What Are Data Centers Actually For?

People keep asking a reasonable question about the data centers being proposed in and around San Angelo:

What are all these data centers actually for?

What data is going into them?

Why do we need so many?

Who benefits from them?

And if San Angelo is going to host this infrastructure, what does San Angelo get in return? Those questions deserve better than either extreme: treating every concern as fearmongering, or treating every data center as something inherently sinister.

I think it helps to start with something much simpler

We build shared infrastructure because specialization works. Nearly everyone could theoretically own a computer powerful enough to run their own applications, store their own data, host their own websites and provide their own online services.

Businesses could do the same thing. They could buy racks of servers, construct secure rooms for them, install redundant electrical systems, manage cooling, maintain backup generators, hire engineers and security professionals, and build connections to multiple telecommunications providers.

Some companies do.

Most don't.

For the same reason that every business doesn't generate its own electricity, manufacture its own vehicles or build its own telecommunications network, we increasingly rely on specialized infrastructure providers to handle computing.

A data center is part of that infrastructure.

What we casually call "the cloud" ultimately has to exist somewhere. Behind the apps on our phones, online banking, streaming services, business software, artificial intelligence, medical systems and countless other services are physical computers running in physical buildings.

Those buildings are data centers.

The value isn't confined to the building

This is where I think some of the discussion about economic benefit becomes too narrow. When we ask whether a data center benefits San Angelo, we naturally ask how many permanent jobs it will create.

That's an important question.

But it isn't the only measure of what infrastructure does.

A highway isn't valuable only because of the number of people employed maintaining the highway. Its greater economic value comes from everything people can do because the highway exists. Computing infrastructure works in much the same way.

A data center doesn't simply support the employees working inside it. It supports businesses that use its computing capacity to serve their own customers, employ their own people and build entirely different products.

I see this firsthand in my own professional life. I work for a software company based here in San Angelo. Like many modern software companies, we don't need to build and operate our own massive data centers. We rely on computing infrastructure operated by companies whose expertise is providing that capacity reliably, securely and at scale.

That allows companies like ours to concentrate on what we actually do—building software and serving customers—rather than building power systems, cooling plants and server facilities.

Multiply that concept across thousands of businesses and you begin to see what data centers actually enable.

Infrastructure creates things we cannot predict.

There is another part of this discussion that is harder to quantify.

Infrastructure often becomes valuable in ways nobody anticipated when it was built.

The electrical grid didn't exist because somebody predicted every appliance that would eventually plug into it. The interstate highway system wasn't built with Amazon deliveries in mind. The early Internet wasn't constructed because somebody foresaw smartphones, YouTube, cloud software or generative AI. Infrastructure creates a foundation. People build things on top of it.

Some of those things become businesses. Those businesses employ people. Employees develop skills. New industries emerge. Existing industries become more productive.

Technology advances, companies advance with it, and people gain opportunities that didn't exist before. That may sound idealistic, but we've watched that cycle happen repeatedly.

That doesn't make every concern invalid

None of this means communities should simply approve every project presented to them. Large data centers consume substantial resources. Depending on their design and location, questions about electricity, water, transmission infrastructure, noise, land use, tax agreements and environmental effects can be completely legitimate.

Communities should ask those questions.We should understand what is being built. We should demand credible numbers. We should scrutinize agreements made with public entities. And we should expect developers and elected officials to explain both the costs and benefits honestly.

Supporting technological development does not require abandoning scrutiny. But scrutiny should also include understanding what the infrastructure is actually for.

Maybe we're asking the wrong question

When people ask, "How does this data center benefit San Angelo?", perhaps there are really two questions hiding inside it.

One is local:

What does San Angelo receive in exchange for hosting this infrastructure and absorbing its local impacts? That deserves a specific answer from each developer and from our local government. But the second question is larger:

Why does society need this infrastructure at all?

The answer is that increasingly, modern society runs on computing.Businesses depend on it. Hospitals depend on it. Governments depend on it. Schools depend on it. Researchers depend on it. Artificial intelligence depends on it. And thousands of companies most of us will never associate with a data center exist because they can purchase computing as infrastructure rather than build it themselves.

That distinction matters.

The real output of a data center isn't simply whatever happens inside the walls of the building.

Its real output is what people are able to build because the infrastructure exists.

And I think that is an important part of the conversation we're not having enough.