Bowie History Atlas

Memories of my father

Some of this I have wanted to set down somewhere for a long time. Whether it belongs on a public page is a question I have not entirely resolved.

A certified genius and a terrible father

A baby grand piano — the kind that occupied one corner of a small Levitt living room, unreachable, unplayed, and eventually sold. Photo: Durand Piano Service · Wikimedia Commons (CC BY-SA 3.0).

My father was a certified Mensa member. He was, by every measurable standard and by the only measure that counted in daily life, a genuine genius. He was also, by the only measure that matters when you’re a child, a terrible father. Those two facts coexisted in him without apparent contradiction, which may itself be a feature of whatever he was, whatever we would call it now. High-functioning autism wasn’t a diagnosis readily available to adults of his generation. Looking back with the vocabulary that exists today, the profile fits: the obsessive interests that cycled through his life like weather systems, the difficulty sustaining ordinary domestic relationships, the genius for technical things and the near-total incapacity for the emotional ones.

I am writing this here because this is the place where I record what actually happened in and around Bowie over the course of a long life, and he was part of that life, and the record should include him. I want to be clear about the limits of this account. I am writing from memory and from documents, not from therapy. I am not interested in pathologizing him or in rendering a verdict. What I want is to get some of this down while I still can, accurately and without embellishment, and let the facts speak for whatever they speak for.

The piano

There was, when I was growing up, a baby grand piano in our small Levitt living room. I have no idea where it came from — probably inherited from someone. I suppose he wanted to learn to play. He never did, as far as I know, and neither my brother nor I was ever pushed toward lessons. The piano was eventually sold. I think about it now when I watch strangers playing public pianos on YouTube — people who simply sat down one day and learned — and I feel something that is not quite regret but is adjacent to it. The piano is the man in miniature: something grand brought into a modest space, full of unrealized possibility, and eventually discarded.

Books

He was a prolific reader — a trait I apparently inherited. Before the advent of the internet, I was constitutionally unable to walk past a bookstore without going in. Back in the mid-1970s, when I was still a small child and he was still around, the bookstore in the area was Crown Books at Hilltop Plaza — a fixture of the Washington-area retail landscape. I believe he stopped at every bookstore he passed, and in them he would strike up conversations with strangers — which was either a charming eccentricity or a social strategy for a man who found formal relationships difficult. Probably both.

The record collection

He had an extensive record collection, at least as it appeared to a child’s eyes. The stereotypical 1970s albums were all there — I remember a particular affinity for Cher — and somewhere in that collection was a copy of Led Zeppelin’s Houses of the Holy — the 1973 album whose cover famously featured two naked sibling children, a boy and a girl, their hair tinted an otherworldly purple-gold, climbing the basalt columns of Giant’s Causeway in Northern Ireland. For a curious child encountering it on a shelf, the cover was — I remember this clearly — fascinating in the way things you don’t fully understand compel, before you have the vocabulary to articulate why.

A man of hobbies

He moved through hobbies the way a hummingbird moves through a garden — with great intensity, for a very short time, before the next flower called to him. He took up sewing: enrolled in classes, bought a machine and all the accessories, and as far as I can remember never produced a single finished garment. He took up photography: acquired a 35mm film camera with a full complement of accessories, then apparently lost interest before the novelty had worn off. He took up cycling — had a road bike with the gear levers mounted on the down tube, flanking the stem, the configuration standard before integrated brake-shift levers changed everything — and he did actually take it on long rides. He just didn’t keep at it.

The things that actually held him

Two things, however, genuinely held him.

The first was electronics. There was a Heathkit store in Laurel, off Route 1, and he was a regular — Heathkit being the company that made its name selling the satisfaction of assembly alongside the finished product, kits for radios, test equipment, and practically anything with a circuit board. Most ambitious was the Heathkit console color television — 25 inches, with an electronic tuner, which was unusual for the time. When he left, he left it behind. We kept it for what felt like forever — not as a television, eventually, but as a table, the way large, heavy things you don’t quite know what to do with simply become furniture. That television, which he built with his own hands and abandoned, is perhaps the most accurate physical symbol of the relationship.

The second was radio. He jumped into CB radio when it was a national craze in the mid-1970s — had a unit installed in the car and another in his bedroom. The CB antenna stood tall alongside the house for years after he was gone; we never got around to taking it down. But CB was always going to be too democratic, too temporary, for a mind like his. What actually held him for the rest of his life was amateur radio — Ham radio, properly licensed, technically demanding, with a culture and a community unlike anything else in the hobbyist world. He had a Ham antenna in the backyard; wires strung between poles — a configuration that generated enough interference with neighbors’ radios and televisions that they complained to the FCC, and he was required to take it down. He knew and practiced Morse code. He was a licensed Ham operator for most of his adult life. When he died, his radios were still set up in his Virginia condominium, the antenna wrapped around the patio door — the workaround of a man who was not going to let a lease agreement stand between him and the ionosphere.

I donated his radios to a Ham Radio association in Annapolis. They sent me an IRS Form 8283 appraising them at over $5,000. I had also read that Ham Radio enthusiasts observe a tradition of issuing a last call for deceased license holders — a final sign-off on their callsign. I contacted someone about arranging that, but never heard back.

He was also a programmer, a passion developed when he was in the Air Force and held until his death. I remember he would communicate with mainframes through a printing terminal connected by an acoustic coupler to a telephone handset — the arrangement where you dialed the number, then physically cradled the handset into a pair of rubber cups on the modem. That was back when “baud” was technically accurate as a unit of measurement. He made serious money throughout his career writing and maintaining niche accounting software — a genuine specialist, the kind of person organizations flew in from around the world because nobody else understood the system.

Mensa, and what I made of those people

He occasionally took us to Mensa family events. I came away from those events with no desire to interact with those people again. I want to be honest about the limits of that perception — I was a child, adult social dynamics are largely invisible to children, and I have no idea whether my impression was accurate.

The spending, and the end of the money

He was also a spendaholic, and the money never lasted. At the end of his life he had filed for bankruptcy, taken a reverse mortgage on his Virginia condominium to service his debts, and was living on Social Security and whatever he could earn reselling thrift store finds. The condominium, when I eventually saw it, looked like a hoarder’s space — a condition that is itself associated with the neurodevelopmental profile I’m fairly certain he had, and one that is no less sad for having an explanation. Whatever he had once owned of value — besides his beloved Ham radios — was gone by then.

The estrangement, and the refusal

He left when I was in sixth grade, which I have written about elsewhere in this account. The estrangement that followed was gradual and then, near the end, complete. When my mother died, I asked him to attend her funeral. I had people ready to drive him — the logistics were arranged, the transportation was offered. He refused. That refusal was the last clear act in the relationship, and it said everything that needed to be said.

He had throat cancer by then — the kind now frequently associated with HPV, though I can’t say that with certainty in his case. He didn’t last much longer. He died of a heart attack in a hospital, at age seventy.

The near-erasure

I did not learn of his death for nearly a year.

The hospital where he died had contact information for my brother and me. I know this because that same hospital had contacted us when he first presented with cancer — we were in their records as next of kin. After he died, they made no such call. He had been living with a roommate — a companion whose precise relationship to my father I don’t know and didn’t press — and that person told the hospital he would handle arrangements. He presented himself to the funeral home as the responsible party, checked the box on the form attesting that next of kin had been notified, and had my father cremated. He had not notified next of kin. He had simply checked the box.

When the reverse mortgage on the Virginia condominium came due following his death, the servicer worked through its contact list. His former girlfriend had been listed as an alternate contact on the loan. They reached her — and that was the first she had heard of his death. She had my cell phone number, evidently given to her by my father at some point. She called me.

I spoke to the hospital social worker, who declined any responsibility. I filed the paperwork and paid the fee to obtain his full medical records — I intended to search them to see how next of kin had been designated — but I never got around to it. I spoke to the funeral home. The director who had handled the arrangements had since died. His successor explained, matter-of-factly, that they were not required to verify that next of kin had actually been notified — they followed standard nationwide protocols, she said, and the form simply required someone to check a box. Whether that is actually standard nationwide protocol I cannot say.

I attempted to consult lawyers about a possible action against the hospital and funeral home for failure to notify. The bar association was indifferent — they are supposed to connect people with attorneys and appeared not to care. The one lawyer I managed to reach told me that without assets to recover there was no case anyone would take on contingency, and that, in any case, they only dealt with Virginia residents. That was the end of it.

The roommate had also been driving my father’s old Toyota. I told him I would contact the sheriff to report that he was operating a vehicle registered to a deceased person without valid insurance, and that as the named executor under my father’s will, I was prepared to enforce that. He surrendered the car without argument. I kept it for several years, let my brother use it as a spare, and eventually donated it to the Red Cross after it developed a fuel line leak. In addition, I took his Ham radios and the filing cabinet containing all of his records — I wanted those records off the street, and under my control. I have no idea what became of the condominium afterward.

Arlington

One thing the roommate had done — and for this, genuinely, I was appreciative — was initiate the process of having my father interred at Arlington National Cemetery. He was a veteran, and that entitlement is a real one; Arlington is an honor, and the process takes months. By the time I learned of the death, the scheduling was nearly complete. Had I learned any later, the interment would have taken place without us. As it was, I had just enough time to contact his brothers and sister, both still living in the DC area, and we attended together.

Arlington gives you the full ceremony. The flag, the rifles, the folding. They presented me with the flag.

That flag hangs in my living room now, in its case, alongside his last driver’s license — the only recent photograph I have of him. He is interred at Arlington, his ashes housed in one of the cemetery’s ossuaries, which is more than he earned by his conduct as a father, and also exactly what he was owed by his country for his service. Both of those things are true simultaneously. I have made my peace with holding them at the same time.

G.W. — a colleague, and something more

Nearly twenty years of federal IT work, the technology that changed everything, and the retirement lunch nobody thought to mention.

