Re: [Int-area] [v6ops] Still need to know what has changed.... Re: IPv10 draft (was Re: FW: v6ops - New Meeting Session Request for IETF 109 - IPv10)

[Khaled Omar <[e-mail]>]

Ok, I have no energy to keep repeating, I'm sorry, read the full draft please again.

Khaled Omar

-----Original Message-----
From: Bless, Roland (TM) <[e-mail]> 
Sent: Friday, September 25, 2020 5:11 PM
To: Khaled Omar <[e-mail]>; Mikael Abrahamsson <[e-mail]>
Cc: IPv6 Operations <[e-mail]>; int-area <[e-mail]>
Subject: Re: [Int-area] [v6ops] Still need to know what has changed.... Re: IPv10 draft (was Re: FW: v6ops - New Meeting Session Request for IETF 109 - IPv10)

Hi Khaled,

Am 25.09.20 um 16:41 schrieb Khaled Omar:
> Roland, the sending host will encapsulate an extension header with two different IP versions, where is the contradiction here? I don't see it.

This is a bit sad, but probably one last try:
An IPv4-only host does only understand a single IP version, namely IPv4.
Inside an IPv4-only host there is NO understanding of a different packet format (IPv10) or IP address format (neither IPv6 nor
IPv10) by definition. Consequently, it cannot encapsulate an IPv4 packet into an IPv10 packet by lack of knowing the format and functionality.
Otherwise you have an IPv4/IPv10 _dual_ stack host, but that is then not an IPv4-only host by definition.

Regards
 Roland

> Khaled Omar
> 
> -----Original Message-----
> From: Bless, Roland (TM) <[e-mail]>
> Sent: Friday, September 25, 2020 4:38 PM
> To: Khaled Omar <[e-mail]>; Mikael Abrahamsson 
> <[e-mail]>
> Cc: IPv6 Operations <[e-mail]>; int-area <[e-mail]>
> Subject: Re: [Int-area] [v6ops] Still need to know what has 
> changed.... Re: IPv10 draft (was Re: FW: v6ops - New Meeting Session 
> Request for IETF 109 - IPv10)
> 
> Hi Khaled,
> 
> Am 25.09.20 um 15:04 schrieb Khaled Omar:
>>>> You don't even have running code to be able to verify that your proposal actually works (it doesn't).
>>
>> Do you have a running code to state this?
> 
> How should one create running code out of a flawed specification?
> The following picture from your draft already shows that it definitely cannot work, thus no code needed:
> an IPv4-_only_ host by definition does NOT support IPv10 and thus CANNOT send any IPv10 tunnel packets. Same for an IPv6-_only_ host.
> 
> IPv10 Host                                         IPv10 Host
>     PC-1                                                PC-2
>    +----+                                              +----+
>    |    |                                              |    |
>    |    |                                              |    |
>    +----+                                              +----+
>   /    /   <--------------------------------------->  /    /
>  +----+              IPv10 Header (Tunnel)           +----+
>                               (3)
> IPv4-Only Host                                    IPv6-Only Host
> 
> Do you see the contradiction here?
> 
> Roland
> 
>> -----Original Message-----
>> From: Mikael Abrahamsson <[e-mail]>
>> Sent: Friday, September 25, 2020 2:43 PM
>> To: Khaled Omar <[e-mail]>
>> Cc: IPv6 Operations <[e-mail]>; int-area <[e-mail]>
>> Subject: Re: [v6ops] [Int-area] Still need to know what has 
>> changed.... Re: IPv10 draft (was Re: FW: v6ops - New Meeting Session 
>> Request for IETF 109 - IPv10)
>>
>> On Fri, 25 Sep 2020, Khaled Omar wrote:
>>
>>> That’s why looking into the transitions solutions became a mandatory 
>>> or a peaceful solution such as IPv10 that will allow both version to 
>>> coexist and communicate until the full migration.
>>
>> No, any change now just resets the clock and postpones the transition by another 20 years.
>>
>> Meaningful support for IPv6 has been available in end-devices since 
>> the
>> 2006-2008 timeframe when Windows Vista was released and around the same timeframe other end-user operating systems gained support as well.
>>
>> We're now in 2020 in a situation where basically every end user 
>> device in use has IPv6 support, even laggards like Smart TVs have 
>> started to gain
>> IPv6 support. Printers have had IPv6 support for 10+ years.
>>
>> For your proposal, you have zero running code and thus zero devices supporting your proposal.
>>
>> You keep making these statements that upgrades are easy. They are not. 
>> Ecosystems take a long time to build. You don't even have running code to be able to verify that your proposal actually works (it doesn't).
>>
>