by Vas Mylko
One day Ingeenee will start hiring, to build Curiosio product. In this post we will recommend useful books for our future Ingeeneers — who will build AI technology and the product — for most curious trips for billions of tourists. Every book will be explained in details: why we recommend it, what we recommend from it. The focus is on width & depth of vision, wisdom & principles, style & lifestyle.
Top Three
If only three books could be recommended at this moment of time, then it would be this triad in this order:
- Programming Collective Intelligence by Toby Segaran
- The Art of UNIX Programming by Eric S. Raymond
- The Practice of Programming by Brian W. Kernighan, Rob Pike
The title of this book is probably confusing in 2018, especially that Web 2.0 marketing decoration in the name. But the content is great. The content is perfect. First, you have to grasp the idea that intelligence is indeed collective. Second, you have to be able to pull all the pieces and process them.
Data sources are so different, processing methods are so different, algorithms are so different. It’s only 300 pages, and it’s written nicely. Not difficult for beginners in the field. Almost everything from the book is needed if you are an Ingeeneer. If we had to recommend only one book, then it would be Collective Intelligence one. This book is about width of vision.
Ever heard about Lindy effect? It’s mental model #18 at Farnam Street. Here is en excerpt from FS: “The Lindy Effect refers to the life expectancy of a non-perishable object or idea being related to its current lifespan. If an idea or object has lasted for X number of years, it would be expected (on average) to last another X years. A classic text is a prime example: if humanity has been reading Shakespeare’s plays for 500 years, it will be expected to read them for another 500.”
UNIX has been there for tens of years. UNIX will be there for tens of years. Linux has been there for tens of years. Linux will be there for tens of years. So sit and learn the traditions and design principles of the system. The system surpassed multiple generations of chips, networks and storage. The system influenced Cloud and Containers, Data Center Operating Systems, HPC systems. We need this all for global scale.
Part I and Part II are mandatory, especially Rules and Interfaces. Part III and Part IV could be skimmed or skipped, though some pieces from them are essential, e.g. Licensing (it’s hard to realize how many programmers, architects and managers are still so weak in open source and free software & free data licenses). This book is about system design & style.
There are no silver bullets. One size does not fit all. Multiple programming languages and multiple design paradigms are needed. Different tools were built to solve different problems. We use multiple tools, we described some in our Dev Stack. The tools change more frequently than design, design changes more frequently than style. Hence, focus on style.
This books is very practical. It shows most efficient use of programming tools in different contexts. It emphasizes on style. You will learn: how to see alternatives, how to make decisions, how to understand low-level & high-level capabilities, how to debug, test, profile, what is quality. It’s very thin — under 250 pages. It’s written by gurus who built software industry and still building it. This book is about wisdom and style.
Top Seven
Top Three are the same, with four more, in this order:
- Programming Collective Intelligence by Toby Segaran
- The Art of UNIX Programming by Eric S. Raymond
- The Practice of Programming by Brian W. Kernighan, Rob Pike
- Beautiful Data by Toby Segaran, Jeff Hammerbacher
- Hackers & Painters by Paul Graham
- Artificial Intelligence by Stuart Russell, Peter Norvig
- Quantum Programming by Unknown
Humans are most efficient at grasping patterns. To assess a situation, to make a decision — we have to see it. Seeing is believing. With Big Data there is nothing else better than data viz. You got to learn how to visualize the data, to see what the data is.
This book is totally practical. (It’s same Toby Segaran who wrote Collective Intelligence.) It’s a collection of case studies, how people solved real problems, and how data viz helped them. They used different tools over different data. If non-visual programming is less or more disconnected (the program could work for a while without the programmer’s interaction), the data viz is very interactive. You try this and see something, you decide to try that to change the view, then you see something else, and so on and so on.
Once Don Norman said: “Attractive things work better.” We could re-phrase to “Beautiful things work better”. If your data doesn’t look good, there is big probability that the data is incomplete or wrong. And vice versa, if your data looks beautiful, you have probably made something useful. This book is under 350 pages. The book will help you understand and communicate ideas & results.
This books is for width of vision and for cultural introduction. Into the culture of hackers. We are hackers, so should be you. In many many cases you will have to hack it to make it work at right time at right place. Solo virtuoso. The book is written in easy style, vivid, real, only ~200 pages. It’s about lifestyle.
This is not a book, this is a whole text book for the whole semester or two. Good news — many chapters could be read in arbitrary order. Bad news — there are so many chapters and pages.
Curiosio works on top of Travel AI technology. AI emerges from Complexity. So here we are — to build AI you must stand on the shoulders of the giants, who built it before you. Many algorithms will be reused, and you will also invent and design new algorithms. This book isn’t easy or entertaining. It’s a hardcore foundation for your future inventions. The book is for width & depth of vision.
Quantum Programming is the seventh book. We don’t know how much time it will take you to get here. By the time you are here, you got to master your own style — learning to learn, design of design, meta. Geeks, come on, whatever works. You are hackers, make it work when we reach out to you.