A mentor, though we never used the word

There was a man I worked with for nearly twenty years who functioned, in some important way I could never quite name, as a mentor. I don’t think he thought of it that way. We never discussed it. The word itself never came up in any conversation I can recall. But when I think about what I learned about navigating a federal IT career — about how to approach technical problems, how to figure out what an organization actually needed rather than what it said it needed, how to do the work without requiring acknowledgment for doing it — I keep arriving at the same source. I’ll refer to him here as G.W.

He was a math major. You could tell, or at least I could tell, in the way a person’s mind works when they’re bored and turn to equations in the margins of things rather than doodling. That kind of mind — one that reaches for structure automatically — turns out to be useful in precisely the way a federal IT shop needs people to be useful: methodically, without drama, on problems that don’t come with clean specifications.

Before the network — the world we came from

When I first came to the federal government, I was in the Office of Refugee Resettlement. ORR ran Wang minicomputers — the dominant office computing platform in federal agencies in those years, and the machine that defined what “word processing” meant before personal computers changed the vocabulary entirely. The workflow was the one that prevailed everywhere at the time: professionals wrote documents by hand, and secretaries and clerk typists — including me, for a stretch — typed and printed them on the Wangs. I had a Wang terminal at my desk. ORR also had a handful of IBM XTs by that point, and one staff member had even brought in a bootleg Apple II clone, which occupied a desk with the quiet, slightly illicit status of a thing that wasn’t supposed to be there but clearly worked.

The Wang was administered by the executive secretary to the director of ORR — an arrangement that made organizational sense but had practical limits. She wasn’t really qualified for the role, and she knew it. What I discovered was that ORR had purchased a full suite of Wang software, including a spreadsheet application — packages that, at the prices software commanded in those years, had probably cost thousands of dollars. I wanted to install and use them. She was terrified I would break something, and she said no. The software sat unused. We used the Wangs exclusively for word processing, which was a fraction of what we had paid for. It was the kind of institutional waste that happens when the person controlling access to a system is more afraid of it than the people who want to use it.

There is a footnote worth adding here: WordPerfect, the PC word processor that would later become central to my daily work, was deliberately modeled on the Wang interface — the same function-key structure, the same underlying logic. For anyone who had trained on a Wang, sitting down at WordPerfect for the first time was less a learning curve than a recognition.

In 1986, ORR was consolidated with several other HHS programs into the newly created Family Support Administration. G.W. came from a different organization within that same consolidation — one that had been running not just Wangs but dedicated AB Dick word processors and IBM System/36 midrange computers with terminals, the multi-user centralized setups common in larger government offices of that era. That is how we ended up in the same office, both of us arriving at roughly the same moment the personal computer was arriving too.

I was seventeen or eighteen at the time, having started my federal career at fifteen through the Stay in School program — a detail that tended to surprise colleagues who assumed, reasonably enough, that I simply looked young. G.W. was considerably older, with a career already behind him that I knew only in outline.

What I had going for me, beyond youth’s impatience with slow things, was that I had come in with actual programming instruction behind me. My high school computing classes had included COBOL on terminals, then BASIC on TRS-80 Model IIIs, and eventually the Tandy 1000 IBM XT clones with their included DeskMate software. At home I had a Tandy MC-10 — an extremely cheap home computer in the tradition of the Timex Sinclair, aimed squarely at families who couldn’t afford a Commodore or an Apple — and later a used TI-99/4A bought off a classmate for next to nothing — he had upgraded to a Commodore 64 and had no further use for it. I still have both, wrapped in layers of plastic for posterity. I spent countless hours typing BASIC programs printed in magazines, and eventually began writing my own. I did my math homework by writing programs to solve the equations for me, which was either resourceful or shortsighted depending on how you look at it — in retrospect, probably the latter, since homework exists to make you practice the thing, not just arrive at the answer.

That background meant that sitting down at a Wang terminal, or at one of the early PCs, was not a confrontation with something foreign — it was a recognition. I learned those systems fast, genuinely fast, in the way that only happens when the underlying concepts are already there waiting.

The office recognized this by putting me in charge of training new staff on the Wangs. This did not make me universally popular. There is a particular kind of resentment that flows from older, more senior staff toward a teenager who has just been assigned to teach them how to use their own equipment — a resentment I understood, even then, as not really being about me. It was about the technology arriving faster than professional dignity could absorb it. I taught the classes anyway, and tried to be useful about it.

The arrival of the IBM XT, and what G.W. built

The first IBM XTs arrived as curiosities — a scatter of machines, not yet a system. When management decided the office should build its own PCs rather than procure them — a decision that seems extraordinary in retrospect but made a certain fiscal sense at the time — G.W. led that effort. I assisted, but the distinction matters: he designed the approach, established the configurations, and did the heavy lifting. I was young and learning.

What G.W. built next was the foundation everything else ran on. He chose Banyan VINES as the network operating system — a sophisticated choice, and one that larger organizations and US government agencies adopted precisely because of its advanced directory services at a time when competitors hadn’t grasped why that mattered. VINES ran on Unix on Banyan-branded servers; later, as the technology evolved, we migrated to VINES on Windows NT on generic hardware. G.W. designed the original network architecture, established the file structure, configured the servers, and built the connectivity between our Washington headquarters and the regional offices. The network that everybody used every day — that they took for granted the way people take electricity for granted — was something G.W. had conceived and built.

My role was to support what he had built: installing and maintaining the PCs, supporting the applications — WordPerfect, Lotus 1-2-3, Lotus Symphony, Ashton-Tate dBASE — and later helping with the network itself as the infrastructure grew and needed more hands. We moved through the successive eras together: Banyan VINES Mail, then Beyond Mail for Banyan, then the migration to Windows 3.1 and the Microsoft suite, then Windows 95, and on from there in the familiar cadence of upgrades that defined federal IT through the 1990s.

The cabling, the regional offices, and work nobody asked us to do

For the physical infrastructure, professional crews handled the initial installations. But our office was housed in modular furniture — the kind that gets rearranged with reasonable frequency to accommodate reorganizations, new hires, and the general organizational restlessness of federal agencies — and every rearrangement meant the network cabling had to move too. IBM Token Ring Type 1 and Type 3 cabling didn’t move itself. G.W. and I would come in on weekends and do it. We did this dozens of times over the years. We never received overtime or any additional compensation. We never asked for any. It simply needed doing, and we did it.

We also flew out together to ACF’s regional offices — Atlanta, Boston, Chicago, Dallas, Denver, Kansas City, New York, Philadelphia, San Francisco, Seattle — to build out the network infrastructure at each location. The pattern was the same each trip: arrive, install network cards and cabling over the weekend and evenings, then spend the working days teaching the regional staff how to use what had just been installed. Chat functions, email, file sharing — things that feel obvious now but that required explanation and practice when they were new. I taught the majority of those classes, which was work I genuinely enjoyed. I had studied the application manuals — software came with substantial printed documentation in those days, binders thick enough to use as doorstops — and knew the applications thoroughly enough to answer questions I hadn’t anticipated.

What I built — and the machines I considered mine

Beyond the teaching, I had built a few things of my own. I developed automation in the WordPerfect macro language for filling out acquisition forms — thick multi-part carbon forms that could only be printed on a specific brand of heavy-duty impact printer — eliminating a genuinely tedious manual process. I wrote a small PC-tracking application that ran at network login, periodically quizzing users about their PC assignments and recording hardware specifications; this predated the commercial software that eventually replaced it. I wrote a utility that allowed DOS users to select network printers from a centrally managed list at login, before Windows made that capability native. These were small tools in the context of an enterprise network. They worked, and people used them.

I was also, I’ll admit, a collector. Not officially — everything was government property and I understood that — but I had an interest in computing history that the people around me did not share, and I had quietly accumulated what I thought of as a personal museum: machines that represented significant moments in the technology’s development, kept back from the surplus process because I believed they deserved to be preserved. The Apple Lisa was the crown of it — a machine so fragile in institutional memory that it was treated as an object not to be touched, brought out occasionally and handled with the reverence usually reserved for things behind glass. I touched it a few times. It was eventually surplused, long before the rest of the collection, and I didn’t fight it at the time. The Compaq luggable — the suitcase-sized portable that looked like it had been designed for someone who found airline carry-ons insufficiently burdensome — was part of it too, along with other machines whose names I no longer remember clearly.

When I was temporarily assigned to another building for a year, my colleagues surplused all of it in my absence. They had no way of knowing what those machines represented to me, and I had no official claim on any of them. I was resentful for longer than was probably reasonable. The Lisa was already long gone. But I had thought of those machines as mine — in the way that people think of things as theirs when they are the only ones who understand their value — and losing them to a surplus auction felt like losing something that should have been a museum.

Chicago, and the creation of ACF

G.W. and I were both in Chicago when the announcement came. On April 15, 1991, the Administration for Children and Families was formally created — FSA and the Office of Human Development Services merged into a single new agency. The reorganization we had been working within now had a new name and a new shape. We continued on together in it for years, the network expanding as the agency grew.

The shared services reorganization, and the retirement lunch

Eventually HHS determined that the IT shops of the smaller operating divisions — the “small OpDivs,” of which ACF was actually the largest — should be consolidated into a shared services model. Staff were reassigned to the new office. G.W. went with it. I was offered a position as well, but the role on the table was in COOP — Continuity of Operations Planning — which was neither my area of experience nor my interest, and I said so. I stayed in ACF. I was the only member of ACF’s IT staff to do so.

Some time after the move, I was in conversation with a couple of contractors working for the new shared services office. They made an offhand comment about G.W. — dismissing him as though he didn’t know what he was doing, the implication being that his lack of formal IT certifications was disqualifying. They had no idea I had worked with him for nearly twenty years. They had no idea what he had built.

G.W. retired not long after. I heard later that there had been a retirement lunch. I was not invited. I was not told. I learned about it after the fact, from someone who mentioned it in passing and assumed I already knew. I had worked alongside that man longer than anyone at that lunch had, had been in the buildings and in the regional offices and in the weeds of the infrastructure with him through every phase of what he’d built — and nobody thought to tell me there was a lunch.

I was hurt. I was upset about it for a long time. I still think about it.

