest to us even while we move. Don't get stuck in the mode of living with what's been forced on us by lack of functional network-wide mobility management in today's laptops and PDAs. > Thus far the Internet services have all been start-ups. IMHO, they still > will be for a long time. They are still looking for the best feel, which > leads back to my previous comment. And the kids are learning fast to > "reengineer human intuition" to what is available. The text messaging > tornado was driven by kids, not business users. None of this supports your contention related to the topic at hand. In fact, service providers across the Internet would be a lot happier if their clients did not have to URP all the time. > As a system architect I respect your arguments, but the wicked service > providers seem to have triumphed over them all... Nowhere in my text will you find the characterization that service providers are wicked. In fact, they tend to deploy the standardized protocols that help them make the most money. It's our job to get those standardized protocols available to them in a way that does not burn all of us in the future. > Making it a MUST wouldn't change anything. There is no Internet Police > that could enforce a MUST. Exactly. We can't make anyone implement IPv6. We can't make anyone implement Mobile IPv6. We can't make anyone implement these protocols, no matter which protocol features we MANDATE or RECOMMEND. But we can foster the expectation that we make good engineering judgements about what needs to be done, if vendors implement the protocols at all. Promulgating systems that are certain to break will not help us to foster this expectation. > We can't even claim that they will lose business, if they don't comply, > because there is no MIPv6 business to lose. So, how is it going to help us make the business if we promote false expectations about the need for various features? > Out of the three known tools of statesmanship, Threats and Blackmail > are powerless. The only one left is Bribing. Therefore I was asking, > if there is a business reason for the fixed, free-for-all, financed-by-ads > services to support MIP. What will they win, if they do it? Even in these dark days of worldwide loss of human lives, property, and civil liberties attributable to the intentional and/or misguided actions of certain of the world's leaders, I am not yet this cynical about the characterization of statesmanship. So I suggest that we continue along with the business of engineering useful protocols. Others in our company, and in other companies, will take up the task of creating systems using these protocols, and they may be able to continue trust in our results if we don't ourselves forget the purpose(s) of the work. I view good IETF work as enlarging the Internet pie, so that we can all get bigger slices. I reckon the service providers will make tons of money from IPv6. Why not? Why not help them, by enabling users to avoid URPing? Regards, Charlie P. -------------------------------------------------------------------- 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] --------------------------------------------------------------------
