The Mountain Eagle went online recently.

That's the simple version.

Subscribers can log in at MountainEagle.net and find the newspaper. Someone who moved to Florida can still read the same paper they used to pick up locally. People can listen to regional radio, submit events, find local organizations, buy an advertisement or list something for sale.

If that's all anyone ever notices, I'm fine with that.

It took me a long time to understand why.

I started thinking seriously about local infrastructure years before I had any involvement with the Mountain Eagle. Some of that thinking followed Hurricane Irene, which hit this region hard in 2011.

Flood damage is relatively easy to see. Houses are gone. Roads wash out. Businesses close.

Other losses happen slowly afterward.

Some people rebuild. Some don't. A family that was already considering leaving gets one more reason. A business loses enough customers that the numbers stop working. Fewer children enter the school. Someone retires and the thing they knew how to do isn't picked up by anyone else.

Eventually a place can lose more than population.

It can lose relationships.

That distinction became more important to me over time because communities aren't collections of independent pieces.

A newspaper depends on readers, advertisers, reporters, businesses, governments and community organizations. A local business depends on customers, employees, banks, suppliers and other businesses. Volunteer organizations depend on people having enough connection to one another that someone still shows up when something needs doing.

Most of this doesn't require anybody to design it.

People live together long enough and relationships accumulate.

The hardware store knows who can fix something. Someone at the diner knows whose kid just came home from college. The newspaper remembers the argument the town had fifteen years ago when everyone involved in the current argument was younger.

That works remarkably well until enough pieces disappear.

Then something strange happens. The institutions may still exist, but the connections among them become thinner.

I originally thought technology might help.

My first attempt didn't work.

That was useful.

I had approached the problem too much like a software developer. There were features people should have, information that could be organized, services that could be connected and technical problems that could be solved.

Most of those observations were individually reasonable.

Together, they produced the wrong thing.

The mistake was treating the community as something the software needed to create.

The community was already there.

That changed the problem.

The Mountain Eagle has been publishing a newspaper for decades. Radio stations already broadcast here. Businesses already serve customers. Churches, fire departments, libraries, schools and local governments already maintain relationships with people. Families already exchange information in ways no technology platform will ever completely see.

We didn't need to manufacture a digital community.

We needed the technology to recognize the one that already existed.

That sounds like a small distinction.

It changed almost everything.

The newspaper became the obvious place to begin because newspapers already do this kind of work.

A local newspaper doesn't merely produce articles. Over enough time, it accumulates relationships with a place.

A reporter knows which meeting matters this week. A business buys an advertisement because it wants local customers to know something. A family submits an obituary. Someone sends in a photograph. A town clerk provides a notice. A reader calls because something printed last week was wrong.

The newspaper sits in the middle of all of that without owning most of it.

That's valuable.

So instead of building another website that happens to publish a newspaper, we started building outward from what the newspaper already was.

Subscribers needed access to the paper, so we built subscriptions.

Businesses needed advertising, so we connected advertising.

Organizations needed somewhere to publish events and information, so we built directories and calendars.

People listen to local radio, so the radio belongs there too.

Members can submit things. People can list items for sale. Journalists can publish. Print readers can follow a QR code into something that doesn't fit on paper.

None of these features is particularly revolutionary.

The relationship among them is more interesting.

A business isn't merely an advertiser in one database and an event organizer in another. A church isn't merely a name appearing in an article. A journalist isn't merely text beneath a headline.

They're participants in the same region.

The website can begin to reflect that.

This is also why we've kept the underlying infrastructure local.

The servers are ours. The software is ours. The newspaper archive isn't being entrusted entirely to a social network that might change its rules next year. Outside services can still be used where they're useful. Payment processors are useful. Search engines are useful. Social media is useful. AI is useful.

They don't need to become the Mountain Eagle.

That boundary matters more to me now than it did when I started.

A useful outside service should extend what a local institution can do without quietly becoming the institution itself.

That gives us some interesting possibilities later.

The same relationships that connect readers, businesses, organizations and publishers could eventually support things we haven't built yet. Some ideas may work. Others probably won't.

I'm deliberately less certain about that than I once would have been.

We have something better than a theory now.

We have a newspaper.

People are using it.

That means reality gets to participate in the design.

If people use the directories but ignore some clever feature I spent months imagining, the directories win.

If subscribers mainly want to log in, find this week's PDF and leave, that's useful information too.

Especially that one.

It would be very easy for someone building technology to regard the PDF as an old form that eventually needs to disappear.

I think that would misunderstand what we're doing.

The PDF works because the newspaper already works.

A reader knows what it is. The publisher knows how to produce it. Advertisers know how to buy space in it. Reporters know how to fill it. A person who moved away twenty years ago can open it and immediately understand what they're looking at.

The digital system doesn't need to defeat that.

It can preserve it while allowing other things to grow around it.

That has gradually become the larger idea.

A community doesn't become stronger by replacing everything old with something new. It also doesn't survive by refusing to change.

Sometimes the useful work is figuring out which relationships still carry information and giving them another form in which to continue.

If this approach works, perhaps it can travel.

Other newspapers could share infrastructure. Local publishers could cooperate without becoming branches of the same company. Some technology can be common while the institutions using it remain locally owned and locally accountable.

I don't know yet what scale makes sense.

That's one of the things we'll find out.

For now, there is a much simpler test.

A grandmother can tell her grandson in Florida that he can go to MountainEagle.net, read the hometown newspaper and listen to WIOX.

He doesn't need to understand the servers.

He doesn't need to know what database is underneath the directory or how the radio player works.

He certainly doesn't need my twenty-year explanation of why any of this seemed worth building.

He just needs the newspaper to be there.

Maybe that's the better measure of the infrastructure anyway.

When it's working, people get to think about something else.