What the relationship was

I have no idea what G.W. made of any of this — of me, of the years of working together, of what if anything our collaboration meant to him. He was not a person who disclosed that kind of thing, and I was young enough for most of those years that the asymmetry of the relationship probably made it awkward to discuss in any case. Maybe it was simply a professional relationship to him: two people who happened to work on the same infrastructure for two decades and happened to be in Chicago the day the agency they worked for got a new name.

But I came up in IT in the way people came up in it before there were certification programs and career ladders and formal mentorship tracks — by watching someone who knew more than I did, by doing the work that needed doing, by being in the room when the decisions were made, and by accumulating, over years, an understanding of what good judgment looked like. That understanding came from somewhere. Most of it came from G.W.

The contractors didn’t know that. Nobody at the lunch knew that. The surplus crew certainly didn’t know that when they cleared out the room. That’s the nature of knowledge that was never written down and never formally acknowledged — it simply exists, or it doesn’t, in the minds of the people who were there.

Unpaid Overtime — The Federal Employees You Never Heard About

A personal reflection on my decades at the Administration for Children and Families and its predecessor, the Family Support Administration.

In the essay that appears just before this one I wrote about G.W. and the years we spent building and maintaining the ACF network. I mentioned, almost in passing, that we came in on weekends when the modular furniture was rearranged. That was true, but it was only one small piece of a much larger pattern. For most of my federal career I worked a great deal of unpaid overtime. Not occasionally. Routinely. And I was far from the only one.

When we traveled to our regional offices to install or upgrade the network infrastructure we flew out on Saturday, spent Sunday installing PC network cards, loading software, configuring and installing the servers, setting up accounts, and making everything work. In a couple of cases the professional cabling crews had terminated connections incorrectly and we had to re-terminate them, but cabling itself was rarely the issue. In Philadelphia on one trip we stayed at the site until well past midnight because the work simply wasn’t finished. The hotel was next door, so at least we could stumble back and get a few hours of sleep before starting again. I asked management once whether all this extra time would ever be compensated. The answer was polite but unmistakable: travel itself was considered compensation enough. If I wanted to keep traveling, I should stop asking the question. I was young, eager, and very dedicated. I never asked again.

That same work ethic stayed with me for the rest of my career. I never saw the job as something you clocked out of at 5:00 p.m. if the work still needed doing. On September 11, 2001, I was leading a scheduled PC upgrade project in one of our operating divisions. While nearly everyone else clustered around the few televisions in the building watching the horror unfold in New York, I kept working. I remember telling someone that a B-25 bomber had struck the Empire State Building in 1945 and the building had survived — a comparison I already sensed was inappropriate even as I said it. I didn’t yet grasp how different the scale and speed of a passenger jet would be. When the third plane—American Airlines Flight 77—hit the Pentagon and smoke became visible from our building off L’Enfant Promenade, the office emptied. Metro stations turned into madhouses. People who lived in Virginia began walking across the bridges. I stayed. I figured the building was probably the safest place I could be, and I was able to finish far more of the upgrades than I would have if everyone had remained. The same thing happened after the Virginia earthquake on August 23, 2011. Most people went home. I stayed behind, noticed that the breaker on the roof-mounted drycooler serving our computer-room air conditioners had tripped, went up to the roof, and reset it. Had the servers lost cooling they would have thermal-shutdown and at least some of the RAID arrays would have failed. I was glad I had been there.

There were dozens of smaller examples—late nights, weekends, holidays—when I simply stayed until the job was done. In most cases I was not alone; plenty of other people in the office went above and beyond as a matter of course. But in the crisis moments I was often one of a very small handful still at our desks.

The largest single project of my later career was the multi-phase move when our lease on the building we had occupied since 1988 expired and we returned to the Mary E. Switzer Building where I had started in 1985. I built a Microsoft Access database to track every seat assignment, every move date, every phase. Program offices marked up printed floor plans by hand; I entered the data, let the system flag conflicts, then generated the receipt sheets that employees had to sign when their boxes and hardware arrived at the new location. I coordinated with the VoIP team to make sure phone numbers were ported properly and phones were live on the right day using the move dates in the database. I did many dozens of other tasks that were never formally assigned but simply needed to be done. On some days I worked fourteen hours or more. Most of those hours were uncounted and unpaid. We did receive a few time-off awards for the move as a team, and I received an individual award as well—grateful for both—but the bulk of the extra time was simply absorbed.

None of this is a complaint. The overtime helped me earn promotions. I received a number of awards over the years, including one presented to me personally by Secretary Sullivan on stage during the annual awards ceremony. The experiences were genuinely meaningful to me. I loved the work. I loved solving problems, building systems that people relied on every day, and knowing that the small tools and procedures I created actually made someone’s job easier. I was proud to be a federal employee.

What did bother me—deeply, at times—was the casual contempt I heard from pundits and commentators who painted all federal workers as slackers. I knew the slackers existed; every office had them and we all knew who they were. But the vast majority of people I worked with were conscientious, often quietly heroic in the small ways that keep large organizations functioning. The bad apples simply loomed larger in the narrative. When DOGE arrived and the public narrative hardened into sweeping claims that federal employees were largely slackers and part of a bloated bureaucracy, it felt like a personal erasure of the decades I and thousands like me had given. We weren’t the story they wanted to tell.

I had already been seriously thinking about retiring at the end of 2025. When the Fork-in-the-Road buyout offer appeared, it was an easy decision. I took it. I would have preferred to leave in better times, under better circumstances, with a different tone in the national conversation. But the past is the past. I spent decades doing work I believed in, often well beyond what anyone asked or paid for, and I have never regretted the dedication itself—only the occasional sense that it went unnoticed.

That is the part of federal service the public rarely hears about. Not every government employee was a slacker. Some of us loved the job, gave it everything we had, and would do it again in a heartbeat. We just never made the evening news.

Story songs — the golden decade, a Staunton afternoon, and Ren

On the oldest form of popular music, the decade that did it best, and why an independently-made album from a sick man in a small Welsh village is carrying something forward that most people don’t know was lost.

Something heard before it was understood

My father’s record collection, which I’ve described elsewhere in these pages, accumulated without evident curation — the popular artists of the era alongside classical, the whole thing organized by arrival rather than intention. Whether any of it shaped my taste is a question I can’t answer with confidence. Children absorb what’s in the air of a house without being able to name it later. What I can say is that somewhere in those years of growing up in Bowie, I developed a strong preference for songs that told stories over songs that simply expressed emotions. A song that described the arc of a life, the crash of a relationship, the fate of a character I’d never meet — that interested me in a way that a song about how somebody felt, without context or consequence, did not. I may have encountered that preference at home, through records I didn’t choose. I may have encountered it on Sunday evenings at Allen Pond Park.

The Sunday evening concerts at the Setera Amphitheater were a ritual of those years — free, attended by families spreading blankets on the grass. The offerings ran to bluegrass and folk as often as anything else, which was appropriate for a community at the geographic margin between the DC suburban sprawl and the Appalachian cultural territory that begins somewhere around the mountains of western Maryland. Bluegrass is, fundamentally, a storytelling music. The old songs — the murder ballads, the disaster songs, the tales of wayward sons and abandoned women — existed to carry narrative across distances that literacy couldn’t cover. The story was the point; the music was how you made it stick.

That tradition ran directly into the popular music of the 1970s. And the 1970s, whatever else they were, were the golden decade for the story song as a mainstream American art form.

I know this partly from having lived through it — and partly from what I’ve watched happen on YouTube over the past several years. The reaction video is its own genre now: someone watching a piece of music for the first time on camera, recording their response, and posting it. What is striking about reaction videos for 70s story songs is how genuine those reactions consistently are. Younger viewers encountering “American Pie” or “Cat’s in the Cradle” or “The Wreck of the Edmund Fitzgerald” for the first time are not performing surprise — they are surprised. They are moved. The songs land on people who have no nostalgia for the era, no prior attachment to the artists, and no reason to pretend. That is the test of whether something was built to last, and these songs pass it.

The decade — and the question of the hook

Don McLean’s “American Pie” (1971) — nearly nine minutes long, a metaphorical history of rock and roll from the 1959 plane crash that killed Buddy Holly, Ritchie Valens, and The Big Bopper through the disillusionment of the late 1960s — established that popular audiences would sit with a story that took its time and meant more than one thing. Radio stations were initially reluctant to play it because of its length, where the norm was closer to three minutes. It became a number-one hit anyway. The industry’s categories had encountered something that didn’t fit them, and audiences found it regardless.

Harry Chapin’s “Taxi” (1972) — a seven-minute narrative about a cab driver who picks up a former lover and the conversation that doesn’t happen — pushed further in the same direction. “Cat’s in the Cradle” (1974) compressed a man’s entire failure as a father into a structure that reversed its own meaning as it went. “Operator” (1972) was a phone call to a long-distance operator that became a meditation on loss and dignity. “The Wreck of the Edmund Fitzgerald” (1976) was a ballad in the oldest tradition — a formal lamentation for the dead, the kind of song people once wrote about specific disasters to give them shape and meaning, now played on FM radio.

These songs had varying relationships with the commercial hook — that repeatable, radio-friendly phrase designed to make a song stick in the shortest possible time. “American Pie” has a chorus that radio eventually accepted, though the song’s entire value was in its verses, which told the story and which no radio format was built to accommodate. “Taxi” had no conventional hook at all. Neither did “Edmund Fitzgerald.” I have never cared particularly whether a song had a hook. The hook is essentially the price of admission to the commercial format — the concession an artist made to get played. The story songs I’ve loved most tended to treat that concession as optional, or skip it entirely.

