Hacker News Reader

Comment by eastbound | Hacker News Reader

There’s a beautiful conference (which I’ve lost) saying that science and technology can regress. We always talk about “the progress” but things do regress all the time; Tests I’ve cleared for employment can’t be passed by newcomers, the NASA wouldn’t be able to put a rocket on the moon today, and there is no downloadable Jira competitor anymore.

In fact the speed of innovation is, pretty much, equal to the speed of maintenance. Nothing gets maintained, it’s either new or collapsing, but no-one enjoys the middle part. Which is sad. It’s a form of inflation.

Boeing is perfectly right to design a new plane right now. Engineers who interned on the 787 have bought a house on the countryside a few years ago.

View on HN

This here should trigger a red flag when you hear anyone say "The government should be run like a business".

Businesses kill themselves all the time from the loss of institutional knowledge. We see stuff like "government spends X a year paying people Y to build almost no product Z". Instead, we're paying people Y to be ready to build Z when something goes terribly wrong like war.

There are fixes for this: When I was at the Boeing ASL (pre MDD merger), there were plans to have a group that regularly created a "paper airplane" every two years based on market conditions and product line gaps. If the economy could support a launch, great, you've got a head start. If the economy is down, well, at least you've got the institutional capability to design airplanes.

It wasn't a particularly huge cost, and before Harry Stonechisler took over there was money for that type of work within a profitable BCAG.

I don't disagree with your larger point, but specifically in the scenario you outlined, the highly skilled, highly capable people that actually can produce product Z will not be satisfied with that kind of arrangement. All you will have is a daycare for bureaucracy barnacles.

When they say run like a business that doesn't mean that it should be done like a reckless business. Plenty of businesses invest into things like disaster recovery or insurance which they may never need.

Running like a business means providing services to profitable customers. This is, surprisingly, the exact opposite of what government is supposed to be doing. The USPS would be plenty profitable if it didn’t have to maintain the infrastructure to deliver a letter from Honolulu to Barrow for 50 cents. Medicare? Not the best business model. Roads? Forget it.

>means providing services to profitable customers

So Uber wasn't a business for its first 14 years? There are more objectives a business can have than maximize immediate profit. There are other metrics it can try and move.

Uber was a Business trying to gain a market share in hopes to cement itself as competitor or even become monopoly. It's also clear Uber was building massive ride hailing network so when self-driving cars became a thing, out with human drivers, in with computer drivers. Obviously, that didn't happen.

Government does not really have that. All their businesses have 100% market share so to run it profitable, they would have to stop doing unprofitable thing. Like providing healthcare to poor people or delivering from Hawaii to some rural area.

Nope. When people say governments should run like a business, they always mean the most reckless version of business you can think of.

It's even the only interpretation of that phrase that makes sense, because if you take the optimizing into large risks interpretation away, there's no other way those two can act in similar ways.

Is it this one?

https://www.youtube.com/watch?v=ZSRHeXYDLko Preventing the Collapse of Civilization / Jonathan Blow (Thekla, Inc)

YES. Yes it is. I need to learn his arguments by heart to demonstrate it myself.

> In fact the speed of innovation is, pretty much, equal to the speed of maintenance.

Instead, the most important innovation are the ones that reduce maintenance needs.

Automating farms, moving mechanical computation into general purpose processors, simplifying science theories so that people can learn in a semester stuff that took decades to mature... All of those have a tremendous impact. All of those speed everything up, and make room for more innovation to appear.

> and there is no downloadable Jira competitor anymore.

How much are you or your enterprise willing to pay for that?

See, the economics just don't make sense. Give me some VC money, a small team of experts in the domain and some runway and I'll give you a self-hosted alternative to Jira.

But the cost for you would be too high and I will go out of business trying to sell software while everyone else is renting it out.

Reply on news.ycombinator.com