31 May, 2025

It all started as a simple as possible website for my off-screen project (Moto Jazz Roulette).

My objective was to sell the product without much effort, because I was about to buy a similar one, also like this (poor video, poor site, expensive product). So I tried to create one by myself.  I started to investigate how to reproduce it with the materials that I have in my garage, in the neighborhood, and inside Brazil.

I was very happy with the final product, which led to the next step: creating a replicable version. The prototype was ready in August 2024.

Since there was no expiration time for the steel roulette, there was no reason for me not to try to sell. I'm the worst seller ever, so I can still keep trying it for the next decades.

Once I replicated nine units, I stored them in a closet. I thought that to open a virtual store, I should have at least four different products to sell. I force myself to create something else with the leftover pieces. Shuffle Lights was born one month later. Another barrier that I placed was the 200USD price. I can produce and sell only at this price. There are other small products of the family, but they didn't pass this price barrier.

For Shuffle Lights, I just replicate two pieces. I like to keep very low expectations.

When I reached this point, I started to look for a domain. It impressed me that shufflemachines.com was available. My first thought was to buy and wait for the main product seller to buy me back. So I had this in my mind when I was drawing the logo for the site. I focused on the initials, so if it was necessary to change, I could find another combination of words.

I bought the domain and started working on the website. My first search was for a solution that could combine Python development, games, and the website. I realized the need for a dedicated server, and that was out of the question. Since the goal was to sell a roulette wheel, I couldn't afford the cost of a server to just play around.

Since I didn't want anything too complex, I started with a platform that helps create websites really quickly and offers several possibilities. My first test was to create the Dice page. I thought the response time was really bad. I took the generated code and asked ChatGPT to review it. It was after a lot of trial and error that I realized something had changed. It was the "Eureka" moment! The response was absurdly fast. This served as the basis for all the other scripts that came later.

Since there are no courses on the market that use JavaScript for games, I was probably doing something exotic and maybe wrong. I don't like doing anything exotic, so it needs to be tested a lot before I can say that this is a good path to follow.

I paid for the standard license and started organizing things.

I feared mixing my professional and private life with this venture, so I didn't want to expose myself. The first goal was to test whether anyone liked the roulette wheel, so I tried to eliminate the bias of my story and my image. It would have to be someone outside my family and circle of friends. I was careful to only comment on what was necessary, because I didn't want to sell in an artificial way. My ideal client would buy the roulette wheel without question and would be pleasantly surprised by what they ended up discovering. Or they would like the website and would buy the roulette wheel to help maintain the website. I would be a customer for this store, so I didn't want to validate it with acquaintances, because the simple fact that the person knew me could lead them to say that the roulette wheel is cool, and even buy it. This concept is well explored in the book “Anti-Fragile”, which I read in 2024.

By 2023, I already had almost 100 detective stories with a science fiction theme, so I tried to create a version for people to play this type of game alone. I hadn't seen anything on the market, so I was very happy when I created “Galactic Mysteries”. I thought that announcing a detective game was a bit cliché, so the idea was to keep it a bit hidden. I'm not sure if it was a good idea.

This excitement led me to want to open the store before the four products. I thought it was too cool to keep just for myself, so I brought a science fiction book that I wrote for children in 2023, which is an adaptation of a short story I had written when I was part of a group of science fiction writers called “Dream Factory”. It is a low-cost product, the lowest possible for publication.

“Hangman” came next and gave me an idea. I was trying to do something with a game I had the opportunity to learn about in 2017, called “Torto Reverso”, and switched to “ShuffleWords” for this project. I already had a machine that generated custom grids, and it was a matter of recovering some old backup versions to be able to run it again. I tried to make a version for PostgreSQL, since it was still written in Oracle, but it didn't work.

At that point, I was very happy, and several ideas emerged about what could be done with similar logic to what I had already consolidated. I had to hold myself back from working too much on the site. Each achievement made me more excited. I was structuring it to create customizable grids for the games, so the garden aspect was forming. My idea was to spend time making grids for the games. I didn’t intend to launch the site for a specific date or as a finished art. The idea is to keep working on it over the next few decades.

I realized I needed to work on a background for the website and revived a project from a mural I painted for a friend. I got excited and created the logo using elements from the background to maintain the visual identity. I had already worked in advertising between 1999 and 2006, so I went back to creating visual art, which made me soooo happy. I printed some stickers with the logo and stuck them on the shelf where the tools and pieces are. I need to make some t-shirts and put them in the store too.