Dan Fogelberg belongs in this company, though the commercial label on him — “soft rock,” “adult contemporary” — obscures what he was actually doing. “Same Old Lang Syne” (1980) is a story song by any definition: a chance meeting in a grocery store on Christmas Eve, an ex-lover, a shared beer in a parking lot, two lives narrated in four minutes without a false note. “Leader of the Band” (1981), which he wrote for his own father, a band director at a high school in Peoria, Illinois, is an elegy that manages to be grateful and bereft simultaneously. Fogelberg came to his storytelling through folk and, eventually, through bluegrass — High Country Snows (1985), recorded in Nashville with Doc Watson, Ricky Skaggs, and Jerry Douglas, became one of the best-selling bluegrass albums ever made. It grew out of years spent listening to bluegrass tapes on long Colorado drives and an unexpected sit-in with Chris Hillman at the Telluride Bluegrass Festival. The thread from Appalachian tradition to folk-rock to adult contemporary to a full bluegrass album is a coherent artistic biography.

I would include the Talking Heads in this conversation, though the argument requires a moment. “Psycho Killer” (1977) is a character study that works by implication and dread. “Once in a Lifetime” (1980) is a running commentary on how a man arrives at a place he didn’t intend to be. “Life During Wartime” is a fugitive’s internal monologue. Byrne was approaching the same territory from a more abstract and literary angle — the story told from inside a consciousness rather than above it. That is still a story.

Staunton, 2025, and Allen Pond, 1975

Last summer I was in Staunton, Virginia — partly for its history and partly because the Shenandoah Valley is bluegrass country in a way that Prince George’s County is no longer. I learned by chance that a free bluegrass concert was happening in a public park, and I went. I happened to have a folding chair in the car; I pulled it out, found a good spot, and settled in. There were hundreds of people there across three performing bands, with food sold in a pavilion and a raffle running through the evening. It was the Setera Amphitheater fifty years later — different park, different state, same music, same families, same summer evening. I sat in that chair and felt something I had not felt in a long time: the recognition of something very old being practiced by people who understood what it was.

The demographics of the DC suburbs have changed since those Allen Pond evenings. The connections to Appalachian folk tradition have largely dissolved into the broader cultural fabric of a major metropolitan area. The Staunton concert reminded me that the tradition didn’t disappear — it survived in the places it had never stopped being the local music. I left with that feeling still active: the sense that something worth caring about had carried on somewhere, even as it was absorbed or lost in the places I grew up.

A sick man from a small village

The last piece of this essay is about an artist I encountered entirely by accident — or rather, by the algorithm that functions as accident in the age of YouTube. His name is Ren. He is Welsh, he lives in a small village, and he has Lyme disease, which he contracted several years ago and which has not been adequately treated and may never be. His music is made in his home, produced independently, and distributed almost entirely through YouTube and Bandcamp. He has never had a record deal. His largest audience has been assembled without radio, without music press coverage, without the machinery that built the careers of the artists I described above.

What he does, specifically, is tell stories. Long ones, carefully constructed, with characters and arcs and the kind of detail that requires the listener to pay attention and rewards the listener who does. “Hi Ren” (2022), the song that first circulated widely, is a twenty-minute conversation between Ren and a version of his own illness, staged as a debate and a struggle. “Sick Boi” (2022) is an autobiographical account of what happened to him — the initial infection, the misdiagnosis, the loss of his career as a session guitarist, the years of not being believed, the financial collapse, the attempt to rebuild. It is about twelve minutes long. It is one of the most complete and emotionally specific pieces of music I have encountered in my adult life.

His most ambitious project is the album RENegades (2024), which runs to something like two hours across its full tracklist and tells a single continuous narrative story. It is structured like a concept album in the tradition of The Wall or Tommy, except that it was made by one person, in a village, with an illness, without label support, and distributed to a global audience of people who found it the same way I found it: by following a recommendation and sitting with it long enough to understand what it was.

The reaction video community has found him too. Watching someone encounter “Hi Ren” for the first time produces essentially the same thing I described watching happen with the 1970s story songs: genuine surprise, genuine emotion, genuine engagement with something that required patience to understand and repaid it. The work is landing on people who have no prior attachment to him, no reason to be moved, and who are moved anyway.

The story song did not disappear. It went underground, or it went independent, or it went to a village in Wales where a man with a chronic illness was not going to wait for industry permission to make the work he had in him to make. The golden decade I grew up in produced the radio-broadcast version. What Ren is doing is the same impulse operating without the radio — and perhaps more honestly for it.

From a bulletin board to an AI-assisted atlas — on the blank page, and what happened to it

On running a government BBS in the 1990s, the long gap that followed, a neighbor’s offhand comment about French soldiers on Generals Highway, and what this technology actually is and isn’t.

The bulletin board

For a period in the 1990s, I ran a MajorBBS multi-line bulletin board system for my office at the Administration for Children and Families. The MajorBBS, developed by a Fort Lauderdale company called Galacticomm, was built on a technical breakthrough that let a single PC handle dozens of simultaneous dial-up connections — something that had required mainframes not long before. We used it to make official guidance, rules, and regulations available to the public. Someone with a modem and communications software could call in, navigate the menus, and download what they needed.

I was proud of it. I had been dialing into bulletin board systems since I was a kid — the BBS was the internet before the internet existed, a place where you could exchange messages, download files, and find your people across a telephone line — and being handed the opportunity to run one myself, for a federal agency, was genuinely exciting. It was an entirely solo effort. I am not sure the senior executives at the agency fully grasped what had been made possible: documents that had previously been printed, warehoused, and mailed on request — we had staff whose job it was to maintain physical stocks of publications and fulfill individual mailing requests — could now be obtained electronically, on demand, at any hour, by anyone with a modem. That is a quiet revolution by the standards of the time, and I understood it as one even then.

What made the MajorBBS historically interesting was that it became one of the platforms through which the BBS world and the emerging internet briefly touched. Third-party TCP/IP add-ons, and eventually Galacticomm’s own Internet Connection Option, gave MajorBBS systems the ability to provide dial-up internet access via SLIP and PPP. Some early internet service providers ran on exactly this kind of software. It was a short-lived bridge: the World Wide Web arrived and the BBS world collapsed with remarkable speed. The introduction of inexpensive dial-up internet service and the Mosaic browser offered something a BBS simply could not — global access, a universal interface, and a publishing model anyone could use.

Before the BBS was decommissioned, I wrote software to convert existing BBS content into simple HTML pages and publish it to the web. It was rudimentary work by any standard, but it was genuine web development. Eventually web publishing matured, it was handed to a dedicated web team, and that was the natural end of my involvement. I had some later exposure to tools like Macromedia Dreamweaver — the web development tool that passed to Adobe in the company’s 2005 acquisition of Macromedia — but I never published anything serious, never had a reason to learn Javascript or CSS, and my career had moved on to other areas of IT. The technical skills lay dormant, atrophying from disuse. For more than two decades, I had no particular occasion to go back.

The neighbor’s comment

The thing that changed it was an offhand remark from a neighbor. A friend of his had mentioned that the French Army marched down Generals Highway during the Yorktown campaign — that they’d passed through what is now Bowie. I didn’t know whether that was true. I started looking into it, and the looking-into-it turned into something I hadn’t expected: a sustained and genuinely pleasurable research project using AI as a collaborator. What I found was that AI could hold the shape of a complex historical question across a long conversation, pull in primary sources, check facts, and help me understand things I’d previously only skimmed the surface of.

That led to the military history visualizations — interactive timelines covering the evolution of ground vehicles, aircraft, weapons systems, naval vessels, and campaigns including Yorktown. Each visualization has a Surprise button that jumps randomly to a new entry; each entry links to its Wikipedia article or primary source. The YouTube algorithm, which has strong opinions about what I should be watching, kept surfacing new military history material, and I kept going back to add it. These felt like something worth sharing. That observation — these are good enough to put somewhere — opened a different rabbit hole entirely.

The rabbit hole

Sharing the visualizations meant building a website. Building a website with AI turned out to be genuinely feasible in a way it hadn’t been before — I could describe what I wanted, review what was produced, correct it, refine it, and arrive at something that looked and worked the way I intended, without needing to learn CSS or JavaScript from scratch. The Yorktown research had already pulled me toward Bowie history, and once the infrastructure existed, the history chapter followed naturally: Eva Cassidy, the Rochambeau wagon train route, Governor Oden Bowie, Allen Pond Park, the Levitt development, Belt Woods. Each topic supported by primary sources, newspaper archives, and reference links — the kind of documentation that I’ve always found more satisfying than the unsourced claim.

Everything on this site could theoretically have been researched using Google and Newspapers.com — and both of those tools were, in fact, part of the process. But it would have taken much, much longer. AI didn’t replace those sources; it collapsed the distance between question and answer, held the context across long research sessions, and helped me understand what I was looking at when I found it. The difference is something like the difference between having a knowledgeable colleague to think with and working through a card catalog alone.

And then, somewhat to my own surprise, the essays.

The blank page

I had been carrying some of these thoughts for years. The memories of my father, the G.W. career piece, the story songs essay — these were things I’d wanted to put somewhere and never had a practical mechanism for doing so. The obstacle was never the material. It was the blank page: that specific paralysis that sits between having something to say and having said it. What AI gave me was freedom from that paralysis, and the feeling that came with it — of being able to finally get things down, to shape them, to publish them somewhere they would last — is difficult to overstate. These are things I am genuinely glad exist now that didn’t before.

I want to be precise about what that means, because it is frequently mischaracterized. The essays on this site were not written by AI. They were wordsmithed in a ghostwriter arrangement — the ideas, subjects, and emotional content are entirely mine. I provided the topics, the specific memories, the judgments, the corrections. I told AI to verify and fact-check, and it provided details I’d simply forgotten but recognized as accurate when I saw them — the name of a piece of software, a date, a technical specification. That is no different from what a good editor and fact-checker would have done in any traditional publishing arrangement. Going back and forth until a passage reads right, catching grammatical errors and sentences that don’t flow, insisting on accuracy — that is work, and I did it for every piece on this site.

Keeping the mind alive

The same collaborative process extends well beyond this website, and has become one of the more unexpected pleasures of retirement. Since I first began using AI several years ago, I have had hundreds of conversations about current events and their historical antecedents — using the past as a lens for understanding the present, and the present as a reason to go back and learn the past more carefully. It is the kind of thinking I always wanted more time for during a decades-long federal career, and retirement has finally provided it.

