Mercury is tidally locked to the Sun — but not synchronously. It settled into a 3:2 resonance: captured but still spinning. That distinction maps onto the only realistic vendor strategy most organizations have.
Mercury takes 59 Earth days to spin once on its axis and 88 Earth days to orbit the Sun. Work out the arithmetic of sunrise to sunrise and you get a day that lasts 176 Earth days — exactly two Mercury years. A single Monday-to-Monday on Mercury takes longer than two full trips around the Sun.
This isn’t a quirk of astronomy. It’s a clue.
Mercury is tidally locked to the Sun, but not synchronously — not in the 1:1 ratio that glues our Moon’s same face toward Earth forever. Mercury settled into a 3:2 spin-orbit resonance: three rotations for every two orbits. Captured by gravity, but still spinning independently. According to Correia and Laskar’s work in Nature, Mercury reached this state within 10–20 million years of formation — fast by astronomical standards (Nature 429:848–850, 2004). The capture is front-loaded.
The distinction between 1:1 and 3:2 matters more than it seems. Because the physics that locked Mercury is the same physics that locks an enterprise into its vendor stack. And Mercury’s outcome — not escape, but preserved rotation — maps onto the only realistic strategy most organizations have.
Tidal locking begins with deformation. When a moon orbits a planet, the planet’s gravity isn’t uniform across the moon’s diameter — the near side feels a stronger pull than the far side. That gradient stretches the moon into a slight oval, raising tidal bulges on opposite ends. If the moon rotates faster than it orbits, the bulge leads the line connecting the two bodies. The parent body’s gravity tugs back on that leading bulge, creating torque. The torque slows the rotation. Energy dissipates as heat through the constant flexing.
The critical detail: the bulge IS the mechanism. The moon isn’t chained in place — it’s reshaped until its own structure enforces the lock.
Vendor lock-in works identically. An organization that adopts a major platform doesn’t get chained to it; it gets reshaped by it. Staff certifications align with the vendor’s curriculum. Data formats nest inside proprietary schemas. Workflows assume the platform’s specific capabilities. Disaster recovery plans depend on the vendor’s tools. Each adaptation is a tidal bulge — a structural deformation that makes the organization’s shape match the vendor’s gravitational field.
Nobody notices the lock happening because the reshaping feels like optimization. You’re getting better at using the tool. Your team is more efficient. The integration is tighter. That’s all true. It’s also how tidal locking works: the moon is most thermodynamically stable when the bulge faces directly toward its parent body. Efficiency and capture are the same process.
The time it takes for a body to become tidally locked follows a formula with a brutal exponent. Among the variables — mass, rigidity, orbital distance — one dominates:
Locking time scales with the sixth power of orbital distance.
Double your distance from the parent body and locking takes 64 times longer. Halve it and locking is 64 times faster. This isn’t a gentle slope. It’s a cliff.
Translated to vendor lock-in: “orbital distance” is depth of integration. A company using a cloud provider for commodity storage is at a great distance — loosely coupled, easy to migrate. A company running custom machine learning pipelines on proprietary APIs, with data in vendor-specific formats, staff certified in vendor-specific tooling, and CI/CD wired through vendor-specific orchestration, has halved its orbital distance several times over. Each layer didn’t add a linear increment to switching costs. It multiplied them by a factor closer to 64x per halving than the 2x most executives assume.
This is why “we’ll migrate later” is an exponential bet against yourself. Every year of deeper integration isn’t one year harder to undo. It’s a move closer to the parent body, and the math of tidal capture says the energy required to escape grows at the sixth power of how close you’ve drifted.
Our Moon completed the tidal locking process billions of years ago. It now rotates exactly once per orbit — perfect 1:1 synchronization. The same face always points at Earth. The far side exists, but we never see it from the ground.
Even in this total lock, there’s a remnant of freedom. The Moon’s slightly eccentric orbit causes it to wobble — a phenomenon called libration — letting us see about 59% of its surface over time instead of exactly 50%. It’s a marginal liberty. Shadow IT of the celestial kind. Locked organizations similarly preserve tiny freedoms: a team running an unapproved SaaS tool, a developer writing scripts in a non-standard language, an engineer keeping personal notes in a format the vendor doesn’t own. These librations don’t change the fundamental capture, but they reveal something about it: even total synchronization can’t eliminate every degree of freedom. It can only make them irrelevant.
