Johnny

Picture a Tuesday night bowling league at the lanes on the edge of town. Eight lanes, eight teams. The scoreboards light up every week, the pins reset, the snack bar sells nachos. From the parking lot it looks like a healthy league.

Here is how it runs. The league pays its winners in free lane time. The top teams practice for free all month, so they win again, so they earn more free practice. Spots for next season go to last year's teams first. A new team has asked to join every fall for nine years. There has never been an open lane.

So is the league alive? The lights say yes. The games start on time. Nothing is broken. Still, no new team has bowled there since 2017.

A money network can end up in the same place. A network here means a shared money system run by many computers. Block producers are the ones who add each new page of records, called a block, and get paid for it. A network can keep making blocks on time and still be closed to newcomers. I use a three part test for that. Do a few producers do most of the work? Are they paid in the same coin that keeps their seat, so each reward grows the lead? And is there no force pushing coins back out to new people? Meet all three, and uptime tells you very little.

Bitcoin miners get paid in bitcoin, but they pay their power bills in dollars. That bill pushes many of them to sell coins every month, and those coins land with new holders. That leak runs in a healthy direction.

Now the fair pushback. The test depends on uneven starting points, and those can fade. Several networks people call closed have run fine for years. Mining has its own concentration too, in chip makers, cheap power regions and big pools. I mine, so I see some of that from my own shelf.

What would change my mind? Watch four numbers over a few years: how much the top producers control, how many new producers show up, how much supply sits outside the biggest holders, and how independent the operators are. If all four improve on a network, the test fails there, and I would say so out loud.

Eighteen years after the whitepaper, I check new entrants before I check uptime. A league with open lanes is the one I want to bowl in.

If this changed what you check first on a network, Zap ⚡

Picture a recipe card in your grandmother's handwriting. Pot roast, written in 1961, taped inside a cupboard door. Every holiday someone cooks from it, and every holiday somebody wants to change it.

One cousin wants less salt. Another wants to swap the beef for something cheaper. A third says the oven times are old and should be updated. Each change sounds small. Some of them are good ideas.

The card lets anyone in the family check whether the dish on the table is still the dish. Without it, the recipe becomes whatever the loudest cook remembers. After enough holidays, the family loses track of what changed and who changed it.

Networks work the same way. A network here means a shared system that many strangers use, like money that moves without a bank in the middle. The rules it starts with are its recipe card. Those first rules are called design principles. They are the promises the system made before anyone had a reason to bend them.

The promises a system starts with are the ones it can defend later. A feature added years afterward usually arrives as an option. Options get used by a few people, and the few who use them start to look strange to everyone else. "We will add that later" sounds like a promise. It rarely holds.

Now the fair pushback. Recipes should improve. Ovens changed. Some families learned the old way was unhealthy. Treating the card as sacred can be plain stubbornness in a nice apron, and it can keep a good dish from reaching more people.

I take that seriously. A network that never changes anything will break. Changing a cooking time is a different kind of change from swapping the main ingredient, though. One adjusts a setting. The other turns it into a different meal.

What would change my mind? A big network adding a core protection years after launch, like privacy for everyone by default, and most people using it within a few years. If that happened, I would have to admit late additions can stick.

Eighteen years ago today, nine pages went out to a cryptography mailing list. I read them in 2015, and I still use them as my card. When someone proposes a change to bitcoin, I ask one question first: is this a new cooking time, or a new main ingredient?

If this gave you a way to sort the next big debate, Zap ⚡

Picture an air pump at a trailhead. You roll up with a soft bike tire, drop in a quarter, and it fills you up. Nobody works there. The pump is a small box with a compressor, a coin slot and a little computer inside.

Now the town wants to drop the coin slot, because nobody carries quarters anymore. So they look at a card reader, and they hit a wall. Many card plans charge the seller a flat fee, often around 30 cents, plus a small slice of the sale. On a 25 cent sale, that fee is bigger than the sale. The pump would lose money on every tire.

So the pump has two choices. Charge a dollar, which feels like a rip off for ten seconds of air. Or go free, and hope the town keeps paying for repairs.

And the fee is only half of it. To accept cards at all, someone has to apply for a merchant account, pass a review, and agree that the payment company can pause the account if the sales look odd. A pump that does a hundred tiny sales on a sunny Saturday looks odd.

This is why I care about the rails machines use to pay and get paid. A rail is the path money travels on. For a machine, the rail needs to handle tiny amounts, let a new box join without asking anyone, and treat every box the same.

Bitcoin's Lightning Network was built for that job. Lightning is a layer on top of Bitcoin where payments move through channels, which are shared balances two parties lock up ahead of time. A payment of a few cents can settle in seconds for a fee of a fraction of a cent. The pump would hold a small wallet, show a code on a screen, and get paid. Nobody has to approve it.

The pushback deserves a fair hearing. Running Lightning well still takes work. Someone has to keep the wallet online, manage those channels, and deal with a box that gets stolen with sats inside. Most towns would rather call a vendor than learn any of that. And plenty of people will tell you a card is good enough.

Here is how I would know I'm wrong. If card networks start clearing a 25 cent payment for well under a cent, for any machine, with no application and no review, the case for a separate rail gets much weaker.

I have held bitcoin since 2015, and a quarter of air at a trailhead is the kind of small, boring job where I expect it to win.

If this made machine payments click for you, Zap ⚡

Picture a high school rowing team picking who sits in the stroke seat. That seat sets the pace for the whole boat, so it matters who gets it.

One coach runs a test every Monday at dawn. Every rower pulls two thousand meters on a rowing machine, and the fastest time takes the seat for the week. Last week's win counts for nothing. Next Monday, everyone pulls again.

Another coach gives the seat to the rower whose family gave the most to the booster club. The kid in that seat gets noticed by college scouts, which brings a scholarship, which frees up more money for the booster club next year. The seat pays for itself.

Bitcoin works like the first coach. Miners earn the right to add the next block of transactions by spending real electricity on specialized machines. That is called proof of work. The cost comes from outside the system, and it has to be paid again for every block. Nobody can coast on yesterday's power bill.

Many other networks work like the second coach. They use proof of stake, where the people who lock up the most coins get picked most often to add blocks, and they earn new coins for doing it. Those new coins can be locked up again. The say you have tomorrow grows out of the say you hold today.

I have run miners myself. The power bill shows up every month whether I find a block or not. Control has to be bought again and again from the physical world.

The other side has a fair case. Proof of stake uses a tiny fraction of the energy. Mining also has its own snowball, since large miners buy machines in bulk and get cheaper power deals than a small miner like me. A big enough operation can grow its share too.

How would I know I am wrong? If, over many years, the share of mining held by the top few pools grew faster than the share of staking held by the top few holders, the rowing story would stop holding up. I would also change my mind if a miner could keep control without spending fresh power on each block.

Until then I trust the Monday morning test. Bitcoin asks every miner to prove it again, about every ten minutes, and I keep a node running at home to check that they did.

if this made the Monday test click for you, Zap ⚡