Et Cetera


The point of a journey is not to arrive.

Neil Peart wrote my operating philosophy into a Rush song called Prime Mover in 1987. I’ve been using it as a permission slip ever since.

I started wandering the online world in the mid-80’s with a dial-up modem attached to an Atari 1200xl with no particular plan, and I’ve stuck to that plan with real discipline. When the world wide web arrived, almost by accident I bought this domain. It has evolved over the years, but landed eventually as a weblog. What you’ll find here is three decades of it, written down as it happened: daily links to tools that were going to change everything and didn’t, hot takes that cooled badly, arguments I’d lose today, quips, half-baked concepts, and the occasional post about a road trip or a gong.

The digital age never shipped with a compass. I checked. So some of this meanders, some of it is beside the point, and a surprising amount of the beside-the-point stuff turned out to matter. The essays behind the pillars started here, as a link or a hunch, years before they became a theme.

If you know what you’re looking for, search. If you don’t, welcome. You’re doing it right. Wander on.

The eras

The link blog, 2005 to 2008

Daily links, the Web 2.0 tool parade, and the first arguments about software as a service.

The crash, 2008 to 2011

Finance, market resets, and the downturn seen from inside enterprise software.

The long middle, 2012 to 2019

Fewer posts, longer thoughts. VRM, personal clouds, the decline of the cookie, and an internet that remembers too much.

The return, 2020-

AI, markets, moats, and so much more.

The thread

Kevin Kelly framed the future in The Inevitable as twelve verbs: becoming, remixing, filtering, sharing, questioning, beginning, and the rest. Technology is something that keeps happening, and the rest of us are endless newbies, forever starting over as the tools change under us.

I’d like to report that thirty years online made me an expert. It made me a very experienced beginner. Read this archive in order and you can watch the verbs at work: links became feeds, feeds became streams, streams became agents, and I explained each one with total confidence before relearning it from scratch. The book I’m finishing is about what an enterprise does with that same problem, at scale, with a budget.

From the archive

  • Remembering September 11, 2001

    It’s hard to imagine that a decade has passed, yet it seems just like yesterday. Ten years ago, on this day, I boarded an early morning flight from Pittsburgh to New York’s LaGuardia airport. I was beginning my weekly travels across the northeast a day late, delaying the routine Monday departure for Boston to be home for an extra day. A decade ago, I was working for Siebel Systems on…

Subscribe

Catch the next idea while it’s still early.

Subscribe →

Everything, newest first

  • Bay Area Tech Economics Seminar with Abhishek Nagaraj

    Abhishek Nagaraj on using large language models as a stand-in for human-subject experiments in social science.

  • DevRev Leadership Circle: trust, outcomes, and the slow death of the seat license

    Notes from DevRev Leadership Circle: Adobe’s Anil Chakravarthy on trust and Mayfield’s Gamiel Gran on why buyers want to pay for outcomes instead of seats.

  • How AI Is Changing the Economy: New Evidence on Work, Growth, and Economic Transformation

    A Stanford Digital Economy Lab conference on new evidence about how AI is changing work, productivity and growth.

  • AI in Science Collaboration Hackathon at Stanford

    A one-day Stanford hackathon on how researchers and their AI agents find collaborators and work together across fields.

  • VRM Day 2026: contracts, fiduciaries, and who your agent works for

    Notes from VRM Day at the Computer History Museum on MyTerms, digital fiduciaries and who your AI agent works for.

  • Remembering Om Malik

    I just saw the news, where else, come across my RSS feed, that Om Malik has passed away. There are very few bloggers, reporters, and consummate insiders in Silicon Valley. On all those fronts, Om was an OG. His posts, reports, and photography were among a few constants on my RSS feed for more than two decades. His unfiltered views on things were always worth a read. He will be…

  • The Gong Was the Easy Part

    There was a Zenith console television in our family room, the kind built into a slab of wood furniture that weighed more than the couch, and a good part of my childhood happened in front of it. That’s where I met Charles Nelson Reilly and Jamie Farr and Jaye P. Morgan, though I couldn’t have told you their names at the time. They were just the funny people who turned…

  • Understanding Sector Change: The Role of Business Models in AI Adoption, Part 2

    Part 1 of the sector clock established that the sectors restructuring fastest share a common profile: digital-native workflows, clear and measurable units of value, relatively low regulatory barriers to workflow change, and competitive dynamics intense enough to punish non-adoption on a short timeline. Financial services, legal, and professional services meet most of those conditions. The restructuring in those sectors is advanced, compounding, and past the point of reversal. The sectors…

  • Understanding Sector Change: The Role of Business Models in AI Adoption

    The four clocks in this series have been running at roughly the same speed. Capital concentration, labor bifurcation, and the productivity trough are all happening simultaneously, which is what makes the current moment structurally distinct from previous technology transitions. But the fourth clock, sector restructuring, doesn’t run at a uniform speed across the economy. It runs fast in some industries and slow in others, and the difference is not primarily…

  • Faster Isn’t the Same as More Productive

    The Productivity Clock In 1987, economist Robert Solow made an observation that became one of the most quoted lines in technology economics: “You can see the computer age everywhere except in the productivity statistics.” It took roughly a decade for that to change. Computers had arrived. Organizations were using them. Productivity, at the macro level, was not moving. The explanation, when it finally arrived, wasn’t that computers didn’t work. It…