That's when I needed to create a name for the products. Moto Jazz Roulette came up very naturally, and I immediately liked it. The problem is that there wasn't anything with Jazz or motorcycles in the store yet. I was forcing my brain to do something with the Enduro theme, since it's something my father practiced and managed to pass on his love for trails to me. I spent about 2 months thinking about incorporating this theme with what I already had. While I didn't have anything in mind, I ended up creating with other themes. Next came “Enter in Time”, “Tarot”, "Galactic Journey of the Hero”, “Galactic Journey of the Fool”, “Character”, “Comedy”, “Fantasy", “Stoics”, “Samurai”, “Paradoxes”, “Bias”, and “Fallacies”.

In addition to “Enduro”, I wanted to create a “canastra” style card game. I thought a lot about mixing groups of cards that could be counted at the end. That's when I had the idea of ​​generating Supertrump style cards for the main fictional starships, and that's how “Galactic Aquisition and Profit” was born. With the same dynamic came “F1”, “NBA”, “Soccer”, “Reggae”, “Jazz”, “Blues”, and “Rock”. I focused on the names of artists, athletes, and objects with the same theme. Apparently, this sounds like Dungeons & Dragons. The truth is that I haven't had the opportunity to play this type of game yet.

This batch included “Color Sequence,” “Alfa Sequence,” “Memory,” and “Music Notes Sequence.” I was inspired by the game I used to play as a kid, “Pense Bem.”

I already had “Jazz,” which matched the name of the roulette wheel, but I still needed "Enduro”. It seemed like an impossible challenge, so I left it aside. That's when I started looking for something with a grid and billiards. After several tests, I came across the first version of "Mini-Golf”. It was exciting when I thought of using the traveling cashier problem on the grid. The first version of “Enduro” had been created!

Everything was great until I remembered that I had taken a break from social media because of work and politics. To keep my life private, I decided to create an account for Shufflemachines, but I realized that I would have to have a business to do so, and I wasn't sure if I would have any customers yet.

On social media, I just observe and sometimes compliment cool stuff. My discretion and the post for the detective game (without mentioning the game's name) created a distorted image of the store's purpose. Apparently, no one believed that it would really be a roulette and free games.

I paid for some ads to promote:

“Moto Jazz Roulette”, “Caco Matrix”, “Galactic Mysteries”, “ShuffleWords”, “Enduro”, and “Enter in Time”.

During this period, “Checkmate,” “Save the Whales,” and “Crack the Chord” (changed to “Discover the Chord”) came out, inspired by what I was observing in games on the market. Improving the first version of “Enduro,” I came up with “Jack-o'-lantern” and then “Spaceship.” I really liked being able to add hidden passwords and wormholes. The number of space-themed emojis helped, too.

At this point, I tried to add gravity to "Jack-o'-lantern", and after a lot of work, I came up with a version called “Mission”. The idea was to use the script to create different versions with the same mechanics. The first one that came to mind was “Dodo Quest,” where a bird runs to escape extinction. I ended up leaving the idea aside to work on what seemed more logical.

Monkey Maze came about because of the similarity between the monkey emoji and the game mechanics. There are always monkeys in my backyard, too. I tried not to use bananas to avoid the obvious, and I used strawberries, which I also have plenty of at home. Other things that are quite common are scorpions, cacti, rainbows, and butterflies, so the neighborhood I live in is well represented in the game.

That’s when I realized the need and possibility of synchronizing posts on different platforms. It was very powerful to see how publishing on different platforms creates a kind of musical chord. Posts that complement each other and I also tried to maintain a poetic pause. I published only on Wednesdays, Thursdays, and Fridays. The only direct message other than to the sellers was a critique of Moto Jazz Roulette - noted for future improvements.

About indirect messages. This site is almost invisible and has zero hits. Even at the best times when I was advertising, there was no access or permanence on the site to justify keeping it active. However, the channels I consume public information seemed to be synchronized with the updates I was making on the site.

Some indirect messages were strong enough to make me create other styles and sports. “Rap”, “Pop”, “Screen”, “Samba”, “WCT”, “NFL”, “MLB”, “NHL”, “Boxing”, “Tennis”, “Golf”, “Volleyball”, “Basketball”, “Gymnastics”, “Tango”, “Art”, “Chef”, and “Photography”. I felt like I was in an encyclopedia and learned a lot from what could be created following similar logic. Most indirect messages were positive, but I'm unsure about their source.

I set out to post one original post per week from August 26, 2024, and kept doing so until January 10, 2025, when I no longer felt comfortable keeping the account open and closed the store with a 160-character password. It was my attempt to step away to see if I would miss it, too.

After 3 weeks, things returned to normal, and I restarted working on the garden.

I also brought the "Holidays around the World" app project back to life. "Spinn", "Metronome", and "Mini Music Box" were the latest outstanding achievements, which unlocked other good use cases.

So far, so good.