Re: [arch-d] ETSI launches new group on Non-IP Networking addressing 5G new services

[John Day <[e-mail]>]

One minor correction.  

TUBA/CLNS was an ISO effort. It is true that it does have a ITU number but nearly all OSI standards had both an ISO number and an ITU number. 

The ITU consistently opposed connectionless and CLNP at every opportunity they had starting with the OSI Reference Model to keep connectionless out of OSI. The first amendment to the OSI Reference Model was to add connectionless and the ITU fought that tooth and nail.  It was only through the efforts of *computer companies,* not ITU members, that connectionless and CLNP was there and the work progressed.

The ITU never wanted CLNP and the IETF obliged them.

There was more CLNP deployed and operational in the Internet in 1992 than IPv6 in 2014.

The ITU owes the IETF an immense debt of gratitude for keeping this issue uncertain for the past 28 years.  This has greatly facilitated the ITU's great success in bringing the IETF even closer to the ITU direction.

Congratulations.

John



> On Apr 17, 2020, at 06:45, Victor Reijs <victor.reijs=[e-mail]> wrote:
> 
> Hello all of you,
> 
>> -----Original Message-----
>> From: Architecture-discuss <[e-mail]> On Behalf Of
>> S Moonesamy
>> Sent: Friday, 17 April 2020 11:08
>> To: Toerless Eckert <[e-mail]>; [e-mail]
>> Subject: Re: [arch-d] ETSI launches new group on Non-IP Networking addressing
>> 5G new services
>> 
>> IPv6 was promoted as "secure by default" as IPsec was part of it.  
> 
> [Victor Reijs] 
> I can remember that when we together had a fully operational TUBA/CLNS service in Europe and US (really 90s). The missing of IPSec was one of the reason not to choose for TUBA/CLNS (beside that it was an ITU standard). TUBA/CLNS was only providing longer (flexible length) addresses. While IPv6 should do indeed secure stuff. Not much has materialise in all these years during which IPv6 needed to be developed/deployed...
> But ok that is in the past.
> 
> I now read in the ETSI ISG-NIN group that the major issues not to use IP inside B5G is de additional cost of spectrum (which is quite limited) due to the large headers in IPv4/v6 and many encapsulations happening. Furthermore congestion control and QoS are really not supported and that might cause problems for real-time or time critical applications. The ICANN report (5G Technology) says that IP can all do that, but why are the ICANN arguments not (yet) convincing the mobile operators?
> I think the specifications are quite clear defined by ETSI NGP's KPI (https://www.etsi.org/deliver/etsi_gs/NGP/001_099/012/01.01.01_60/gs_NGP012v010101p.pdf ) so it must be 'easy' to determine if IP can or cannot do it.
> 
> 
> All the best and stay healthy,
> 
> Victor Reijs
> Research engineer: future internet
> 
> Attendance (CEST): Mon 8.15-16.45 | Tue 8.15-16.45 | Fri 9.00-17.30
> 
> 
> 
> SIDN | Meander 501 | 6825 MD | Postbus 5022 | 6802 EA | ARNHEM | The Netherlands
> T [phone] | M [phone]
> [e-mail] | www.sidn.nl
> _______________________________________________
> Architecture-discuss mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/architecture-discuss