Politicians love a good infrastructure crisis. When an internet outage hits Day 3, MPs rush to the microphones, local news outlets run sob stories about disconnected households, and the collective outrage machine starts chanting for heads to roll on a platter. They call it "disgraceful." They demand multi-million-dollar fines. They act as if a temporary digital blackout is a collapse of civil society.
They are entirely wrong.
The lazy consensus among public officials and tech journalists is that 100% network uptime is both achievable and desirable. It is neither. In fact, demanding flawless uptime is the single biggest threat to long-term digital resilience.
I have spent fifteen years architecting enterprise networks and dealing with the fallout of massive systemic failures. Here is the uncomfortable truth the telecom giants will not admit and politicians are too naive to understand: continuous uptime breeds fragile systems. Outages are the brutal, necessary stress tests that keep society from building digital houses of cards.
The Myth of Five-Nines Uptime
Every service-level agreement (SLA) promises the holy grail of 99.999% uptime. It sounds impressive. In reality, it is a marketing gimmick that creates a false sense of security while actively draining budgets that should be spent on true redundancy.
When a network operates flawlessly for years, two things happen. First, the team running it gets soft. They forget how to handle a catastrophic failure because they never see one. Second, the businesses relying on that network stop investing in their own backup plans. They assume the pipe will always flow.
Imagine a scenario where a major regional hub stays up for a decade without a single blip. Local hospitals, emergency services, and financial institutions rely solely on that primary connection. When the inevitable black swan event occurs—whether it is an accidental fiber cut by a rogue backhoe or a sophisticated state-sponsored cyberattack—the entire region collapses into chaos because nobody remembers how to operate offline or switch to an alternative network.
An outage is not a disgrace. It is a reality check. It exposes the hidden dependencies and single points of failure that accumulate during times of peace.
The High Cost of Eradicating the Last One Percent
To move a network from 99% availability to 99.99% requires doubling the infrastructure spend. To move it that final fraction of a percent costs exponentially more. We are talking about billions of dollars funneled into redundant hardware, overlapping fiber routes, and round-the-clock engineering staff just to prevent a few hours of downtime a year.
Where does that money come from? It gets passed down to the consumer in the form of higher subscription fees, or it gets siphoned away from expansion projects. By demanding absolute perfection in existing networks, politicians are actively delaying the rollout of fiber and satellite coverage to rural communities that have no high-speed access at all.
We are sacrificing broad accessibility on the altar of localized perfection. It is bad economics, and it is worse public policy.
Why Redundancy Should Be Your Problem, Not the Provider's
When the internet goes down, small business owners complain about lost revenue. But if your entire business model collapses because a single internet pipe stops working for forty-eight hours, you did not have a reliable business. You had a gamble.
Relying on a single internet service provider (ISP) for mission-critical operations is negligence. True resilience does not mean praying your primary provider never fails; it means designing your operations so that a failure is merely an inconvenience.
- Dual-homing is mandatory: If connectivity dictates your survival, you need connections from two distinct providers using different physical pathways (e.g., fiber and cellular or satellite).
- Decentralized data architectures: Local caching and offline capabilities should be built into every critical software stack. If your point-of-sale system cannot process a transaction without talking to a cloud server thousands of miles away, the software design is fundamentally flawed.
- Regular chaos testing: Companies like Netflix popularized the concept of intentionally breaking their own production environments to ensure their systems can self-heal. Telecom operators and enterprise clients need to adopt the same mindset.
Admittedly, this approach shifts the financial and operational burden onto the end-user. It requires investment, planning, and a departure from the convenient lie that technology never breaks. But it is the only way to build a society capable of weathering a true infrastructure crisis.
Stop Demanding Fines and Start Demanding Transparency
When an outage drags into a third day, the immediate political reaction is to threaten the telecom operator with massive financial penalties. This satisfies the public desire for retribution, but it solves absolutely nothing.
Fines simply remove capital from the exact companies that need to invest in fixing the physical infrastructure. It incentivizes operators to patch up broken networks with quick fixes and temporary workarounds just to stop the bleeding, rather than taking the time to rebuild a compromised section correctly from the ground up.
Instead of demanding financial penalties that vanish into government coffers, we should demand total architectural transparency.
We need public mapping of fiber routes, open reporting on hardware vulnerabilities, and clear documentation of how traffic is routed during an emergency. If consumers and businesses can see exactly how fragile a provider's network layout is, the market will naturally reward the operators who build for resilience rather than those who just promise high uptime numbers on a glossy brochure.
The current outrage cycle is lazy. It allows politicians to look tough while ignoring the systemic vulnerabilities that they themselves overlook when approving zoning laws, construction permits, and infrastructure budgets.
Stop crying over a dead connection. Buy a backup satellite terminal, rewrite your fragile software, and learn to value the occasional blackout for what it truly is: the only honest audit your technology stack will ever get.
