One must have chaos to be able to give birth to a dancing star

“I am a disciple of the prophet Dionysus ” — Friedrich Nietzsche

Information Technology recommendations on data structures: the more structured, the more clean and standardized, the better. It increases productivity, makes everything more organized. Standards diminish rework and confusion with data. Excel spreadsheets are the worst thing in the world, because the user will create columns, insert empty lines, change the header: it will be a mess.

This discourse seems perfect.

But, also as Dionysus’ disciple, I say NO.

  • The more chaotic the database, the better.
  • The dirtier, the better.
  • The more the user messes up, the better.

Apollo x Dionysus

The German philosopher Friedrich Nietzsche had the remarkable ability to combine philosophical ideas with a poetic style. One of his most famous metaphors is the contrast between the Greek gods Apollo and Dionysus.

Apollo is the god of the Sun, Harmony, Medicine. He represents the Order: the beautiful god, tall, strong, symmetrical, organized. He is the god of the arts, giving precise shapes to the sculptures, creating order in chaos.

Dionysus is the god of wine. He represents the Chaos. Fat, short, ugly, drunk, crooked, everything in him is bad. He is the ecstasy, the drunkenness. Born of hunger and pain, he reborns each spring and spreads joy wherever he goes.

Order x Chaos

The order seems better than the chaos. But the fact is that we can bring order only to a very small part of the world. The universe, infinitely greater, will never be known by humans.

I imagine the two gods as two arms: chaos and order, complementing each other. We should have the humility to recognize that there are premises that will always be out of any model .

I have a very strong Apollonian side, by my background in engineering, math Olympiads, etc. But I also have a very strong Dionysian side, which makes me very skeptical of everything that wants to organize too much, optimize too much, giving little room to the unforeseen.

Why I like chaotic data structures?

I work extensively with innovation, creating new tools, new processes, new ideas.

When working with a new project, the client don’t have the slightest idea of ​​what he wants. He knows only the symptom (say, analysts lose too much time to generate the report), and assumes that he knows the solution (automatically generate a report).

But is this the real problem? No one knows, this must be discovered.

Sometimes, he even does not need the report he thought he needed. Anyway, in 100% of cases, it’s useless to structure databases to try to solve efficiently the wrong problem.

As Peter Drucker says:

There is no greater waste than to solve with great efficiency a problem that didn’t needed to be solved.

The recommendation of structured databases is only for mature processes, those already established and that will change little.

For innovations, the more prototypes, the better. The more quick-wins, simple, fast, flexible and therefore chaotic, the better. The dirtier the excel, the better. The more the client messes up, the better.

No new music arises from the order. No new picture emerges of order. No new idea arises from the order. Only from chaos.

One must have chaos to be able to give birth to a dancing star” — Friedrich Nietzsche

More forgotten lore at:

Portuguese version:


An abnormal man looking for an interesting number

After taking the picture below, from the building written “normal”, the abnormal guy of the photo — me — remembered an “interesting” paradox.

Suppose I list the natural numbers in order:

  • 1
  • 2
  • 3
  • 4

Now let’s say something interesting about each of these numbers:

  • 1 is the first number of all, it is divisor of all the others
  • 2 is the first and only prime number
  • 3 is the first odd cousin
  • 4 is the first perfect square

Let’s say the numbers with the an interesting property are called “interesting” numbers.
And the numbers that are not interesting are the “normal” numbers.

Using this definition, a list would look like this:

  • 1 is an interesting number
  • 2 is an interesting number
  • 3 is an interesting number
  • 4 is an interesting number

Now suppose the number x is the first “normal” number in the list.

  • 1 is an interesting number
  • 2 is an interesting number
  • 3 is an interesting number
  • 4 is an interesting number
  • x is a normal number

But if x is the first “normal” number, it is an “interesting” number because it has an interesting property: to be the first “normal” number.

On the other hand, if we consider x an “interesting” number by having the property of being the first “normal” number, it is no longer a “normal” number and now it is an “interesting” number, this way losing the property of being the first “Normal” and then ceasing to be “interesting” …

What a mess! It’s not “interesting”?

To tell you something interesting, this problem is the “Paradox of Richard’s Numbers,” described by the mathematician Jules Richard in 1905.

This link ( tells more details about Richard’s paradox, but in a less interesting way than here.

Another similar paradox is the “Liar Paradox”. A man who only tells lies says “I’m lying.” But as he only lies, he will be telling the truth in this statement. But if he speaks the truth, he is not the one who only tells lies.

These paradoxes “bugs” not only the minds of ordinary human beings, but also the minds of the greatest mathematicians in history.

The Austrian mathematician Kurt Godel demolished the foundations of all mathematics in 1931, with its Incompleteness Theorems, by proving that mathematics can not at the same time be Complete and Consistent. That is, mathematics has limits. Godel found a “bug” in the foundations of mathematics — it can not at the same time get rid of these bizarre paradoxes and answer True or False to all its propositions. Godel used a sophisticated version of Richard Paradox to prove it.

It’s a long story, which involves mind giants like David Hilbert and Bertrand Russell, and it’s for another day.

By the way, I think the author of the building light wrote “normal” in an “interesting” way only for the building not to be “normal” anymore, and thus to confuse our head …