[enum-wg] repost: Proposal for non-geographic ENUM E.164 UPTS for the general public

hi!

>     Chris> can you think of a practical solution to satisfy the ITU's needs?
> 
> I'm not yet sure what the requirements are: ie why you want RIPE NCC
> to operate some E.164 number range for UPT and why this can't be done
> by someone else who already has existing E.164 numbers. From an ITU
> perspective, I think this has to be clear before any allocation of
> E.164 space will be considered.  In other words, you need to explain to
 > ITU that you have a need for space -- how much?? -- that isn't or
 > can't be addressed elsewhere.

ic. as explained earlier, i think the RIRs (that's not just RIPE) are a 
good choice as independent number coordinators in the internet. also, 
they - and with regard to ENUM, RIPE - have proven to be fit for such tasks.
this is why the RIRs seem to be a very good choice for such an operator 
to me.
as explained earlier, i beleive that (an)other organisation(s) could 
provide for the same services, but i don't see a better alternative. 
please feel free to suggest some.
there is currently noone who holds non-geographic ENUM-enabled E.164 
space available to the general public.
how much numbers will be needed is an interesting question, but with no 
doubt such a global prefix for voip will be used by a very large 
community (how many voip-users do we currently have and what is the 
growth-rate?).

> You may also need to show how this
> number range will interwork with the rest of the telephony world.

hm. not sure about that, in fact cc-prefixes interact really badly with 
the voip-rest of the telephony world...
but i think this would be a process of evolution.
actually i think such a number space would present a big chance for 
evolution and convergence.

> An
> explanation of how other telcos and carriers can route calls to/from
> this number range could be helful.

a telco could build its own ip gateways (over here i don't know of any 
big telco without ip, but i know of a lot of productive pstn switches 
that already have voip gateways build in today), or route to other 
telcos providing gateway services for them.
calls from this number space are not a problem at all, this already 
works canonical for cc-numbers over here.

> Bear in mind too that talk of VoIP
> implies telephony by-pass, which is something that makes telcos and
> some regulators *very* uncomfortable.

i prefer to keep out of regulatory discussions (of course while still 
not being ignorant about it, but i beleive that's simply a different topic).
to illustrate what i mean: i certainly won't switch to 'ipv9' because 
china wants to make it mandatory. i wouldn't like - and thank god i'm 
not forced - to be excluded from the internet because of regulations in 
china. the same is true in this case. if a regulator prohibits providers 
and/or users in a country to use global upts (i wonder why anybody 
should or would?), that shouldn't keep the rest of the world from using it.
i guess we all know that making phonecalls on the internet doesn't 
depend on ENUM or E.164 number delegations. also i don't think it's a 
sound claim or the ITU's aim to suppress good voip solutions to keep 
telcos away from competition.
also, my experience shows that many telcos have also discovered that 
voip provides interesting solutions for themselves.

> Going to ITU and saying "give me
> a pile of E.164 numbers for VoIP so I can eliminate revenue for the
> telcos who pay your membership fees" isn't likely to get a warm
> reception.

well who wants that.
on the other way round, do you think it would be a better idea to create 
a private company that sells numbers from such a space to users, to fund 
ITU? i think this doesn't really have a lot to do with funding of ITU.
and isn't telephone number space a kind of public resource? and isn't 
ITU an UN organ in the end?

> And the regulators will want to know about what will be
> done about access to emergency services, universal service
> obligations, QoS, lawful intercept, blah, blah, blah.

right. all this is in discussion over here (and lots of other places, 
too), but thank god it didn't keep providers from providing enum-enabled 
voip-services and users from using them. i don't see why it should. 
regulations have to be followed, but i don't see why this should keep us 
from creating a technical means to run voip a lot more smoothly and to 
provide for a good means of convergence in this area.
i don't see a direct connection between non-geographic enum-enabled upts 
  and the mentioned issues, they are general voip issues.

>     Chris> but i also think it wasn't
>     Chris> a bad choice for ITU to delegate tier 0 operation to RIPE.
> 
> You're playing with fire here. It was IAB who delegated e164.arpa to
> RIPE NCC. [Not RIPE.] I believe ITU were not consulted about this and
> they're still a bit annoyed about that. ITU would like to be listed as
> the administrative contact for e164.arpa, which is another sore point.

...and in the end different points. i've read some letters from the ITU 
TSB to RIPE (NCC) and vice versa regarding an agreement on e164.arpa 
delegation, so i doubt that ITU wasn't involved at all, though there 
might have been less than perfect proceedings. but then again: i don't 
want to shift control or anything like that, i want to have E.164 space 
that is useable for voip, i.e. a global non-geographic enum-enabled 
prefix available to the general public. the +878 prefix is part of 
E.164, it is defined as non-geographic space and UPTS. E.164 is from the 
ITU, there seems to be some sort of consent about the need for such 
space in the ITU itself, and also a block has already been delegated, so 
it's hard for me to imagine that this idea is even something new to the 
ITU. actually i'm a bit surprised that there hasn't been such a proposal 
until now.
(if you ask me, i don't see why ITU shouldn't be listed as admin-c of 
e164.arpa - but i think this is off-topic in this discussion.)

btw: is there anybody from the ITU on this mailing list? that would be 
very helpful, i guess.

>     >> There would also be a conflict of interest given that RIPE NCC
>     >> is already operating the registry for e164.arpa. So if the NCC
>     >> was to apply for a non-geographic E.164 code and get it, this
>     >> would probably need to be administered by a separate
>     >> organisation from that running the e164.arpa registry.
> 
>     Chris> i guess it's a sub-optimal idea to allocate the block to
>     Chris> RIPE and RIPE only, the community of RIRs probably in the
>     Chris> form of the NRO would be a better idea, as the space is
>     Chris> meant to be used globally.  
> 
> I think it would be very helpful if you could sketch out a model for
> how this proposed organisation would be structured and funded, what
> the roles & responsibilities would be, etc, etc.

the proposal was meant to explain that the RIRs could do the job with 
the means they already have developed (whois, dns, a bunch of good and 
working procedures with just very little adaptions), and how workload 
for RIRs could be reduced to a minimum to keep funding needs minimal.

> Bear in mind there could be legal issues too. Like the RIRs extending 
 > their monopolies or going beyond their charter obligations,
 > impact on EU competition regulations, etc, etc.

the RIRs - let's simply say the NRO as the community of RIRs - are 
organizations to provide for coordination of number resources on the 
internet.
looks to me like a perfect match.
the RIRs are not profit-oriented organizations and are 
non-discriminatory. talking about a monopoly doesn't seem especially 
realistic in this case. but i don't want to discuss wordings, what other 
kind of organization could do the job in a better way?

>     Chris> actually the work RIPE did up
>     Chris> to now as tier 0 operator demonstrates the high
>     Chris> qualifications of RIPE as an enum-operator at the tier 0
>     Chris> and tier 1 levels quite convincingly.
> 
> Sure, the NCC does an excellent job at operating the Tier-0
> registry. This is to be expected. However that pretty much rules them
> out of operating any Tier-1 (or Tier-N) infrastructure.

i beleive it were a bigger benefit for voip-users and voip-providers if 
the RIRs operate a +878* tier1 infrastructure than the tier0 infrastructure.
can you point me to the ruling that prohibits the tier0 operator from 
operating as tier1, too? this is certainly a point that has to be 
considered and i wasn't able to find info about that yet.

kind regards,

	Chris Heinze