notes from the fcc ip transition workshop

The FCC’s Wireline Competition Bureau held a two-day IP Transition Workshop on July 15–16, 2026, at the Commission’s headquarters in Washington. The framing was regulatory — the workshop supports the Commission’s proposed reforms to accelerate the move off legacy TDM voice: accelerating all-IP interconnection, completing the transition of intercarrier compensation to bill-and-keep, and reforming the High-Cost Universal Service Fund for an all-IP future. Chairman Brendan Carr opened the first day with the throughline that if the industry gets the transition right, it strengthens the communications infrastructure, encourages investment, and improves resilience. What follows are my notes — an account of the first day, with a general overview of the four panels and a longer look at my own panel on consumer protection, and a shorter summary of the second day, which was given over to the economics of the transition.

The recording of day one is on the FCC’s channel: IP Transition Workshop, day 1. My panel begins at 5:58:00.

Panel 1 — the right interconnection framework for an all-IP world

The opening panel stayed on the practical mechanics of actual IP interconnection rather than the theory. Much of the conversation orbited the universal-service constraint — keeping services and rates reasonably comparable under section 254(b)(3) of the Act as the underlying network changes — and how an interconnection framework should be structured so investment and resilience improve rather than fragment. Panelists spanned the policy and carrier spectrum: USTelecom, Public Knowledge (Harold Feld), Range, Conterra Networks, and Inteliquent/Sinch. The recurring tension was the familiar one: how prescriptive the Commission should be about the framework versus letting commercial arrangements settle it.

Panel 2 — considerations in transitioning to IP interconnection

The second panel brought the regional and rural carrier perspective — FirstLight Fiber, AT&T, SDN Communications out of South Dakota, Verizon, and Nsight. The honest theme here was that the easy part is largely done: much of the network has already migrated, on the order of 75–80%, and what remains is the hard tail — the routes where TDM interconnection persists because IP interconnection is not yet economical or technically feasible. Panelists walked through the ongoing operational challenges, the dependence of everything on the underlying Ethernet/IP transport, and how funding mechanisms like BEAD intersect with the buildout that the last stretch of the transition depends on.

Panel 3 — public safety considerations in transitioning to IP interconnection

The public-safety panel focused on the NG911 transition — its current status and the barriers facing 911 authorities that want to move to an all-IP NG911 system. The panel drew on NENA (the National Emergency Number Association, representing tens of thousands of public and private PSAPs), NASNA and state 911 leadership, and carrier voices from Verizon, Bandwidth, and Intrado. The through-question was how to close the gap between the PSAPs and providers that are ready for NG911 and those still on legacy infrastructure, so that no part of the emergency-calling system is stranded as the rest of the network moves to IP.

Panel 4 — consumer protection and the IP transition

My panel closed the day. It was co-moderated by Chris Laughlin (deputy chief, Competition Policy Division, Wireline Competition Bureau) and Michael Scott (deputy chief, Consumer and Governmental Affairs Bureau, Disability Rights Office) — a pairing that set the tone, because it deliberately put consumer protection and accessibility in the same conversation. Alongside me were David Bahar (Telecommunications Access of Maryland), Rebekah Johnson (Numeracle), Joshua Ruby (Granite Telecommunications), and Jim Tyrell (TNS).

The consumer-protection half of the discussion was grounded in where call authentication actually stands, and the honest answer is uneven: STIR/SHAKEN signing is near-universal between the Tier-1 carriers but drops off sharply below them, full A-level attestation is still the exception rather than the rule, and bad actors have every incentive to route through the seams. The seam that kept coming up was TDM in the middle — a legacy hop between two IP endpoints that strips the signed identity and the rich call data along with it, so that even a call that was properly signed at origination can arrive stripped of its authentication. That gap is not incidental to the IP transition; it is one of the strongest arguments for finishing it.

what I discussed

I introduced myself in my Somos role — VP of System Engineering, at the company that administers the North American Numbering Plan, the toll-free number system, and the reassigned-number database, and that spends a lot of its time on the robocall and fraud problem. I also wear the standards hats: co-author of the core STIR/SHAKEN specifications and industry co-chair across the IP-NNI Task Force and the STI-GA Technical Committee.

My main argument was that the IP transition lets us solve two problems with one solution. Bringing call authentication and all-IP together — with certificates as the trust layer — is not two separate initiatives competing for attention; it is one architecture. It also validates a great deal of the work the industry has already done. The telephone number is how people identify themselves on the network, and an all-IP fabric lets us make the association between a number and the entity authorized to use it clear, cryptographic, and anchored at the certificate level rather than inferred hop by hop.