The Kurdish situation is one example among many: sustained conversations about the Kurdish populations of Iran, Iraq, Syria, and Turkey — one of the largest stateless groups in the world — trying to understand how communities in each country think about independence, and whether there is any universality across those groups that could support a viable Kurdish state. Those same conversations extended into the historical divisions of multi-ethnic states across Eastern Europe. I’ve used AI to think through why certain modern conflicts echo earlier ones, why some peace settlements hold and others don’t, why geography shapes politics in ways that ideology rarely can. I’ve used it to understand the music industry well enough to articulate why an artist like Ren, whose work is discussed in the previous essay, isn’t more widely celebrated. These are not research projects with defined endpoints. They are ongoing conversations — the kind of intellectual engagement that keeps the mind from going quiet.

That freedom — to follow a question as far as it goes, to learn something genuinely new in the afternoon because something on the news that morning made me curious — is what retirement is supposed to feel like, and AI has been a significant part of making it feel that way. It fills the days, and fills them well.

Not a crutch

What AI provides to me is not the ideas. It provides relief from the technical burden — the blank page, the grinding effort of getting a first draft down, the fact-checking that would otherwise require hours of separate research. The press is full of commentary about AI making people intellectually lazy, atrophying the skills and habits of mind that serious thinking requires. That has not been my experience. AI has made me more curious, not less — more willing to pursue a question because I know I have a capable tool to pursue it with. It is not a crutch. It is a tool, and the distinction matters.

The prompt I used to start this essay — the one that contained the BBS, the gap, the neighbor’s comment, the visualizations, the Surprise button, the Bowie History rabbit hole, the Kurdish conversations, the Ren conversation, the ghostwriter model, the blank page, the word count instruction, and everything else — was 756 words long. The subsequent prompts refining, expanding, and correcting the text added another 315. Together, the instructions that produced this essay totaled 1,071 words. That is the work I put in. The essay is mine.

Ownership — on autonomy, investment, and what happens when you walk away

A career-long pattern of taking things on, and what that actually costs.

The autonomy that came with the work

Looking back at the projects I’ve written about here — and at many more I haven’t — I can see a pattern I didn’t fully recognize at the time. I was given a remarkable degree of autonomy. Not as a stated policy or explicit grant of authority, but in the practical sense that when I identified something that needed doing and took it on, people generally let me do it. That autonomy was both a gift and, in ways I only recognized later, a burden. Because when you’re given that kind of latitude, and you’re the kind of person who responds to it by investing fully, you tend to feel like you own the thing — and once you feel like you own something, what happens to it after you’re gone matters to you in a way it may not matter to anyone else.

The helpdesk years — and what a knowledge base is actually for

My early career was in what I’d describe as front-line IT, though the title never quite matched the work. I was, in practice, the lead technical person in the office during the DOS era — installing and repairing PCs, swapping out hard disks, ISA cards, and power supplies, and keeping the printers running. That last category was its own world: replacing LaserJet fusers, cleaning rubber feed rollers, clearing jams, vacuuming out spilled toner with a toner vac. It was dirty work — PCs, printers, and CRT monitors were all extremely heavy, and none of it was done at a clean desk. Keyboards were another matter entirely: crusty, stained things that required a thorough scrubbing before you touched them and a thorough washing of your hands afterward. And then there were the office hoarders — colleagues whose cubicles and offices were stuffed floor to ceiling with paper and accumulated debris of unknown vintage, which made any hardware call in their space a small adventure. I didn’t have the title of lead tech. I had the job.

For ticket management and knowledge base, we used a product called HELP!Desk, from Coastal Technologies — a DOS-based helpdesk application that did exactly what it needed to do. When I administered that system, I was deliberate about one thing above everything else: the tickets had to be useful to the next person who saw them. Resolution information had to be detailed enough that someone encountering the same problem three years later could read the ticket and actually solve it. A ticket that said “fixed” and nothing else was, to my mind, no ticket at all. The knowledge base was the point. The ticket was just the container.

I wrote elsewhere in these pages about the software I developed to track PC assets across the office. I imported that data directly into HELP!Desk — it used an easily accessible dBASE-type database. The goal was to ensure that asset information was integrated with the helpdesk, so that when a tech opened a ticket for a user, the hardware information was already there, accurate and current. No one required it. No one asked for it. It was simply the right way to build the thing.

The handoff, and what followed

As the agency grew, the helpdesk software and its data were eventually handed off to someone else. This is how organizational change works: responsibilities migrate, roles get reassigned, and the person who built something is no longer the person tending it.

The person who took it over didn’t share my priorities. The accuracy of the data wasn’t important to them. The knowledge base wasn’t a resource worth maintaining. Within a relatively short time, the system had become something else entirely — a tracking tool, basically, logging that a ticket existed and had been closed, with nothing much in between that would be useful to anyone. From a distance, and without any authority to intervene, I watched something I had invested years of genuine effort into become essentially hollow. Looking back, this was an early lesson in what it means to feel responsible for something you no longer control. I was more disappointed by this than I probably should have been, and I say “probably” only because I’m still not entirely sure the disappointment was disproportionate.

Remedy — and the performance-based contract

We eventually moved from HELP!Desk to Remedy, an enterprise IT service management platform that was, by any honest measure, far more product than we needed. We were running, at best, a two-tier shop, and Remedy was architected for large, multi-tiered helpdesk operations — three or four tiers out of the box. We deployed it in the default configuration and used it accordingly, which meant paying for a great deal of capability we never touched.

I would have loved to take that product on. I knew what could be done with it, and I believed I could shape it into something that actually served the organization the way HELP!Desk had. But by then, helpdesk operations had been turned over entirely to a performance-based contract — which was, at the time, fashionable in federal IT management circles as a path toward accountability and cost control. The theoretical model made sense. The execution, in my experience, was frequently a problem.

The company that won the contract required its technicians to wear suits — which meant people regularly crawling under desks and into wiring closets in dry-clean-only clothing. One of the techs, to his credit, kept a shop apron at his desk and put it on whenever he had to work on a system. The suit requirement was, in a word, ridiculous.

The Team Lead role was assigned to a colleague whose background was in contracting and acquisition, not technology. She was good at what she was trained to do. She was not equipped to evaluate what the contractors were telling us, which was a real and recurring problem. Performance-based contracts in that era were often structured in ways that invited abuse: the performance metrics were too broadly drawn, the incentives were tied to self-reported outputs, and most government-side staff lacked the technical grounding to push back effectively when the contractor’s account of their own performance didn’t match observable reality. I could tell when something didn’t add up. I was not in a position where anyone was asking me. I could see how the tools and contracts could be shaped into something better; I simply no longer had the authority to do it.

A colleague who understood the feeling

ACF acquired a knowledge base system called RightNow Web, a cloud-based customer service and knowledge management platform. I’m fairly certain we were among the earlier government customers. A colleague of mine managed and championed that product, and she had about it exactly the same sense of investment and ownership that I’ve been describing. She used to say, with no small amount of exhaustion, that it consumed an enormous amount of her time and energy. At some point I offered to help, and the offer was declined. I don’t hold that against her — I understand the impulse perfectly, because I felt it too. What I can say honestly is that I don’t remember ever turning away someone who offered to pitch in on something I was running. Whether that made me more collaborative or simply less territorial, I’m not sure. Unlike my own projects, she also had contractor support to direct, which was a resource I rarely enjoyed.

RightNow Web served the agency well until Oracle acquired the company in late 2011 — a deal valued at roughly $1.5 billion — and, as tends to happen with enterprise software absorbed into Oracle’s portfolio, the price increased substantially. ACF walked away. The content migrated to plain HTML. I wasn’t surprised. By then I had learned how often good systems, once absorbed into something bigger, are left to wither or priced out of reach. She understood, maybe better than most, what it meant to pour yourself into a system you didn’t ultimately control.

The voicemail system — and the software nobody knew we had

At some point I became involved in managing our in-house voicemail infrastructure. We were running an Octel system at the time, connected to a Centrex telephone environment with both analog and ISDN lines. Octel Communications, founded in 1982, built systems widely regarded as the most reliable on the market — designed to integrate with virtually every major PBX and engineered for hardware and software redundancy. In my experience, that reputation was fully warranted. I have never, before or since, worked with hardware that ran as reliably over as long a period. The thing was a tank.

When I started learning the system in earnest, I found something unexpected: the Octel had network capability that had never been enabled. The system was running in a completely standalone configuration — no network connection, no integration with anything. And we owned software that would allow users to access their voicemail over the network and export messages as audio files. We had owned that software since the hardware was purchased, years earlier. As far as I could determine, nobody had ever attempted to implement it.

I made it a project, got it done, and moved on. It seemed like a natural thing to take on, not a particularly heroic one, but it was exactly the sort of unattended capability I felt responsible for.

The value of that implementation wasn’t fully apparent until the office relocation. The Octel was being decommissioned as part of the move — we were transitioning to VoIP — and there was no mechanism to migrate voicemail content from the old system to the new one. The technology simply didn’t exist. If users wanted to keep a message, they had to export it themselves before cutover. Because the network client was already in place, they could. I’ve thought about that more than once: if nobody had taken on that implementation — and for years, nobody had — there would have been no option to offer.

The naming change — seeing the problem first

Around the same time, HHS pushed a small-OpDiv network naming reorganization. ACF was required to move to the HHS enterprise standard naming scheme of first.last@acf.hhs.gov. On paper, it was just an email address change. In practice, because we were already using PIV cards for Windows logon and the User Principal Name lived in the Subject Alternative Name on the certificate, it meant our logon identifiers were about to change too — which in turn meant re-issuing or re-encoding every PIV card we had.

I pointed this out to the department-level lead, who was not a technologist, who told me I was wrong. Then I explained the problem to my boss. He wasn’t sure I was right either, but, as he told me later, he’d noticed I was right more often than I was wrong. He went to bat for me. It was eventually confirmed that the PIV certificates would have to change, and the result was an all-hands effort to get more than 1,800 staff scheduled for PIV updates. I didn’t do the re-encoding; what I did was see the collision coming and insist that someone take it seriously. It was one of the rare times when someone with authority trusted my understanding soon enough to avoid making a bad decision even worse.

