[stir] Re: [art] Re: For those of you who follow this kind of stuff.

[Roman Shpount <[e-mail]>]

Pierce,

I never called STIR/SHAKEN a success.

If you look at it as a mechanism to reduce the number of robocalls, it
failed completely. It can easily be used to combat robocalls, but I don't
know if anyone implementing it actually cares about it. It was always more
about going through the motions, regulatory compliance, and possible
deniability.

If you look at it from the point of view of ease of deployment, it was
horribly designed with complete disregard for how current carrier networks
operate. It is a networking, implementation, and administrative nightmare.
I would not blame this group for creating a sub-par standard, but it was
driven more by political necessity than by reasonable design considerations.

If you consider it in terms of providing helpful information in combating
robocalls, it has clear utility. The most useful information is the
information about the certificate used by the signer. It can easily be used
to limit valid signers for phone numbers, preventing call spoofing. CNAM
can be used for this with minimal effort. Adding a campaign registry to
register high-volume originating numbers will provide a way for legitimate
call centers to whitelist their calls and for terminating carriers to
throttle the traffic. Snowshoeing would still be an issue, but I am unsure
how to address this without conducting a content analysis.

I think the primary reason for the Rich Call Data push is not to combat
robocalls or to present end users with the identity of the remote party. It
is a financial incentive for the wireless carriers, which can charge for
presenting Rich Call Data from the call originator, versus paying for a
CNAM database dip. Wireless carriers attempted to charge their customers
for displaying the caller's identity, but were unsuccessful in selling this
feature. So, they pushed for a new network protocol where they can charge
the call originator for this feature. In my case, I can send a SIP call
with call data directly to a wireless carrier. The wireless carrier
deliberately strips the caller's identity from this call. They are asking
for money (not the caller KYC) to keep it. So, this is not about users. It
is about revenue.

I believe a lot of people in this group are driven by their best
intentions, but it ends up in an administrative quagmire and telecom
revenue fiefdom anyway.

Best regards,
_____________
Roman Shpount


On Wed, Oct 8, 2025 at 9:53 AM Pierce Gorman <[e-mail]>
wrote:

> Well, this is a lively STIR discussion!
>
>
>
> Dear Roman,
>
>
>
> I will offer my view as the network design engineer that led the
> implementation of STIR/SHAKEN at Sprint, and former active participant of
> the FCC’s Call Authentication Trust Anchor working groups of the North
> American Numbering Council.  The fundamental reason STIR/SHAKEN exists is
> because unethical, immoral, or inept service providers are admitting
> illegal robocalls into the PSTN.  RCD, alone, will not change that.
>
>
>
> During the first CATA wg I warned Henning Schulzrinne when he was CTO of
> the FCC that the TRACED Act and STIR/SHAKEN would not be successful in
> bringing about a material decline in illegal robocalls unless the US could
> accomplish “critical mass” of signing and verification.
>
>
>
> I didn’t give a specific metric for critical mass in the context of
> STIR/SHAKEN but suggested something north of 90% of all call originations
> needed to be successfully signed and verified.  Even then success presumed
> analytics meaningfully identified bad calls, traceback based on analytics
> determinations was automated, and that the FCC Enforcement Bureau and/or
> other law enforcement agencies used the information to act swiftly and
> decisively.
>
>
>
> Obviously, that has not happened, and I question whether it ever can.  I
> believe it is impractical to make significant headway against illegal
> robocalling using cryptography and punitive measures alone.
>
>
>
> STIR/SHAKEN as conceived and implemented based on legislative and
> regulatory mandates is a costly and so far, failed, experiment.
>
>
>
> All that said, the TRACED Act and STIR/SHAKEN proved call authentication
> in the PSTN can be successfully accomplished, at scale.  That is very good,
> however, the misses still are:
>
> 1. We should have started with an Out-Of-Band (OOB) approach, possibly not
> even based on SIP, so that we could avoid the problems caused by transit
> networking and heterogenous inband call paths.
>
>
>
> 2. We should have focused on the sources of illegal robocalls, not ALL
> calls.  i.e., we should have focused our efforts on caller identification
> from call centers, as well as service providers (although the definition of
> “service provider” has proved damnably difficult.  Nobody, including and
> especially the FCC, knows how many VoIP call originators exist or how many
> originators aren’t even aware they would be considered a “service
> provider”).
>
>
>
> 3. Caller identification of non-service provider organizations should be
> done by professional KYC vetters who cryptographically attest as to the
> identity of the caller which is what iconective does when issuing an SPC
> token to a registered SP.  This reduces the possibility of collusion
> between service providers and their enterprise customers, and reduces the
> field of vetters that need to be accredited/authorized and held accountable
> by analytics, traceback, etc.
>
>
>
> 4. Caller identification schema need to recognize and accommodate the
> layered identities of modern telecommunications.  Few companies operate
> their own call center, but instead outsource the work to call center
> providers.  Digital identity and trust attribute (e.g., licensed? Insured?)
> credentials can expose the layers of identities and authorizations.  And,
> privacy-preserving proofs are possible where needed.
>
>
>
> 5. We should have lowered the barrier (as you also recognize) to permit
> callers (mostly call centers, but also enterprise mobiles, etc.) to
> register and access STI certificates attesting to their domain-verified
> identity (as a minimum).  Self-registered callers’ calls could be signed
> with RCD, not for the purpose of branded calling, but for the purpose of
> voluntarily identifying themselves as a measure to assist analytics
> providers more than anything to protect those callers’ calls from
> accidental labeling or blocking.  Further, with the identity of the caller
> (at least, and their service provider or providers, plural), robocall
> mitigation takes on a whole new level of capability.
>
>
>
> Respectfully,
>
>
>
> Pierce Gorman
>
>
>
> CONFIDENTIAL
>
> *From:* Chris Wendt <[e-mail]>
> *Sent:* Wednesday, October 8, 2025 1:44 AM
> *To:* Roman Shpount <[e-mail]>
> *Cc:* Brett Nemeroff <[e-mail]>; Richard Shockey <
> [e-mail]>; IETF STIR Mail List <[e-mail]>
> *Subject:* [stir] Re: [art] Re: For those of you who follow this kind of
> stuff.
>
>
>
> Removing art mailing list and only keeping stir mailing list on cc.  Let’s
> keep this on stir list only to avoid spamming art participants.
>
>
>
> RCD, I’m a fan, i guess the party is really kicking in to gear.
>
>
>
> -Chris
>
>
>
> On Oct 8, 2025, at 12:19 AM, Roman Shpount <[e-mail]> wrote:
>
> 
>
> Brett,
>
>
>
> FCC was very deliberate in not specifying the KYC requirements. This being
> said, all carriers introducing traffic to the US phone network should have
> a KYC policy described in the RMD database. Carriers that did not provide
> an adequate policy have been removed from the RMD database and are no
> longer permitted to originate traffic. Additionally, if, as a carrier, I
> can set the A-level attestation for the call based on my KYC policy, I
> should be able to specify the Rich Call Data accordingly, especially if
> this is required when A-level attestation is provided.
>
>
>
> I have a strong feeling that certain providers care more about creating
