On Vendor Lock-In and Cloudflare

Honest Dev say: "Freedom is illusion. Value is real."

This is a collection of thoughts on vendor lock-in, the fear of it, and why I have decided to stop worrying about it.

The Sacred Wisdom

Every developer has heard the gospel. Thou shalt not lock thyself to a vendor. Thou shalt remain portable. Thou shalt be free.

• The Advice
  • "Don't use proprietary APIs."
  • "Always have an exit strategy."
  • "What if they raise prices? What if they shut down?"
  • "Use open standards. Stay agnostic. Keep your options open."
    In the voice of that famous racecar developer. not the lambo guy
• The Reality
  • This advice is correct.
  • It is also easier said than done.
  • Much easier.

The Illusion of Freedom

Some developers advocate for self-hosting. Get a VPS. Run your own servers. Own your infrastructure. Be free.

• The Argument
  • "I use a VPS. I'm not locked in."
  • "I self-host everything. Complete control."
  • "I can migrate anytime."
• The Truth
  • Your VPS runs on someone else's hardware.
  • Your "freedom" depends on Hetzner not changing their TOS.
  • Your "independence" relies on DigitalOcean not raising prices.
  • Your "control" assumes AWS will keep that instance type available.

The uncomfortable truth: Unless you own the metal, the power supply, the network backbone, and the building it sits in, you are locked into someone's platform. You are just choosing which lock.

The Cloudflare Situation

Cloudflare has been quietly building everything. And not quietly. Mostly not quietly.

• What They Have
  • Domain registration. Buy your domain.
  • DNS. Point it somewhere.
  • Workers. Run your code at the edge.
  • Pages. Host your frontend.
  • R2. Store your files. No egress fees.
  • D1. SQLite at the edge.
  • KV. Key-value storage.
  • Durable Objects. Stateful compute.
  • Queues. Background jobs.
  • Rate limiting. Bot protection. DDoS mitigation. SSL.
  • Workflows. (I haven't used this yet)
• What They Are Building
  • Email sending. The last piece. For my current needs.
    When this ships, it is a complete platform.
  • Domain to deployment to email to security. End to end. One dashboard.

The Confession

I have thought about vendor lock-in. I have considered the risks. And I have decided:

• I am going all in.
  • All my domains. Cloudflare.
  • All my sites/APIs. Cloudflare Workers.
  • All my storage. R2.
  • All my databases. D1.
• The Reasoning
  • Even if I am locked in, I have already benefited too much.
  • THE DASHBOARD IS GETTING THE GREATEST FACELIFT EVER.
  • The pricing is aggressive. In the good way.
  • The integration between services is seamless.
  • I am shipping faster than I ever have.

The math: Years of productivity gains versus the hypothetical future where I need to migrate. I will take the gains.

The Acknowledgment

This is not advice. This is a personal choice. Here is what I acknowledge:

• I could be wrong. Cloudflare could change. They could raise prices. They could sunset services. Corporations do that.
• I might change my mind. In a year. In five years. Maybe tomorrow. Preferences evolve. Situations change.
• But today? Today I am putting all my eggs in one orange basket. And I am okay with it.

Conclusion

Vendor lock-in is real. The fear of it is valid. But the fear of it can also paralyze you into building nothing while waiting for the perfect, portable, future-proof solution that does not exist.

Sometimes you pick a platform. Sometimes you commit. Sometimes you accept the trade-off and ship.

Lock me in, Orange Cloud. I am ready.