The office move — a database built after hours

GSA had ultimate oversight of the move; they brought in contractors to assist with move data management, and their proposed approach was a set of spreadsheets. Anyone who has seriously worked with spreadsheets knows that consolidating data across multiple files is a nightmare — version conflicts, manual reconciliation, no referential integrity, no audit trail. It was absolutely the wrong tool for the job. I lobbied management for resources to build a proper relational database and asked our IT contractors for help developing one. Management turned me down. The contractors said they lacked the time and capacity. The Contracting Officer’s Representative agreed with them — a pattern that recurred throughout my career with a frequency I can only describe as dispiriting: contractors declining work plainly within scope, and CORs consistently taking their side.

The GSA contractors had also introduced the concept of “receipts” — documents that employees would sign to certify they’d received their boxes and equipment at their new location. It was a sound idea, and I adopted it directly. But rather than a Word mail-merge, which was apparently what the contractors had in mind, I output the receipts programmatically from the database.

The whole system was built in Microsoft Access, mostly after hours — evenings and weekends, working alone, because it was clear from the outset that I was going to be working alone. Access has never been the recommended platform for mission-critical systems; it is a single-user desktop database at heart, with real limitations in concurrent multi-user environments and in data integrity at scale. I knew those limitations. I built within them, and it worked.

The identity underneath all of it

Being a technologist was never just a job function for me — it was part of how I understood myself. I liked talking with other techs. I liked the specific pleasure of shared technical experience, the shorthand that exists when two people have both worked through the same class of problem. I learned from those conversations and, I think, contributed to them.

What I found consistently difficult was working within structures where the people making decisions about technical work weren’t operating at that level — where the authority was disconnected from the expertise. That gap produced most of the frustrations I’ve described here: the contractor who wore suits, the COR who couldn’t evaluate a contractor’s self-reported metrics, the decision-makers who didn’t understand why a knowledge base needed to be maintained carefully or why a spreadsheet was the wrong tool for a relocation database.

The autonomy I was given was real, and I valued it. What I couldn’t always reconcile was the inverse of that autonomy — the fact that the things I built, once I was no longer the one building them, were subject to decisions made by people who didn’t necessarily understand what they had. In other words, I was given ownership without any guarantee of stewardship after I was gone.

That is not a complaint so much as a structural observation. I encountered it more than once, in more than one form. I suspect anyone who has spent a career caring about the quality of what they produce has run into some version of this too.

My summer of firsts

A solo flight, a grandfather’s silence, a carbon monoxide leak nobody knew about, and a summer at Scout camp — all in the same season.

The solo flight

One summer when I was in elementary school, around the Fourth of July, I took my first airplane trip alone. My mother had arranged everything: a solo flight to visit my maternal grandparents in western Massachusetts. My grandfather had recently retired, and they’d just moved into a brand-new trailer home in a newly built retirement community. I was excited and terrified in equal measure. It was the kind of independence a kid dreams about but doesn’t fully understand until the moment a flight attendant buckles you into your seat and the door closes.

My mother bought the tickets at Gulliver’s Travel in Free State Mall. For reasons I never learned — maybe the travel agent thought it would be more interesting for me to experience both jet and propeller travel — the itinerary was a jet northbound and a small propeller commuter flight home. The northbound leg was only about an hour and a half — Baltimore to Hartford, just long enough for beverage and snack service. I was nervous the whole way, but the flight attendants were wonderful. They brought me up to meet the captain, and I came back to my seat wearing a pair of shiny plastic pilot wings pinned to my shirt. I kept those wings for years; I may still have them somewhere.

Hartford, and Grandma’s errand

Grandma was waiting at the gate in Hartford. She was practical as always and, while we were still at the airport, decided to double-check my return arrangements. The propeller flight simply didn’t exist. I don’t know how she fixed it — whether it cost extra or how many phone calls it took — but the relief I felt was immediate. I had been quietly dreading the smaller plane. In hindsight it was almost certainly a turboprop, but to a small boy it sounded like something that would rattle and shake the whole way home.

The trailer

The trailer was brand-new and laid out exactly as you’d expect: bedrooms at each end, with the living room, kitchen, and dining area in the middle. I slept in the bedroom that contained the utility closet — the one that housed the HVAC unit and water heater and was accessed directly from my room. That detail would matter later.

Riverside Park

We did ordinary things that felt extraordinary because I was far from home and it was just the three of us together. One day we drove to Riverside Park in Agawam, a classic regional amusement park with rides and a big, noisy midway; a local institution that draws families from all over western Massachusetts — It’s now known as Six Flags New England. I loved it until I fell hard coming down a ramp after exiting a ride and scraped the entire side of my torso raw, then loved it again as soon as the sting dulled and the bandages were on. The park first-aid station patched me up, and I wore the bandage like a badge of honor for the rest of the trip.

The fishing trip

Grandpa’s neighbor had a boat, and one morning we went fishing on the local reservoir. It was my first time fishing. We trolled slowly across the water while the two men sat in silence. Grandpa had suffered permanent hearing damage working in an aircraft factory during World War II; he wore hearing aids, but they never seemed tuned to a child’s voice, and I assumed it was easier for him to hear adults. The quiet felt awkward to me, as if I were intruding on something private. At lunchtime they beached the boat and we ate sandwiches on the sand. For some reason I picked up a piece of driftwood and hurled it into the water. The neighbor yelled at me — I still don’t know exactly why he thought it would interfere with the boat — but the moment stuck with me as one of those small, inexplicable adult reactions that children file away without understanding.

Dirt piles and the broken commander

The retirement community was still under construction. Neighboring lots were empty, and there were big piles of dirt everywhere. With no other kids around, I spent hours playing on them. I had brought along my prized remote-control tank — complete with a little plastic tank-commander figure standing in the turret. When I unpacked it at my grandparents' trailer, I discovered that the commander had snapped off. The tank itself was fine, and I spent the rest of the visit running it up and down those dirt mounds like a little general conquering new territory.

Fireworks on the road

We went to the local fireworks display on the 4th — or rather, we stood on the periphery, on a road somewhere, because Grandma wasn’t enthusiastic about getting any closer. I had to plead (and probably cry) to get her to agree to go. It was worth it.

The cough, and what might have been

There was one strange thing about the trailer. Every time I went into that bedroom I started coughing — hard, nonstop coughing that didn’t happen outside. It would stop the moment I stepped back into the living room. My grandparents didn’t discover the problem until after I had gone home. There was a leak; we didn’t smell anything, so it wasn’t ordinary gas. I don’t remember being told what it was but I assume it was probably carbon monoxide. In hindsight, the trip could have ended very differently. I was lucky.

Camp Goshen

I flew home on a jet after all. Almost immediately I was packed off again — this time to my first sleepaway camp. It was Camp Goshen, the Cub Scout camp run by the National Capital Area Council in the Blue Ridge Mountains of Virginia. We rode a bus from Bowie; the camp sat in rolling, hilly country that felt like mountains to a suburban boy.

We slept in army-style tents on wooden platforms — half-walls, canvas roofs, cots, and sleeping bags. Each pod of tents circled a shared picnic table and grill. A teenage counselor (probably a Boy Scout) oversaw us. The communal showers were open-air stalls with walls that didn’t quite reach the ground. We sat outside waiting our turn, a line of boys in towels. Someone later told me I had squatted down at one point and been fully visible to the others. I have no memory of it, but the story circulated among the pod like camp legend.

I was still coughing. The camp infirmary checked me, said I had a temperature, but decided I was well enough to stay. We hiked every day — long trails through the woods. I could walk forever if the pace was steady, but the constant stop‑and‑go of our easily distracted troop of boys drove me crazy. One afternoon I faked a cramp. A counselor took me on a shortcut trail straight back to camp. It turned out we had been hiking in a big loop around the perimeter the whole time.

Breakfast was cooked on the grill at our picnic table — lots of bacon, a smell I can still conjure decades later. Lunch and dinner were in a central dining hall a good walk from our pod. Arts and crafts happened there too. In between activities we just hung out at the picnic table, boys being exactly what boys are: rude, loud, and obsessed with fart jokes — some of which even managed to disgust our counselor.

The diagnosis, and looking back

When I got home my parents took me to the doctor. The diagnosis was bronchitis. I suspect the antibiotics helped, but I’ve always wondered whether the fresh air and constant activity at Camp Goshen — right after whatever I had been breathing in that trailer bedroom — did even more. The cough finally let go.

Looking back now, that summer feels like a small hinge in my childhood. It was the first time I traveled alone, the first time I saw my grandparents in their new life, the first time I really felt the distance between Bowie and the rest of the family. My aunt, uncle, and cousins lived in the same town and saw them all the time; I was the far-away grandson who visited once in a great while. His hearing loss made conversation hard. I regret that I never really got to know him the way I wish I had. The silence on the boat that day wasn’t just about his ears; it was about two generations that didn’t quite know how to bridge the gap.

The trailer cough, the broken tank commander, the dirt piles, the Riverside Park bandage, the fireworks on the roadside, the camp bacon and fart jokes — they all live in the same summer in my memory. It was the summer I learned that independence tastes a little like excitement and a little like loneliness, and that the things that almost hurt you are often the ones you remember most clearly.

Retirement, then and now

That summer I unknowingly stepped into two different dreams of what retirement should look like. My grandfather had worked his whole life on someone else’s time. What he wanted most was simple and quiet: fishing on the reservoir when he felt like it, sitting on the deck of the trailer with no clock to obey. My grandmother hoped for more movement — travel, new places, the things they had postponed for thirty years. They had downsized, sold the family home that my grandfather and uncle had built together with their own hands, and moved into the brand-new trailer in the retirement community. Presumably that freed up resources for the travel and new experiences she wanted. It didn’t last; before long, their marriage ended in divorce.