> new sources of revenue for themselves through regulatory arbitrage than
> about creating a healthy infrastructure to prevent robocalls. A glaring
> example is iConnectiv providing SPC tokens, but not the signing
> certificates, which artificially creates business for specialized
> certificate authorities. Ironically, this business opportunity is so small
> and labour-intensive that no one actually wants to do it, trying to
> shepherd carriers towards the hosted signing solution.
>
>
>
> To summarize, if, as a carrier, I am entrusted with an SPC token, I should
> be trusted to provide the Rich Call Data. If I am not trusted to provide
> Rich Call Data into the network, I should not be introducing any traffic
> into it. If the FCC mandates Rich Call Data, it should mandate that
> carriers accept it without creating walled gardens, with each carrier
> charging a fee to actually accept the data.
>
>
>
> Finally, if we intend to mandate the transmission of personally
> identifiable data with every call, we need to update SIP with a scalable
> and secure transport protocol. Most current carrier SIP implementations
> still use UDP. SIP-over-TLS suffers from head-of-the-line congestion
> issues. SIP is in dire need of a secure datagram-based protocol, such as
> QUIC. I am surprised that no one from the STIR group brought this to the
> SIPCore, so that a more scalable and secure protocol capable of carrying
> Rich Call Data could be standardized.
>
>
>
> Best Regards,
>
> _____________
> Roman Shpount
>
>
>
>
>
> On Tue, Oct 7, 2025 at 8:42 PM Brett Nemeroff <
> [e-mail]> wrote:
>
> Hello Roman,
>
>
>
> In my opinion, US Carriers are unlikely to accept vanilla RCD data because
> of the lack of defined KYC.  RCD is a very good vehicle for delivering the
> RCD, but it depends upon implicit trust of the originating service
> provider. “Vanilla” RCD offered like this to terminating service providers
> gives no assurance to the terminating service provider that the originator
> performed any specific KYC.
>
>
>
> CTIA’s BCID is based on RCD but details an ecosystem with specific KYC
> requirements. Participating in this ecosystem will allow for the delivery
> and native presentation of RCD.
>
>
>
> It’s worth noting that without a defined ecosystem for RCD such as BCID,
> RCD provides little (trust)  benefit over traditional CNAM other than the
> fingerprints of the originating service provider for enforcement purposes.
>
>
>
> -Brett
>
>
>
>
>
>
>
> Brett Nemeroff
> VP of Engineering - Voice
> [e-mail] <%7BE-mail%7D> | 1-512-203-3884
>
> <https://www.numeracle.com/>
>
> *<C2_signature_logo_ccb8cfe8-4171-4801-978c-931782d067de.png>
> <https://www.numeracle.com/>*
>
>
>
> Empowering Calls with
> Identity Management
> <https://www.numeracle.com/insights/entity-identity-management-to-empower-your-calls>
>
>
>
>
> CONFIDENTIAL
>
> *From: *Roman Shpount <[e-mail]>
> *Date: *Tuesday, October 7, 2025 at 7:24 PM
> *To: *Richard Shockey <[e-mail]>
> *Cc: *IETF STIR Mail List <[e-mail]>, [e-mail] <[e-mail]>
> *Subject: *[stir] Re: [art] Re: For those of you who follow this kind of
> stuff.
>
> You don't often get email from [e-mail]. Learn why this is
> important <https://aka.ms/LearnAboutSenderIdentification>
>
> In my day job, I see a lot of robocalls coming through the LEC local
> switches as TDM, as local re-origination with spoofed ANI.
>
>
>
> I would also love to sign Rich Call Data with my SPC token and not have
> wireless carriers discard this data. If I provide the information about my
> customer, I am unsure why I need to pay someone else to sign this
> information.
>
> _____________
> Roman Shpount
>
>
>
>
>
> On Tue, Oct 7, 2025 at 8:11 PM Richard Shockey <[e-mail]> wrote:
>
>
>
> It wont . You mean the legacy TDM/SS7 crap…this is the beginning of
> mandating all SIP in the US realtime US voice network as the British have
> done.
>
>
>
> I would not want to own a Tandem Access network.
>
>
>
> The US industry is pretty clear on this.  You only need to read the FCC
> 17-97 docket at the FCC ECFS website to understand where the players
> actually are.
>
>
>
> This again is my day job.
>
>
>
>
>
> Richard Shockey
>
> Shockey Consulting LLC
>
> Chairman of the Board SIP Forum
>
> www.shockey.us
>
> www.sipforum.org
>
> richard<at>shockey.us
>
> Skype-Linkedin-Facebook –Twitter  rshockey101
>
> PSTN [phone]
>
>
>
>
>
> *From: *Roman Shpount <[e-mail]>
> *Date: *Tuesday, October 7, 2025 at 7:37 PM
> *To: *Richard Shockey <[e-mail]>
> *Cc: *IETF STIR Mail List <[e-mail]>, <[e-mail]>
> *Subject: *[art] Re: [stir] For those of you who follow this kind of
> stuff.
>
>
>
> How would this work with PSTN links?
>
> _____________
> Roman Shpount
>
>
>
>
>
> On Tue, Oct 7, 2025 at 6:59 PM Richard Shockey <[e-mail]> wrote:
>
>
> The United States government is going to mandate Rich Call Data in the
> network.
>
> https://docs.fcc.gov/public/attachments/DOC-415059A1.pdf
>
>
> Richard Shockey
> Shockey Consulting LLC
> Chairman of the Board SIP Forum
> www.shockey.us <http://www.shockey.us>
> www.sipforum.org
>
> richard<at>shockey.us
> Skype-Linkedin-Facebook –Twitter rshockey101
> PSTN [phone]
>
>
>
>
>
>
> _______________________________________________
> stir mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>
> _______________________________________________ art mailing list --
> [e-mail] To unsubscribe send an email to [e-mail]
>
> _______________________________________________
> art mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]
>
>