I think the most disturbing thing about the National Science Foundation’s announcement this week that they will tear down the famous radio observatory at Arecibo is that it didn’t come as a surprise.

Of course, it’s a real temptation to read more into this than you should. Say, the decline of Western Civilization. Or at least of American Exceptionalism. But that’s a trap. No point in going there.

For one thing, radio astronomy is alive and well, thank you very much. The picture at the top of this blog is a testament. The NSF still funds the NRAO and several observatories — although it no longer operates the famous Green Bank telescope in West Virginia:

Funding cuts.

Still, images of the damaged Arecibo telescope in Puerto Rico are hauntingly dystopian:

It will not be repaired. Destined instead to be scrap — demolished, torn down, carted away.

Remember all the talk about infrastructure projects? How badly so much of the country’s roads and bridges are in need of repair or replacement?

Minneapolis, 2007:

Care to guess how long it took to replace this bridge?

A year.

One year. We know how to do these things.

But it had to collapse first.

The last picture made me think of this:

The 737 MAX.

Also famous for looking like this:

Because of software problems.

But enough disaster porn. It’s too easy to find examples, and it’s just one way of thinking about the particular idea I have in mind. Disaster is not the point.

I’m sure you’ve heard the ancient maxim, “talk is cheap”. Well, so are dreams, and plans, and intentions. And abstractions. Like action compared to talk, physical reality is expensive next to abstraction. The abstraction is the beginning, but the hard work comes next: To make it happen. To finish it. To make it right.

To bring something to life in the real physical world is a struggle, a battle that has to be fought and won. Like building Arecibo. Or a highway bridge. Or an airplane.

So the question in my title is really this: Are we are growing too comfortable with abstraction? Too willing to leave the responsibility of physical things to “someone else”?

In my own day to day work, I deal with any number of companies developing new software products for 5G mobile networks — but very few with any appetite for making actual radios or servers.

Is software even a product? The industry is organized around “Continuous Integration” and “Continuous Delivery”, an endless stream of bug fixes and feature updates that continues until the software itself is finally scrapped. Because modern software by its very nature is never finished. It’s an abstraction that never stops evolving, correcting, changing.

But it has to run on something. It has to cause physical things to happen in order for a mobile network — or anything else — to work. Actual signals have to be generated and transmitted and received and decoded. And so someone has to manufacture the radios and the servers, which means design them and build them. And the designs have to be finished and the building has to be right.

Yet, in an almost cosmic irony, the usual profit margin for succeeding in something like this is razor thin — while the potential liabilities for failure are financially devastating.

Hard to blame entrepreneurs for preferring the abstractions.

So, yeah, it’s a fair question. As an industry, as a civilization, are we losing our edge? And is it inevitable?