My own retirement has taken a different shape. After decades in federal service, I now have the time I once envied. Instead of rest or constant travel, I spend my days doing what I could never do while working: digging into Bowie history, building the military visualizations, and writing the essays that fill these pages. The days are full, but they are full on my own terms.

I think of Grandpa now when I sit down to work on these pages. I wonder what he would have thought of a retirement spent this way — not in silence or miles traveled, but in local history and stories finally written down.

Junior High inside the Beltway: Two Years on the Bus

A personal account of forced busing from Bowie to a junior high school inside the Capital Beltway in the early 1980s.

One of the memories that has stayed with me longest — the one that still surfaces sometimes — is the two years I spent at junior high inside the Beltway. It was from about 1980 to 1982, when I was bused there from Bowie for seventh and eighth grade. The building was two stories. The frontage you saw from the road was actually the second floor — cafeteria on one end, gym on the other, library in the middle. Down below was a square of classrooms wrapped around an open courtyard. That lower level is where most of my memories live.

Junior high kids are naturally feral anyway — bodies racing ahead of their brains, everyone trying on new attitudes at the same time. But this school felt different from anything I had known in the neat, orderly suburbs of Bowie. The attitudes and body language — the way the kids who actually lived there carried themselves — felt foreign. They also had this whole system of “hoods” — neighborhood-based cliques and rivalries — that seemed to matter more than anything else. They knew who belonged where, who could stand where, who had to watch their back. Those of us who were bused in from places like Bowie didn’t speak the language; we came from communities where the biggest drama was whose bike got left out overnight. Here the rules were invisible and enforced instantly. Our buses were part of the county’s court-ordered integration plan, one small piece of a system that shuffled thousands of students across Prince George’s County.

Making it harder, very few of the kids I had gone to elementary school with followed me to that junior high. As I’ve described elsewhere in the atlas, families who could avoid the buses often did, so the usual sense of a whole cohort moving together from one building to the next just wasn’t there. I know people who graduated alongside classmates they’d first met in kindergarten; my path broke away from most of my Bowie peers before junior high even began.

The bus arrived well before the first bell — extra time built in for traffic or to let the bus service other routes. For me that early arrival felt dangerous, so I spent it in the library every single morning. It was quiet. It was safe. I could sit with a book and pretend the rest of the building didn’t exist until the bell rang.

Fights broke out in the hallways, and the police were called to the school regularly. In my memory it was almost daily, though I’m sure it wasn’t literally every day. Still, it happened often enough that it became part of the background noise.

Thankfully it only lasted two years. The county phased out junior highs and switched to middle schools (sixth through eighth grade). Ninth grade I went to a school just outside the Beltway, which felt noticeably calmer. Part of that calm, I think, came from the fact that everyone was a little older and more settled, and part of it was that by then this strange new arrangement had become the new normal; we moved together from that junior high into high school, so there was finally some continuity again, just not the kind most kids had enjoyed from elementary school onward.

As rough as those two years had been, many of us were genuinely disappointed when they told us we’d be moving on to high school for ninth grade; we had expected to finally be the oldest kids in the building, and that rite of passage disappeared overnight.

I have a few bright spots I still treasure. In art class we had a real pottery kiln. One assignment was to make a bas-relief plaque based on a comic-strip character. I chose Garfield — the straight-on pose where he’s standing upright with motion lines and extra legs for that cartoon swagger. I still have the plaque on a shelf in my living room. The only problem is I forgot to glaze his eyes white before it was fired. They’re the same orange as the rest of him. Every time I look at it I smile at the mistake and remember the one class where I felt like I belonged.

Shop class was another place I actually enjoyed — until the day a bully was handing out metal rulers. He yanked one out of my hand after I had already taken it, slicing a long gash across my palm. I bled enough to be sent to the nurse’s office. In another incident in the same class a much bigger kid decided I had disrespected him as we were waiting in line for tools. He punched me hard in the back. I just took it. Bused-in kids learned quickly that fighting back wasn’t an option.

There were the teachers, too. I had one young English teacher — a man who seemed to genuinely connect with us — until he was suddenly reassigned to an advanced class. We got an older woman in his place who clearly hated being there and made sure we knew it. I still remember the resentment I felt toward her decades later. And then there was the math teacher who complained constantly that the whole class “stank.” He would spray air freshener around the classroom and even on the students regularly. In hindsight I wonder if it was mold in the room instead of seventh- and eighth-graders who hadn’t discovered deodorant yet.

For years after I left that school I had nightmares — dreams of running around the lower level, heart pounding, trying to find a way out. Fear of what, exactly, I could never quite name. Just the feeling of being out of place in a place that didn’t want you there. I don’t have those dreams anymore.

Looking back now, in retirement, I have the time I never had before to sit with these memories and write them down. The atlas has become the place where I finally unpack the things I carried quietly for decades — the good ones and the ones that still haunt me. It was only two years. But two years at that age can feel like a lifetime when you’re the kid who hides in the library until the bell rings — and this is simply the way those two years felt to me.

Ideas and the People Who Have Them

On academics, the bused, and a Thomas Sowell observation that has stayed with me.

The Mensa room

I have written elsewhere in this account about my father — a certified Mensa member, a genuine genius by any measurable standard, and by the only measure that counted in daily life, a terrible father. He occasionally took us to Mensa family events, and I came away from those events with no desire to interact with those people again. I acknowledged the limits of that perception at the time — I was a child, and adult social dynamics are mostly invisible to children. I still believe in that caveat. But the impression stayed, and it has had nearly five decades to either dissolve or deepen, and it has not dissolved.

What I sensed in that room, even as a child without the vocabulary to name it, was a quality I would later come to recognize in other settings: people who were very smart and knew it, and who had arranged their lives around the comfort of being among others who also knew it. Not all of them, surely. But enough to leave a residue.

I have come to think that many of those people were academics, or at least cut from the same cloth — people whose professional world is built around the production, exchange, and evaluation of ideas. I cannot know that with any certainty; it is an inference from a child’s impression, which is perhaps not the most reliable kind. But the inference has held up in ways I didn’t expect.

The adjunct problem, and what it reveals

Years later in college, that same pattern of insulation appeared in a different room. When I was in school in the early 1990s, a professor said something in class that I have never been able to forget. The topic was unionization, and she mentioned, matter-of-factly, that she was an adjunct who taught at three different schools in the DC area to make ends meet. She wasn’t making a political argument so much as describing a reality she had come to accept — the way someone who has adapted to a situation describes it without expecting anything to change.

I should say that I was not exactly a neutral observer in that classroom. As I’ve discussed in my previous essays, I was working at FSA (we hadn’t been reorganized into ACF yet) at the time, where representatives from NTEU and AFGE were actively campaigning for union representation. I had absorbed their arguments thoroughly enough that when the class held a debate on unionization, I volunteered to come down to the well and argue the pro-union side. I did so with considerable confidence.

What strikes me now is the particular shape of that moment: I was making the argument in front of a woman who had more cause than anyone in the room to make it herself — and yet the people she would have needed to direct it at, the tenured faculty and administrators who presided over the system that kept her commuting between three campuses, were not there to hear it. The debate was available to the students. The reckoning was not available to her.

The situation has gotten considerably worse since then. As of fall 2023, about 68 percent of faculty members in U.S. colleges and universities held contingent (non-tenure-track) appointments — up from about 47 percent in fall 1987 — according to American Association of University Professors (AAUP) analysis of federal data. The two-tier structure is now stark enough that it would be recognizable to a nineteenth-century mill owner: a small class of credentialed, economically secure elites with institutional power and autonomy, served by a large precarious workforce doing equivalent work for poverty wages and no meaningful job security.

What makes this particularly striking is that the tenured class is also, by an overwhelming margin, the class most loudly identified with egalitarian political commitments. The politics espoused and the institution inhabited occupy almost completely separate moral universes, and somehow the people who inhabit both are generally comfortable with the arrangement. The compartmentalization required is considerable. But then, perhaps that is a skill that develops naturally in a professional world where the end product is ideas and not outcomes.

The people who designed the buses

The busing essay I wrote for this collection is the piece I sat with longest before publishing. Two years at a junior high inside the Beltway, bused there from Bowie under the county’s court-ordered integration plan (Vaughns v. Board of Education, filed 1972 and implemented January 1973) — the experience is described there in enough detail that I don’t need to repeat it. What I did not write there, because I was trying to stick to memory rather than argument, is this: the people who designed that system did not, in general, send their own children through it.

This is not a fringe observation — it was widely noted and resented at the time, and it has not become less true in the decades since. The architects of large-scale social experiments of this type have historically tended to live at a remove from the consequences. Court-ordered busing in Prince George’s County continued (with modifications) until 1998. The children of the Prince George’s County families who were shuffled across the county to achieve statistical outcomes lived those outcomes in full. The people who drew the maps did not. Their children were in school elsewhere: private school, or a suburban district that had arranged its affairs to remain unaffected, or simply in a jurisdiction that felt no particular obligation to participate.

I want to be fair here: the intentions behind integration policy were not cynical, and the history that produced segregated schools was real and required a response. I am not arguing otherwise. What I am saying is that there is a specific kind of moral authority that attaches to people who design things for others to experience, and it has always seemed to me that it should require, at minimum, the willingness to subject yourself or your children to the same experience. When that is conspicuously absent, something important has gone wrong in the reasoning — even if the reasoning began from a genuine desire to do good.

The Sowell observation

Thomas Sowell, the economist and social theorist, has written extensively on what he calls the vision of the anointed — a worldview shared by a credentialed class that is certain it perceives problems that others cannot see and possesses solutions that others are too parochial or too self-interested to accept.

The specific observation I keep returning to — paraphrased — goes something like this: the defining characteristic of intellectuals is that their end product is ideas, and unlike engineers or surgeons or anyone else whose work is tested against reality, intellectuals face no comparable accountability when their ideas fail. The engineer whose bridge falls down is answerable for it. The intellectual whose theory produces thirty years of harmful social outcomes has moved on to the next theory.

