There is probably a sensible way to run a small studio.

Pick one thing.

Define the market.

Build the thing.

Ship the thing.

Promote the thing.

Then, if the thing survives contact with reality, build another thing sufficiently similar to the first that everyone understands what sort of things you make.

Starhound Studio has not done this.

At various points, my desk has contained a World War II tower-defense game starring dachshunds, a gothic horror story about inherited hunger, music that probably should not exist, artwork for imaginary bands and real ones, and an experimental machine architecture whose internal geometry I occasionally photograph because apparently this is what my life has become.

This was not the business plan.

There wasn't really a business plan.

There was a question:

What happens if I follow the interesting thing?

That question has caused an enormous amount of trouble.

It has also become the closest thing Starhound has to a development philosophy.

The interesting thing is usually sideways

I don't think particularly well in straight lines.

An idea rarely stays inside the discipline where I found it.

Something I learn while programming a game wanders into a piece of music. A problem in machine architecture sends me reading about fluid dynamics. A visual experiment becomes a fictional world. A joke gets developed far beyond the point where a reasonable person would have stopped and, occasionally, becomes an actual project.

This sounds horribly inefficient.

Sometimes it is.

But spend enough time moving between fields and something strange starts happening:

They start talking to each other.

The borders become less convincing.

Programming is structure.

Music is structure.

Narrative is structure.

Visual composition is structure.

Mathematics is structure.

The materials are different. The rules are different. But many of the questions underneath them aren't.

Where is the tension?

What is carrying information?

What belongs together?

What happens when this moves?

What breaks if I remove that?

And my personal favorite:

What happens if I do the thing I'm apparently not supposed to do?

That question has been responsible for quite a lot.

Build it. Break it. Look at what happened.

I have a bad habit when something works.

I try to kill it.

This is particularly obvious in my work on Sodalis, an experimental machine architecture I've been developing alongside everything else.

Getting a good result isn't enough.

I want to know why I got it.

So I remove a relationship the system appeared to depend on.

I scramble a neighborhood structure.

I destroy information in one place and ask whether it can still be recovered somewhere else.

Change a constraint.

Attack an assumption.

Run it again.

If the result survives, I attack something else.

Eventually either the result dies, or I've learned something considerably more interesting than it worked.

There is a peculiar joy in discovering that something you built was wrong. Failure exposes things that a successful run can happily conceal.

That mentality escaped my machine experiments a long time ago.

Now it's everywhere.

Games get played until something feels wrong.

Songs get rearranged until the part fighting the rest finally reveals itself.

Stories get interrogated.

Art gets thrown away.

Code gets rewritten.

Ideas get subjected to the deeply scientific procedure of:

Okay, but what if I'm completely wrong?

Then I try again.

Occasionally something survives.

Those are good days.

The dachshunds were inevitable

This philosophy does, admittedly, have consequences.

One of them is Wiener Defense.

It is difficult to explain exactly how much work a person can accidentally create for herself by deciding that dachshunds should defend the home front.

It begins innocently enough.

Dachshund tower-defense game. That's funny.

Unfortunately, once you've accepted the premise, certain questions become unavoidable.

What does their propaganda look like?

What does their world look like?

Who are the enemies?

What does the battlefield look like?

What does a dachshund military-industrial complex look like?

And because apparently nobody was available to intervene, these questions received answers.

Then those answers produced more questions.

That's one of the dangers of following an idea seriously.

A ridiculous premise does not necessarily produce ridiculous work.

Sometimes the joke is simply the doorway.

Walk through it far enough and there is an entire world on the other side.

The projects contaminate each other

This may be my favorite part.

Nothing stays politely inside its own box.

Making games changes how I think about systems.

Building systems changes how I think about stories.

Writing stories changes how I think about music.

Music changes how I think about timing and tension.

Visual art changes how I think about information density.

Then something learned in one place wanders back into another project six months later wearing a fake mustache and pretending it was supposed to be there all along.

That cross-contamination isn't an unfortunate side effect of doing too many things.

I've come to think it's one of the reasons to do them.

Specialization can take you incredibly deep.

But sometimes you need to walk sideways before you can see the hole in the wall.

Starhound is the container

For a while, I thought the problem was that I had too many unrelated projects.

Eventually I realized I'd identified the wrong problem.

They aren't supposed to be one project.

Starhound is the thing that makes room for all of them.

A studio doesn't have to represent a medium.

It can represent a way of making things.

Mine happens to involve curiosity, obsessive experimentation, dark humor, strange machines, dogs with questionable military credentials, occasional mathematical violence, and an almost pathological inability to leave an interesting question alone.

There will be games.

There will be music.

There will be fiction.

There will be machines.

And there will be experiments that refuse to fit any of those categories.

Some will work.

Some will fail spectacularly.

Some will be dismantled because they worked and I wanted to know why.

Some will mutate into things I cannot currently predict.

Good.

That's the point.

I don't want Starhound to become a factory for producing variations of whatever worked last time.

I don't want to know exactly what I'm going to be making five years from now.

I want it to remain a place where one interesting thing is allowed to collide with another interesting thing, and we're allowed to find out what happens.

There are probably more efficient ways to run a studio.

There are certainly more sensible ones.

But sensible wasn't really the experiment.

So yes.

We're building too much.

On purpose.

← ALL ARTICLES