Domain Specific Language

Lots of ideas, not much thinking

  1. Lots of ideas
  2. Not much thinking
  3. References

In the year of our lol'rd 2026 we've seen something fAnTaStIc happen. We're witnessing the movement of ideas from the public to the private. I'm not happy about it but also; there's nothing I can do. Not much anyway. Mathematicians are apparently getting idea sniped (but also execution sniped) hard enough for them to go dark. Is everyone happy about this? Is anyone happy about this?

You know, one of the things that really hurt Apple was, after I left, John Sculley got a very serious disease. And that disease - I've seen other people get it too. It's the disease of thinking that a really great idea is 90% of the work. And that if you just tell your - all these other people, you know, here's this great idea, then of course they can go off and make it happen.

Steve Jobs

Lots of ideas

Ideas used to be free or cheap. Some people treated ideas as state secrets, but the truth is most ideas were only worth as much as the work as you put into them. Spend a million dollars to build something, it might be worth ten or a hundred. If you spent zero... probably worth zero.

There's obviously been a shift or pivot or whatever you'd like to call it. Some of the worst people you know are doing a lot of stuff now. Scammers? Check. Social media influencers? Check. DHH? Also check. Are the ideas still cheap? I'd wager that spending a million is still worth a lot with the right idea. But people with a lot leaner wallets are now in a bidding war against you for all of those ideas. I'm sure someone would say that things have been democratized (ick), but I don't know, man.

The thing Steve Jobs said right after the quote above goes hard:

And the problem with that is, is that there's just a tremendous amount of craftsmanship in between a great idea and a great product.

Steve Jobs

Not much thinking

Building software still requires writing source code. One of the most effective levers we have in software development is abstraction. When you use a good abstraction you can almost feel it in your bones. A neat interface that does a lot with a little. Something that fails to hide the underlying complexity of a problem is a Leaky abstraction 1, and I've heard Bryan Cantrill 2 talk about good abstractions as Crisp 3. A leaky abstraction is one that bears an immediate or delayed cost. It might hurt you in the moment, being a pain to use, or it may hurt you a month from now during an outage.

Computers are abstractions all the way down. A struct, a slice of memory, and a cache line all represent the same thing. A function, a pointer, machine code, transistors firing - all the same thing! You're living in the future, child. You can call functions that communicates with the other side of the world like nobody's business. All through the power of abstraction.

Creating a leaky abstraction is laughably easy. Leaky abstractions are a dime a dozen, and probably the cause of your favorite recent production outage. Want to store "a file", "in the cloud", lmao? All abstractions. A crisp abstraction on the other hand has a price. The price is thinking. You internalize the problem, you talk to your colleagues, you ask them to challenge your ideas (or stroke your genius). Grinding your axe. Once your axe has been sharpened on the power of your thoughts, you strike. Strike hard and true.

People will tell you that all you need now is an idea. Or many ideas. I believe much of the software that has been written the last six to twelve months is of the "ideas" variety, not of the "thinking" variety. I'd prefer if people applied the thinking modus of software making, but it seems to have taken a beating lately. Like the old software joke where you have to pick two out of fast, good, or cheap, but you end up picking "fast" twice. It's not cheap, because you're going to pay the price at some point. It's not good, because you had the idea but didn't spend the time thinking.

Thinking can be deep, and thinking can be wide. Thinking can be done at any time of the day, almost for free (bring your own lunch). Yelling your ideas into the void is cool, but software abstractions are usually only as good as the thought that was put into them. If I could only make one recommendation in this string of words it would be putting a lot of thought into them.

Some organisms thrive in the fast-fast environment. Viruses and bacteria don't have much capacity for thinking, so usually fast is good. Humans on the other hand are built for thinking. Grey matter gobbles up a lot of energy, but using it pays dividends. Or used to. Ideas men are like kryptonite to organisms organizations. Look! I'll turn all these ideas into action without having to do all of that stupid dumb thinking that we had to do before. Crisp abstractions? How about no abstractions? At least none that anyone thought about.

The leaking will be left to the cleanup crew.

References