Motivation | Get the Books | Topics | Project Guidelines | Author
Placing an earthling from the year 1600 into our todays world must feel like a journey through Hogwarts! Our personal assistent weighs about 500 gramm, contains a complete map of the whole planet, answers any question asked and presents all kinds of information and media to us immediately. When asking how this is even possible the earthling might get the answer oh, afterall it's all just one's and zero's. Confused but motivated to get behind all of this the earthling might find that the construction of the personal assistants requires just minerals and energy to process them. It is all just a machine that is much more sophisticated than anything built around 1600 but there is no magic, there is only electricity, providing the one's and zero's and a dumb piece of metal. It is explained, that the actual behaviour comes from something called software and the one's and zero's are only interpreted as numbers, the same numbers humanity already knew in 1600, and all you need are computations with these numbers.
If the earthling was born in the year 2000 the only thing that changes is the fact, that he is used to the omnipresent magic. Additionally, it is apparent that the magic is not so perfect, breaks down a lot and once updated, might not work at all. Why? And even being used to fact that software works it is not clear how Google Maps finds a route from A to B.
Curiositas is a project in the spirit of just exploring the space of How Software Works. The goal is to write a network of books, that each take a shot of one aspect in the world of computation and implements enough of it, so that it is not magic anymore.
The network aspect maps to the dependencies that exist in the real world. One can not find the shortest distance from A to B if one can neither represent ways from A to B nor represent a distance and in turn sum all required intermediate distances. But traversal of the network is not restricted. Starting from a high level application and going deeper with the understanding is even the prefered method.
As final note, I feel like there are already good books available that explain the foundational algorithms and datastructures. C++ provides a good and generic set of functionality that is used as starting point. Only the STL of C++ is assumed to exist and no external libraries are used otherwise.
The books are published as e-books and sold for a small payment. Multiple topically related books may be sold in bundles.
The scope of the project is not limited to a specific area, not even to computer science. As long as it has something to do with computation it is in theory "in scope". Of course, not all things of the universe that have something to do with computation are adressed.
The first complex of topics are numbers and their direct applications.
- Bring Your Own Int - How to turn bits into numbers and calculate with them | Code Repo
- In Progress: Hashing data - Foundation for cryptographic signing are hash functions. Having a few hash functions greatly helps with applying RSA cryptography. | Code Repo
- In Progress: Sending Secrets to Strangers - RSA encryption - Implement Basic RSA Encryption without Key Generation | Code Repo
- In Progress: Publicly Exchanging a Secret with full Confidentiality - Diffie-Hellman Key Exchange - Implement Diffie-Hellman for RSA | Code Repo
- each book contains approximately one aspect, like natural numbers or how to encode an image
- each aspect must be implemented with real, functional code - functionality is demonstrated by tests
- each book has its own template repository anyone can fork or clone to get a starting point without a lot of boilerplate work
- the template repository contains the final, finished example code, whose tests are passing in CI
- readers may use whatever language they feel comfortable with, but the books utilize primitive C++-20
- (almost) no math, proofs, complexity analysis and other forms of boredome - just code, try out something and see what works
- books will cover both simple ideas, that don't work and better ideas that do - if the simple idea does not work because of algorithmic complexity, it is of course noted and demonstrated
- each book ends with workable code that shows how computational problems are solved in principle - non of it is supposed to become production code, all aspects are educational
- the author is not an expert on any subjects of the book - they are still researched
All results of the natural sciences always fascinated me. The way things evolve in the universe, the interdependencies of live on our planet and the fact that it is even possible to build the Large Hadron Collider sparke a light in me. I want to understand more of it! Humankinds current publishing methods are astonishing and you can find all relevant information to all unanswered questions as videos on YouTube, as articles on Wikipedia or PDFs for the corresponding literature and papers. You can't just play around with it and that's what's missing for me. Consuming a video about gravity and the way planets move is nice. Simulating, experimenting and playing with gravity with your personal, hand made physics engine is rewarding and fun. As long as I find interesting stuff to try out, I am inclined to fiddle with it. Sometimes it might result in an addition to this project. My name is Jonas Toth, I studied Applied Computer Science and wrote my in computer vision. I am currently working as Embedded Software Engineer in the renewable energy sector in Germany.
This project and all its works are under Creative Commons CC BY-NC 4.0 license unless otherwise noted. You may do anything with the code and text provided, but please keep it open and please don't make any money from it. The books are sold for fair prices and make no use of DRM. It takes a lot of time and effort to create these works. I don't want to rely on an ad infested platform or publish only to DRM and device restricted media. Please do not exploit this.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
