ONCP — Open Network for Content Protocol
A personal framework for open infrastructure in India's telecom stack — starting with what happened to local cable operators after Jio
Started from watching my father's cable business get structurally bypassed — not outcompeted. Since Jio, telcos own all five layers: network, content, device, application, and data. LCOs became dumb pipes with no leverage anywhere in the stack. ONCP is a framework for a five-layer open protocol stack — content delivery, shared infra, device neutrality, identity and billing, edge AI — that gives LCOs a role in infrastructure the same way ONDC gave kirana stores a role in e-commerce. Inspired by ONDC and UPI. Not a startup pitch. A problem I keep coming back to.
At a glance
The Story
The Year Plans Collapsed
My master's plans fell through. Not with a bang — more like a slow deflation that left me in a kind of suspended state. For someone whose identity had been built around strategic ambition, the sudden absence of a forward structure was disorienting. I stopped doing the things that required planning. Stopped reading research. Stepped back from every project that felt too calculated. What I didn't stop doing was observing. That period of stillness turned my attention inward and homeward. And what I found there completely reoriented what I thought I cared about.
My Father's Cable Business
My father is a local cable operator. During those months of detachment from my own plans, I finally saw what had been in front of me for years — the dignity and fragility of what he built. He wasn't chasing growth. He was protecting something — a livelihood, a service, a web of trust with the neighbourhood. I saw the stress of strained margins, the invisible technical debt of aging hardware, and the quiet resilience of someone competing against giants without institutional backing. LCOs are the last mile of India's content delivery story — millions of subscribers, deeply embedded in their communities. And they are quietly being structurally erased. Not outcompeted. Bypassed.
What Actually Changed — Vertical Integration, Not Growth
The telecom industry didn't expand. It vertically integrated and captured the entire digital consumption stack. That's a power shift, not a feature upgrade. Pre-2016: Telcos were bandwidth providers. LCOs handled last-mile TV. OTT was a separate ecosystem. Value chain was fragmented — and that fragmentation gave space for smaller players. Post-Jio disruption: Telcos now control every layer. Not just connectivity — content, devices, applications, data, and increasingly AI. This isn't competition anymore. India is moving toward a controlled duopoly: the Jio ecosystem and the Airtel ecosystem. Two closed loops. Zero leverage for anyone outside.
What telcos now control — all five layers of the stack
| Layer | What Telcos Own Now | Example |
|---|---|---|
| 1. Network | FTTH fiber, 4G/5G rollout, satellite backhaul incoming | Jio Fiber, Airtel Xstream Fiber |
| 2. Content | OTT bundling + own streaming platforms | JioCinema, Airtel Xstream, Hotstar bundle |
| 3. Device | Locked set-top boxes + Android TV ecosystems | Jio STB, Airtel 4K STB — no third-party OS |
| 4. Application | Super apps, payments, commerce integrations | Jio app, Airtel Thanks, embedded UPI |
| 5. Data + AI | Personalization engines, LLM integration (emerging) | Viewing graphs, recommendation AI — all proprietary |
Why LCOs Are Structurally Losing — Not Outcompeted, Bypassed
LCOs have genuine strengths: local trust, physical presence, a last-mile network already built. In the old world, those were durable competitive advantages. In the new world, they are irrelevant to the layers that matter. The real shift: LCOs moved from 'service provider' to 'dumb pipe with zero leverage.' They have no control over content licensing, OTT distribution, device ecosystems, data, or AI. They are being bypassed — by smart TVs that go directly to WiFi, by telco bundles that eliminate the need for a separate cable subscription, by satellite internet that will reach rural India before most LCOs can upgrade. The LCO doesn't lose on price. It loses on stack depth.