I pushed on the numbering side of this too. As the network becomes less geographic, the old static assumptions baked into numbering — a number tethered to a rate center, an identity inferred from where a call appears to originate — stop earning their keep. The all-IP world lets numbering become dynamic, and lets the association between a number and the authorized entity be asserted cryptographically rather than presumed from legacy plumbing. That is the same reform that makes call authentication and the know-your-upstream-provider work actually enforceable: once the binding is explicit and signed, you no longer have to trust the geography or the hop-by-hop inference the old network forced on us.

From there I made the case that TDM in the middle is the thing to eliminate, not work around. The rich information a signed IP call can carry — the identity, the attestation, the branded call data a consumer actually sees — survives end to end only if the path is end to end. Every TDM segment is a place that information drops. On attestation specifically, I made the point that B- and C-level attestation were never meant to be permanent fixtures; they were accommodations for an incomplete network, and the goal is to move the population of calls toward genuine A-level attestation. That connects directly to the Commission’s know-your-upstream-provider work, which I think is beneficial precisely because it pushes accountability down to the providers actually originating and signing traffic.

The part of my contribution I care about most tied back to the accessibility framing the moderators set up. I have spent time on this outside STIR/SHAKEN — co-chairing the NANC Interoperable Video Calling work and serving on the Disability Advisory Committee when real-time text was first taken up — and the point I wanted on the record is that an all-IP network can make real-time text, video relay, multimedia communication, and NG911 native, primary capabilities of the telephone network rather than bolt-ons grafted onto a voice path that was never designed for them. Making rich call data a native part of the network is the same move: identity and context as first-class properties of the call, not accessories. I credited David Bahar directly — my own first real experience of video relay was in conversations with him years ago — and argued that the transition is the moment to design for these users deliberately instead of retrofitting later.

On the “when,” which the moderators pressed, my answer was that now is the time, and that a workable compromise is to have the carriers and the parties most affected come back with concrete milestones rather than an open-ended commitment. The urgency isn’t abstract: the longer the legacy seams persist, the longer the authentication gap and the accessibility gap both stay open.

day two — the economics of the transition

Day two turned to the money, which is where the transition either accelerates or stalls. Commissioner Trusty opened the morning; the two morning panels took up completing the move of intercarrier compensation to bill-and-keep, and the afternoon turned to reforming the high-cost and Universal Service mechanisms — A-CAM, CAF-ICC, the legacy access charges — for an all-IP world. The panelists were the carriers and trades who actually carry the cost: Verizon, AT&T, Lumen, Comcast, and Inteliquent on the interconnection-economics side; NCTA and NTCA on the trade side; and the rural and satellite voices — Pelican Broadband, Golden West, Conexon Connect, NECA, Brightspeed, and SpaceX — on the high-cost side.

The economics aren’t my center of gravity, but they matter, and the honest read is fairly simple. Completing the shift to bill-and-keep is a genuine simplifier. It drains the legacy intercarrier-compensation rules that distort incentives and give providers a reason to keep TDM arrangements alive to protect a revenue flow — take that away, and one of the biggest reasons the last stretch of TDM persists goes with it.

The harder conversation is the one about funding, and it’s the elephant in the room the whole industry recognizes. The recurring ask from parts of the industry is for more time and more support to pay for the transition, and I don’t want to be glib about it: standing up new IP equipment is real capital, the smallest carriers genuinely struggle to raise it, and the business pressures in the voice space have been unrelenting for years. But “we need more time to fund this” is not, on its own, a strong argument. This day has been coming for more than a decade — the tech-transition proceedings are not new, and nobody in this industry can honestly claim they were caught off guard. A provider that hasn’t been planning and budgeting for the end of TDM is in a position that’s genuinely hard to excuse — even if, given how voice economics actually work, it’s not the least bit surprising. The right posture, I think, is real sympathy for the genuinely hard cases paired with a firm expectation that the transition proceeds: targeted support where the need is demonstrable, not an open-ended delay for an industry that has had ample notice.

the throughline

Across both days the same structure kept surfacing from different angles: the all-IP network is not just a faster or cheaper voice network, it is the substrate that makes a set of long-standing goals finally native — verifiable caller identity, rich and branded call data, interoperable NG911, and accessible communication. The interconnection framework, the hard economics of the last stretch of TDM, the public-safety transition, consumer protection and accessibility, and the intercarrier-compensation and high-cost reforms that pay for all of it are really the same question asked at different layers: what has to be true for the network to carry trust and context end to end. That is the question I have been working on in the notebook and the library from the authentication side, and it was good to see it framed as a whole-network transition rather than a single-issue fix.