  • The Workers Who Were Never Hired

    Part 2 of 2: The Labor Clock This is Part 2 of a two-part piece on the labor restructuring underway in the AI economy. Part 1 covered the bifurcation already visible in current data and the historical pattern of who captures productivity surpluses. This piece goes underneath that, to a structural problem that moves on a longer timeline and will be significantly harder to address once it becomes legible. The…

  • Still Flying: Voyager 1 at the Edge of Everything

    Voyager 1 is more than 15 billion miles from Earth. It launched in 1977, before the personal computer, before the internet, before we decided anything older than eighteen months was legacy debt. It was built by people using slide rules. It is still on. Late last week, NASA turned off one of its last active science instruments to conserve power. Two remain: one listening to plasma waves, one measuring magnetic…

  • The Moat and the Mindset: Why You Must Watch Jensen Huang’s Latest Interview

    Dwarkesh Patel’s interview with Jensen Huang is required viewing: an unvarnished, often defensive Nvidia CEO, and a case for a US-led AI stack that echoes the dollar’s reserve status.

  • Who Gets Paid When the Machine Does the Work?

    Part 1 of 2: The Labor Clock This is a two-part piece. The labor restructuring is the most layered of the four clocks in the AI economic reconfiguration, so I’m breaking this up into two parts. Part 1 covers what the data already shows: who is capturing the productivity surplus, and why the bifurcation happening right now looks structural rather than transitional. Part 2 goes underneath that, to a slower-moving…

  • The roadtrip

    Tyler Cowen has one of his regular, random posts over at Marginal Revolution (excellent blog that is worth following): I have driven cross-country four times, at least if you count a 3/4 trip as valid.  I also have driving experience in virtually all states, including Hawaii and Alaska, neither of which would be part of typical cross-country travel. I recommend this mode of transport highly, especially for the United States.  Here…

  • Front-Loaded: The Economic Cost of AI’s Capital Concentration

    The AI infrastructure industry faces a significant financial challenge, needing approximately $800 billion annually to cover interest on investments. As capital concentrates in this sector, it affects broader economic stability, with potential underfunding of critical areas like healthcare and climate resilience. Two paths lie ahead: successful returns or fiscal strain.

  • The Line That Changed Everything

    The more you learn, the more you realize how many “obvious” things were actually invented recently, and usually by accident. Take the idea that time moves in a straight line. Past on the left, future on the right, you’re somewhere in the middle. So obvious it barely registers as an idea. Except it wasn’t always. A fascinating essay on Aeon traces the surprisingly recent origin of this metaphor and the…

  • How Apple Redefined Human-Machine Interaction Over 50 Years

    The mid-1970s marked a transformative era for children interacting with technology, transitioning from gaming consoles to early computers. Apple played a pivotal role in making technology accessible and relevant, emphasizing human creativity and intent. This ongoing interaction established Apple as a cultural operating system, shaping modern understanding of technology’s role in human expression.

  • AI’s running in four different races. We’re in all of them

    AI commentary is everywhere. Clarity is not. Over two decades in enterprise software, I’ve watched technology waves deliver differently than promised. What I’m seeing now isn’t one story — it’s four simultaneous economic restructurings in capital, labor, productivity, and sectors. Here’s a frame for tracking them together.

  • The uncanny valley of empathy

    Here’s an excerpt from my latest Substack post: The aspirational AI customer service deployments today aim to not feel cold or mechanical. They are meant to be warm, patient, apparently attentive. They’ll remember what you said three turns ago. They’ll acknowledge your frustration. They tell you they understand. But is that true understanding, or a very precise simulation of understanding? Read the full article here.

  • Random: Why the House Number Font Dominates American Homes

    Ever wonder why almost every house in America uses that same specific, slightly “fancy” font for its address? In this fascinating deep dive from Dime Store Adventures, we learn that the “House Number Font” isn’t actually a 1920s Art Deco artifact or a secret tribute to Chinese aesthetics, as the internet often claims. Instead, its ubiquity might just be the result of century-old neighborhood peer pressure. As the video perfectly…

  • What happens when the taxpayer disappears?

    The rise of AI poses a significant challenge to funding federal governments, as it threatens to displace a substantial number of knowledge workers who contribute to tax revenue. With the U.S. relying heavily on individual income and payroll taxes, the potential decline in employment raises concerns about sustaining government income and benefits like Social Security. Various tax proposals are being considered, but solutions are lagging behind the problem’s urgency.

  • Open AI’s fundraising approach, and what it reveals

    Om Malik has been doing some of the sharpest behind-the-scenes AI business coverage out there. If you’re not following him at om.co, fix that now. His latest piece: OpenAI is offering private equity firms a guaranteed 17.5% return to join joint ventures. Om’s read is that this is the kind of deal you do when you’ve borrowed against the future and the future is taking longer than expected. The tell…

  • On the underlying infrastructure for the AI boom

    Robert Friedland addressed USC’s Marshall School, warning business leaders about the unseen challenges ahead. He highlighted America’s heavy dependence on China for critical minerals and the urgent need for copper to sustain economic growth. Friedland emphasized that current infrastructure assumptions neglect vital energy and resource realities, suggesting a looming crisis.

  • Customer Experience, AI, and the paradox of efficiency

    The most dangerous thing about AI in customer experience isn’t that it gets things wrong. It’s that it gets things right — efficiently, cheaply, at scale — while quietly optimizing away the interactions that create real customer relationships. Think of a river straightened into a concrete aqueduct: the metrics improve, but the ecosystem disappears. That’s the trajectory we’re on with AI-driven CX today, and most organizations won’t notice until the retention…