Telco vs LCO — control across every layer
| Layer | Telco (Jio / Airtel) | LCO |
|---|---|---|
| Network | Owned + scaled nationally | Local + fragmented |
| Content | Licensed + bundled | None |
| Device | Controlled ecosystem | Commodity hardware |
| Data | Centralized + mined | None |
| AI | Emerging — in-device LLMs | Zero |
| Pricing power | High | Effectively dead |
| Ecosystem | Closed loop | Isolated |
ONCP — Protocol Over Platform
ONDC changed something in how I thought about infrastructure. Here was a government-backed initiative that didn't try to build a monopoly — it tried to build a protocol. A network of networks. Buyers and sellers interoperating through a shared standard rather than being locked inside a platform. The ONCP insight followed exactly that logic: don't compete with Jio. Build underneath it. A parallel protocol economy — one that telcos can't easily absorb because it's publicly governed, federated, and open by specification. LCOs don't need a saviour. They need a protocol layer that turns them from dumb pipes into sovereign infrastructure nodes. The LCO's real advantage — physical presence, community trust, last-mile reach — becomes valuable again only if there's a protocol that rewards it.
The Architecture — Five Open Layers
ONCP isn't a platform. It's a spec — a layered open architecture that any LCO, content creator, or regional broadcaster can build on top of, the same way any bank can build on UPI rails. National coordination, local execution. No gatekeepers. No middlemen between creator and subscriber.
ONCP — Five-layer open architecture
| Layer | What It Does | Infrastructure |
|---|---|---|
| 1. Open Content Network | Standardized OTT distribution API — any LCO plugs in, any creator lists content. No platform intermediary. | Open protocol spec (ONDC-model) |
| 2. Shared Delivery Infra | National CDN layer on public backbone — LCOs stop negotiating bandwidth individually. | BharatNet, BBNL, BSNL, CSCs |
| 3. Device Neutral Layer | Open, regulated set-top OS — no telco lock-in. Any certified device participates. | Android-derived OS, regulated by TRAI |
| 4. Identity + Billing | Direct payments from subscriber to creator or LCO — no platform taking margin. | Aadhaar identity, UPI rails |
| 5. AI Layer | Localized language models, regional content intelligence, community knowledge systems — running at the edge, not in a central data centre. | Edge compute at LCO nodes, open weights |
The Brutal Reality Check
The honest version: this idea has real structural problems that would kill most proposals before they start. Naming them clearly is more useful than pretending they don't exist. The strategic reframe is what matters: the goal is not to compete with Jio. That's suicide. The goal is to build a parallel protocol economy underneath — in the layers telcos are weak or absent: regional content, government and education distribution, rural intelligence, LCOs converted from operators into edge compute nodes. UPI didn't replace banks. It built rails that banks had to use. That's the model.
Four real objections — and the way around each
| Objection | Why It's Real | Strategic Response |
|---|---|---|
| Government execution gap | UPI worked with clear financial incentive + central mandate. Content is messy, political, fragmented. | Define ONE layer first. Regional content API is the cleanest, least political entry point. |
| Incentive mismatch | Netflix won't join. Jio won't open its APIs. Not unless forced by regulation. | Don't build for them. Build for what they ignore: hyperlocal, language-first, rural-first content. |
| LCO capability gap | Even with infra, most LCOs lack tech capacity and scale thinking. | LCO as node, not operator — strip all complexity, give edge hardware. Protocol handles everything above. |
| Consumer indifference | Users want cheap, fast, easy — not a decentralized ecosystem. | Consumer never needs to know. UPI won because the protocol was invisible. ONCP should be the same. |
Where This Goes
The goal isn't to save LCOs. It's about who controls digital distribution in India — and whether that can move toward open infrastructure rather than closed platforms. I'm honest about the gaps: I don't have deep knowledge of telecom business, policy negotiation, or how to build industry coalitions. The consumer side is documented elsewhere in this portfolio. The business and policy mechanics of this space are territory I'm still working through. If you work at the intersection of DPI, telecom policy, or open infrastructure in India — I'd genuinely like to connect. Not looking for validation. Looking for people who know this side of the system.