This is exactly the point I first heard Thomas Sowell make in the Hoover Institution video that introduced me to the idea many years ago: Thomas Sowell on Intellectuals and Society (Hoover Institution / Uncommon Knowledge, ~2009–2010)

Sowell is a conservative economist, and there are people who will dismiss him on that basis before reading a word. That reflex is, itself, a demonstration of the problem he is describing. The observation stands on its own. You do not have to agree with his policy conclusions to recognize that a class of people who shape institutions and policy from a position of permanent insulation from consequences is a class that will, over time, tend to optimize for the clarity and elegance of the idea rather than the welfare of the people the idea is supposed to help.

What ties these things together

I recognize that this essay has assembled some fairly heterogeneous material — a childhood impression of Mensa, an adjunct’s candor, two years on a school bus, a social theorist’s diagnosis. I am not certain they form a single argument so much as a recurring observation: that there is a specific kind of professional and social environment that produces people who are very confident in their ideas and substantially insulated from those ideas’ effects on others. The academic world is the clearest example, but it is not the only one.

I came to these observations not through any formal study — I am not an academic and am not pretending to the vocabulary of one. I came to them the way most people come to their convictions: through things that happened to me, things I noticed, conversations I didn’t expect to remember that I have never been able to forget. The adjunct professor’s offhand remark in a college classroom has held up better over thirty years than most formal arguments I encountered in the same building.

Whether any of this belongs on a public page, as I wrote at the start of the Memories of My Father essay, is a question I haven’t entirely resolved. The atlas is primarily a history, not a political document. But the essays section exists for exactly this reason: to put down the things that living in and around this community over six decades has actually produced in me — not just the memory of what stores occupied the Free State Mall, but what I actually think, while I still have time to write it down.

The house down the road

Or, how the smart home came to Bowie — and how the author followed it there, one gadget at a time.

The smart home was invented, in part, just south of Bowie. That sentence sounds like the kind of thing a local historian says when they want to inflate a footnote into a headline, but in this case the documentation is solid.

In 1984, David J. MacFadyen — an MIT-trained engineer who was reportedly standing in an unfinished house, staring at the customary tangle of electrical wiring, when the idea came to him — brought a proposal to the National Association of Home Builders’ annual meeting in Washington. The technology to unify a home’s electrical, communications, security, and appliance systems already existed, he argued. What didn’t exist was anyone willing to assemble the pieces. The NAHB was willing. Within months, a consortium of thirty manufacturers — GE, Honeywell, and Whirlpool among the charter members — had formed the Smart House Limited Partnership under new federal legislation that allowed competing companies to collaborate on research without violating antitrust law. Congress had, in effect, created the legal framework for the smart home.

The research campus was established at the NAHB’s 51-acre Research Home Park on Prince George’s Boulevard in Upper Marlboro — about 1.5 miles south of Route 214, near the present-day Amazon warehouse. By 1986, two full-size laboratory houses were under construction there: one all-electric, one using gas appliances, representing the two competing utility industries that had both signed on as partners. Baltimore Gas & Electric funded the first completed prototype off-campus — a three-bedroom contemporary in Woodlawn, Baltimore County, finished in 1989 at a cost of over half a million dollars and leased back to the Smart House partnership for two years of testing. A second laboratory, operated by the Gas Research Institute in cooperation with the NAHB, was built in Bowie itself in 1988, outfitted with refinements that included an outdoor gas grill and a gas fireplace, both controllable from inside the house.

Wireless technology was still decades in the future. The entire Smart House concept rested on a single unified cable — carrying power, control signals, audio, and data simultaneously — that would replace the “spaghetti” of separate wiring systems in a conventional house. Each Smart House would contain roughly 150 small computers, each about the size of a pack of cigarettes, serving as controllers. Outlets would not simply deliver power; they would recognize what was plugged in and supply only the voltage the device requested. A child’s finger in a socket would receive no shock, because a finger cannot identify itself as an appliance. A washing machine could flash a message on the television screen when the load was done. The house could be called from a car phone and instructed to start dinner.

The wiring standards were written into the 1987 National Electrical Code. The spokeswoman predicted that fifty percent of all new homes built between 1995 and 1997 would be Smart Houses. None of this happened. The technology proved difficult to mass-produce, appliance manufacturers resisted retooling, and consumers were put off by the combination of high cost and unfamiliarity. By 1992, Smart Houses were selling in only selected markets. By 1997, the Limited Partnership dissolved. The future had been built, tested, documented, and set aside.

The essays elsewhere on this site describe a pattern that began in childhood and never really stopped: a father who communicated with mainframes through an acoustic coupler in the 1970s, a teenager typing BASIC programs from magazines on a Tandy MC-10, a young federal employee who was put in charge of training staff on Wang terminals not because anyone assigned him the role but because he was the person in the room who already knew how the machine worked. Being a technologist was never just a job function. It was part of how I understood myself.

A significant part of that identity was formed at Radio Shack. There was a location at Pointer Ridge Plaza, at the intersection of Pointer Ridge Drive and U.S. Route 301, that I visited as often as I reasonably could — and sometimes more often than that. When that location closed I moved on to the Hilltop Plaza location off Route 450 and Racetrack Road. I bought gadgets, cables, adapters, and electronic components there regularly — components to experiment with, parts for projects that ranged from practical to purely curious. Radio Shack occupied a particular cultural niche that has never quite been replaced: a store where the staff actually understood what they were selling, where you could walk in with a half-formed idea and leave with a component that made it real, and where browsing the catalog was itself a form of continuing education in what technology was becoming.

So when Radio Shack began carrying X10 home automation products, I was paying attention — and for a very specific reason. The ability to turn lights and appliances on and off anywhere in the house, on a schedule or at will, seemed to me genuinely and practically useful. Not as a novelty. As a tool. X10 products first appeared in Radio Shack stores in 1978, grown from a Scottish engineering company called Pico Electronics that had developed the protocol in 1974. By the late 1980s the system had matured into something a technically inclined homeowner could actually deploy: plug-in modules for lamps and appliances, wall switches, timers, and — the detail that made it feel genuinely futuristic — an alarm clock that could turn lights on and off anywhere in the house on a schedule. I installed a Radio Shack X10 system and considered it a minor miracle.

It was far from ideal. The control signals traveled over the existing household electrical wiring, and because the system relied on power lines it was susceptible to electrical interference — signals would frequently fail to cross from one circuit to another, leaving a lamp stubbornly lit or dark while the controller reported success. Appliance modules would miss commands. Troubleshooting an X10 system required the patience of a person who genuinely enjoyed troubleshooting. I was that person. What X10 offered, beneath its unreliability, was a genuine glimpse of what the Smart House researchers up the road were working toward in harder and more expensive form: a home that responded to commands, that could be scheduled, that extended the reach of human intention beyond the light switch on the wall.

That same orientation extended to software. Egghead Software — founded in 1984 and eventually grown to over 200 stores nationwide, with locations in the Washington metropolitan area — became another regular destination throughout the late 1980s and 1990s. My federal job involved researching and evaluating software products, which meant I arrived at Egghead already familiar with a wider range of applications than most customers. Over time I found I could explain some of the business productivity software better than the store staff could — the use cases, the limitations, which competing product solved which problem better. On at least one occasion, an Egghead staff member asked if I wanted a job. I took the question as a compliment and declined. Egghead closed its remaining stores in January 1998 and moved exclusively to internet sales, filing for bankruptcy in 2001 when its domain name was acquired by Amazon. It was the kind of loss that feels small until you realize nothing has replaced it.

The decision to step back from the bleeding edge of consumer hardware is a recent one — only in the last five years or so. Before that, the pattern was consistent: buy early, accept the rough edges, live inside someone else’s unfinished product, and wait for the firmware to catch up with the promise. I have accumulated, over the years, a fair amount of technology that never quite worked as advertised — including several Kickstarter products that arrived late, underperformed, and are now in a drawer somewhere. The lesson accumulated slowly. Early adoption has a cost that isn’t only financial: it is the cost of time spent troubleshooting things that should simply work, and of enthusiasm redirected from using technology to managing it. I still maintain that investment in computer software, where the bleeding edge tends to be more navigable. But for hardware and consumer devices, I learned — eventually — to let others work out the kinks first.

Virtual reality was the exception. My entry point was the Oculus Go, released in 2018 — a standalone headset at a mainstream-accessible $199 that required no PC and no external sensors. It was Oculus’s first all-in-one device, and for media consumption and lighter VR experiences it worked well enough to make the case for going further. Going further meant the Rift — the PC-tethered headset that represented the full high-end VR experience — which in turn meant first acquiring a gaming PC capable of running it. That was a significant investment on top of the headset itself, and it underscored a recurring theme in early adoption: the advertised price is rarely the whole price. The Rift delivered on its promise, but the setup complexity and the tangle of sensors and cables felt like a step backward from the Go’s simplicity. The Quest line resolved that tension. The original Quest in 2019, Quest 2 in 2020, and Quest 3 in 2023 each delivered a progressively better standalone experience that eventually matched and then exceeded what the tethered Rift had offered — without the PC, without the cables, without the sensors on the wall. I purchased each in succession. By the Quest 3, the gap between what VR promised and what it delivered had narrowed to something approaching a genuinely useful tool rather than an expensive demonstration.

The arc from Smart House to Quest 3 spans roughly forty years of the same underlying project: making the built environment and the tools within it responsive to human intention. The Smart House researchers in Upper Marlboro and Bowie envisioned a home that recognized its occupants, managed their utilities, protected their children, and called for help when they couldn’t. The execution required a unified cable, 150 cigarette-pack computers, and the cooperation of every appliance manufacturer in America — and it failed. What actually succeeded was wireless, distributed, and arrived through a completely different industry. The smart home the NAHB imagined in 1984 arrived anyway, eventually, sold at retail for thirty dollars as a smart plug, controlled from a phone, and requiring no cable at all. The Radio Shack at Pointer Ridge Plaza, where I first bought the X10 modules that were the consumer-grade premonition of all of it, closed years ago — another technology distribution channel that outlived its moment.

MacFadyen wasn’t wrong about the destination. He was wrong about the road.