Meanwhile, the Moon recedes from Earth at about 4 centimeters per year — tidal energy slowly pushing the locked body outward. The lock persists, but the relationship erodes. Customer satisfaction with locked-in vendors follows the same trajectory: still captured, but drifting.
What happens when a locked body’s parent changes its behavior?
In November 2023, Broadcom completed its acquisition of VMware. What followed was the most dramatic demonstration of vendor lock-in consequences in recent enterprise history. Licensing costs increased by 800–1,500% (The Register, 2025). A representative 10-server environment that cost $40,000–$43,000 per year before the acquisition jumped to $200,000–$270,000 (Software Pricing Guide, 2025). Perpetual licensing was eliminated. The minimum core requirement per CPU rose from 16 to 72. Late renewals incurred a 20% surcharge.
But the most striking number is structural, not financial. The number of authorized VMware Cloud Service Providers dropped from over 4,500 to approximately 13 (Software Pricing Guide, 2025).
In orbital mechanics, that’s not just increasing gravitational pull — it’s clearing the neighborhood. Eliminating every smaller body that offered an alternative orbital path.
Gartner found that 74% of IT leaders began exploring VMware alternatives (CIO Dive, 2024). Yet Gartner’s own projections suggest only about 35% of VMware workloads will actually migrate to alternative platforms by 2028. That 39-point gap — between wanting to escape and actually escaping — is the tidal lock. The organizations that most need to leave are the ones most reshaped by what they’re trying to leave. Migration costs for mid-size environments run $50,000–$200,000 over 6–12 months (Software Pricing Guide, 2025), and that’s before you account for the human costs: retraining, process redesign, institutional knowledge that exists only in the shape of the old platform.
The sunk energy can’t be recovered, just as a tidally locked moon can’t reclaim the heat dissipated during its capture.
Mercury proves that total synchronization isn’t inevitable. Its orbital eccentricity — 0.2056, the highest of any planet — prevented 1:1 capture. At perihelion, Mercury moves too fast for its spin to synchronize; at aphelion, too slow. The 3:2 resonance is the stable compromise: captured but not synchronous.
The resonance produces observable strangeness. Near perihelion, the Sun appears to move backward in Mercury’s sky for about eight days. During the deepest gravitational engagement — closest approach, maximum tidal force — the partially locked body experiences its environment differently than a fully locked one would. This is the paradox of partial coupling: at the moments of greatest intensity, you see things the Moon never can.
Multi-cloud and open-standards strategies are Mercury’s 3:2 resonance. They don’t escape the gravitational field. A multi-cloud organization is still captured — still paying the coordination costs, still shaped by the platforms it uses. But it preserves independent rotation. The operational weirdness is real: different APIs, inconsistent identity management, multiple billing systems. That weirdness is the backward-moving Sun. It’s the price of not being the Moon.
Cloud providers, for their part, have built gravitational wells with deliberate asymmetry. Dave McCrory coined the term “data gravity” around 2010 to describe how data attracts services and applications — the more data you store, the harder it is to leave. The pricing structure literalizes the metaphor. Ingress is free or cheap. Egress — getting your data out — costs 5–6 times what storing it costs. Azure charges $0.087 per gigabyte to move data out versus $0.018 per gigabyte to store it; Google Cloud charges $0.12 versus $0.020 (Backblaze; CloudOptimo). Moving 50 terabytes costs $3,500–$7,000 in egress fees alone, before any migration tooling or staff time.
In 2024, Google Cloud, Azure, and AWS announced they would waive egress fees — but only for customers migrating entirely off their platforms (ConsoleConnect, 2024). We’ll give you escape velocity, but only if you leave the orbit completely. No partial unlocking. Mercury’s 3:2 resonance isn’t on the menu.
Three places, ordered by how much they matter.
