Good morning! And welcome to "Kon Codes Kobuta".
My name is Pyry Kontio, or Kon for short, and this is a live programming stream where I hack away, developing Kobuta, a new data format for tabular data.
This is an introduction video of a sort, where I'm going to tell you the basics of this stream and of the Kobuta format.
So, a short self-introduction. I'm a programmer living in Tokyo, Japan. I'm originally from Finland, but I've been living here for a year and something for now.
I develop stuff for both work and for hobby, but lately I wanted to try something new; and that's how I got into streaming. Doing it for the first time, I'm excited and nervous but I hope it's going to be great time.
Of this stream, I'm expecting a kind of a motivational boost and a concentration boost that comes from showing your work to others.
I very much enjoy programming in Rust, so that's going to be my main language in this stream too.
I'm trying to stream at least once a week in English and once a week in Japanese, but I've got no hard schedule. If you want to get notified some hours before I start, follow me on Twitter!
This introduction is scripted, but normally I'm planning to just hack away with the code, verbalizing my thoughts and explaining my thinking process while coding. And hopefully interacting with you too!
So, that being said; about Kobuta. What's the thing I'm making here!?
Kobuta is a new file format, or a data format for tabular data. Think of something like CSV, comma separated values, or Excel files. So there's going to be rows and columns.
Kobuta is meant for storing, exchanging and processing data.
The reason why I started developing a new format is because I'm fed up with the major formats I use at work. Those formats are CSV, JSON and Excel.
Unlike CSV and JSON, Kobuta is a binary format. That means that you can't edit it with a text editor. That's a bummer, but I think of it as a sacrifice I have to do to enable many other things that CSV and JSON are incapable of.
To make up for it, I'm meaning it to be really easy and fast to convert back and forth with existing formats.
The main features of Kobuta is that it's meant to be very performant to read, write and edit. It's designed modern hardware in mind.
Additionally, it's strongly typed, and it carries it's own schema, which means that you set the type of the each column of the data: this is an integer, this is a floating point value, a text string, a boolean truth value and so on. I think that's sorely missing when moving data between databases using CSV.
You can read about the specifics of the format on the readme page of the Github project of Kobuta. Be sure to check that out!
So, there we go. I hope you join me in developing Kobuta - a new data format for tabular data.