First: Tidal locking has no agency. Moons don’t choose their orbits. Organizations do — or at least they did, before the lock set in. This means the preventive window for organizations is real and actionable in a way it never is for celestial bodies. Mercury couldn’t have chosen a wider orbit. A CTO can. The moment of architectural decision — which cloud, how deep, what standards — is the moment that determines whether you end up in a 1:1 or 3:2 resonance. After that, physics takes over.
Second: In celestial mechanics, the parent body doesn’t benefit from the lock. Earth gains nothing from the Moon’s synchronization. Vendors benefit enormously. The entire pricing structure is built to leverage capture — ingress free, egress expensive; deep integration rewarded with discounts that increase dependency. The gravitational field is being actively tuned by someone who profits from your lock. Moons don’t have to contend with that.
Third: Tidal locking is permanent absent external perturbation. Vendor lock-in is, in principle, reversible through sufficient investment. The Moon will be locked for billions of years. An organization with enough budget, executive will, and time can migrate off anything. “Enough” is doing enormous work in that sentence — but the door exists, even if most organizations never walk through it.
In orbital mechanics, escaping a gravitational lock requires external perturbation — a giant planet disrupting the orbit, a collision providing new angular momentum. Without external force, tidal locking is thermodynamically irreversible. The energy has already dissipated as heat and cannot be recovered.
The organizational equivalents are regulation, market disruption, and crisis. The EU Data Act functions as a giant planet — an external gravitational force compelling data portability. The 2024 egress fee waivers were a response to that regulatory mass, not a gesture of goodwill. Market disruption — a new competitor offering radical portability, or open-source alternatives reaching parity — provides the angular momentum of a collision. Organizational crisis — acquisition, near-bankruptcy, a leadership change that resets institutional inertia — forces the rebuild that nobody would have chosen voluntarily.
But escape doesn’t undo the sunk costs. Regulation doesn’t un-train your certified staff, un-write your proprietary integrations, or un-format your data. It creates escape velocity. It doesn’t erase the heat.
The Pluto-Charon system offers one more cautionary image: both bodies are tidally locked to each other. Neither can rotate independently. Some enterprise relationships reach the same state — vendor and customer so deeply co-dependent that neither can change without the other. Mutual lock-in. The most stable configuration, and the hardest to escape.
Here’s the sentence worth carrying out of this essay: if your organization is deeply integrated with a single vendor, the energy required to switch is not twice what it was three years ago. It’s governed by the sixth power of how much closer you’ve drifted.
Mercury’s lesson isn’t that escape is easy. Mercury never escaped. Its lesson is that the moment of capture determines the outcome — that the eccentricity of your orbit at the point of initial lock determines whether you end up in synchronous rotation or something more livable. Once Mercury settled into 3:2, it stayed there. Once the Moon settled into 1:1, it stayed there too. The architecture of the relationship is set early and enforced by physics.
One Mercury day is still two Mercury years. The planet is still locked. But when the Sun crawls backward across Mercury’s sky during those eight days near perihelion — during the moment of deepest gravitational engagement — it sees something the Moon never will.
A different angle.
Sources: A. Correia, J. Laskar, “Mercury’s capture into the 3/2 spin-orbit resonance as a result of its chaotic dynamics,” Nature 429:848–850, 2004. The Register, “VMware pricing,” 2025. Software Pricing Guide, “VMware by Broadcom: Licensing, Pricing, and Packaging Guide,” 2025. CIO Dive, “Gartner: 74% of VMware customers exploring alternatives,” 2024. Backblaze; CloudOptimo (cloud egress pricing comparisons). ConsoleConnect, “Cloud egress fee waivers,” 2024. D. McCrory, “Data Gravity,” blog post, c. 2010.
The tidal forces in this essay don’t stop at cloud infrastructure.
Every AI agent whose decision trail lives inside a single vendor’s logging system is drifting toward 1:1 lock. Chain of Consciousness creates portable, vendor-neutral provenance — every agent decision signed and timestamped in a chain you own, not one you rent. Your agent can change models, change clouds, change orchestration frameworks, and its decision history comes with it. Mercury’s lesson applied to agent trust: preserve your independent rotation before the lock sets in.
Verify an agent’s decision chain · Explore vendor-neutral trust infrastructure · pip install chain-of-consciousness